HUMAN–LIVESTOCK INTERACTIONS,SECOND EDITION

The Stockperson and the Productivity and Welfare of Intensively Farmed Animals

This book is dedicated to the late John Barnett, our friend and colleague, who contributed so much to our research and to animal welfare research in general.

HUMAN–LIVESTOCK

INTERACTIONS,SECOND EDITION

The Stockperson and the Productivity and Welfare of Intensively Farmed Animals

Paul H. Hemsworth

Animal Welfare Science CentreUniversity of Melbourne andDepartment of Primary IndustriesAustralia

and

Grahame J. Coleman

Animal Welfare Science CentreMonash UniversityAustralia

CABI is a trading name of CAB International

CABI Head Offi ce CABI North American Offi ceNosworthy Way 875 Massachusetts AvenueWallingford 7th FloorOxfordshire OX10 8DE Cambridge, MA 02139UK USA

Tel: +44 (0)1491 832111 Tel: +1 617 395 4056Fax: +44 (0)1491 833508 Fax: +1 617 354 6875E-mail: cabi@cabi.org E-mail: cabi-nao@cabi.orgWebsite: www.cabi.org

© CAB International 2011. All rights reserved. No part of this publication may be reproduced in any form or by any means, electronically, mechanically, by photocopying, recording or otherwise, without the prior permission of the copyright owners.

A catalogue record for this book is available from the British Library, London, UK.

Library of Congress Cataloging-in-Publication DataHemsworth, Paul H. Human–livestock interactions : the stockperson and the productivity and welfare of intensively farmed animals / Paul H. Hemsworth and Grahame J. Coleman. – 2nd ed. p. cm. Includes bibliographical references and index. ISBN 978-1-84593-673-0 (alk. paper)1. Animal industry–Moral and ethical aspects. 2. Livestock–Research–Moral and ethical aspects. 3. Animal welfare. I. Coleman, Grahame J. II. Title.

HV4757.H46 2011 174'.9636–dc22

2010022610

ISBN-13: 978 1 84593 673 0

Commissioning editor: Rachel CuttsProduction editor: Shankari Wilford

Typeset by AMA Dataset, Preston, UK.Printed and bound in the UK by CPI Antony Rowe, Chippenham, UK.

v

Contents

Preface vii

1 Introduction: the Stockperson as a Professional – Skills, Knowledge and Status 1

2 Farm Animal Welfare: Assessment, Issues and Implications 21

3 Human–Animal Interactions and Animal Productivity and Welfare 47

4 Attitudes of Stockpeople 84

5 Stockperson Behaviour and Animal Behaviour 103

6 A Model of Stockperson–Animal Interactions and their Implications for Livestock 120

7 Changing Stockperson Attitudes and Behaviour 135

8 Conclusion: Current and Future Opportunities to Improve Human–Animal Interactions in Livestock Production 153

References 169

Index 189

This page intentionally left blank

vii

Preface

Human–livestock interactions are the topic of this book because there is an ever-increasing body of evidence which demonstrates that these interactions may result in profound behavioural and physiological changes in the animal, with consequences on the animal’s performance and welfare. Furthermore, these inter-actions may also infl uence the stockperson to the extent that job-related character-istics, such as job satisfaction, motivation and commitment, may be affected with implications for the job performance and career prospects of the stockperson.

When we wrote the fi rst edition in 1998, there was an emerging appreciation of the infl uence of the stockperson on the productivity of livestock. Our studies and those by M.F. Seabrook in the late 1970s and early 1980s on the implica-tions of human–animal interactions for farm animals and studies by W.B. Gross and P.B. Siegel on the implications of human–animal interactions on experimen-tal animals at a similar time demonstrated the effects of handling on animal behaviour, physiology and productivity. These early studies stimulated subse-quent research in the 1980s and 1990s on the human characteristics respon-sible for these effects. However, progress in understanding these relationships in livestock production was relatively slow at this time compared with other devel-opments in the fi eld of animal science. The study of stockperson characteristics in livestock production creates a number of problems that are generally not encountered when studying other more traditional areas of livestock production. Stockperson characteristics are not as amenable to study as other factors such as nutrition, housing, genetics, etc., because of our limited ability to: (i) manipulate individual characteristics; (ii) control others not under direct study; and (iii) study humans in commercial situations. These problems were exacerbated by the lack of interest shown by psychologists in this important area.

The early research on human–livestock relationships was initially conducted because of its implications for farm animal productivity, since fear responses to humans were shown to reduce farm animal productivity through stress. How-ever, with the ever-increasing interest in animal welfare and a better appreciation

viii Preface

of the role of stockpeople in determining animal performance and welfare, there has been a substantial amount of research conducted on human–livestock rela-tionships since the 1990s.

Our intention with the fi rst edition was to stimulate interest and exploration of the subject of human–animal interactions by animal scientists and industry personnel interested in the role of the stockperson in determining animal perfor-mance and welfare. We emphasized that this subject is particularly relevant to those with responsibilities in the areas of staff training and selection in the live-stock industries. This in turn is also relevant to the livestock industries’ efforts to attract and retain desirable staff.

The research on human–livestock relationships is multidisciplinary in nature and much of the material presented in the two editions is a mixture of agricultural and animal science and psychology. This presents something of a dilemma when attempting to identify the target audience for this book and, therefore, how to pitch the material so that readers will not fi nd it too technical within a particular discipline but, at the same time, will gain insight into the processes that are being described.

As with the fi rst edition, this edition is aimed at those people who have an interest in human–animal interactions in livestock production, perhaps as train-ers, managers of livestock farms, students and academics seeking an introduc-tion to the subject. We have attempted to make the book as self-contained as possible, by giving a brief account of the theories or principles underlying the research discussed.

The fi rst half of this book contains a detailed review of our empirical knowl-edge of human–animal interactions and their human and animal effects. With this review of human–animal interactions in livestock production together with a review of the development of the theory underlying this empirical research, the second half of the book leads into an examination of the opportunities to manip-ulate these human–animal interactions. If characteristics of the stockperson are important determinants of human–animal interactions, opportunities exist to improve animal performance and welfare in those situations in which the human–animal relationship is poor. The main opportunities to improve these key human characteristics are through stockperson training and selection.

A major motivation in writing this book is to provide a consolidated account of the role of stockpeople so that the contribution of the stockperson to farm animal welfare and productivity can be seen to have an importance similar to that of many of the factors which are traditionally considered in animal hus-bandry. It is our intention to stimulate interest, understanding and exploration of the subject by animal science students, animal scientists and industry personnel interested in the role of the stockperson in determining animal productivity and welfare. The subject is particularly relevant to those with responsibilities in the areas of staff training and selection in the livestock industries. This topic is also relevant to the livestock industries’ efforts to attract and retain desirable staff.

It is the contention of this book that human factors contribute to farm animal welfare and productivity to an extent similar to that of many other factors, such as housing, and that human factors need to be recognized as a routine compo-nent of animal husbandry. To some this may be obvious and not worth saying,

Preface ix

but for others, it remains less obvious or even contentious. It is the purpose of this book to establish the credibility of this point of view.

As we conclude in the fi nal chapter, this is a relatively new direction for industries in which stockpeople regularly interact with livestock. Much has been done to improve genetics, nutrition, health and housing but efforts to target the stockperson, who performs such a key function, have just begun. We should not underestimate the role and impact of the stockperson on animal productivity and welfare. To do so will seriously risk the productivity and welfare of our livestock. It is likely that both the livestock industries and the general community will place an increasing emphasis on ensuring the competency of stockpeople to manage our livestock: the livestock industries’ interests in this topic are thus likely to be motivated by both animal productivity and welfare and the general community’s interest in animal welfare.

A substantial amount of the research reported in this book was conducted by the authors and many Australian and overseas colleagues. One very prominent contributor to this work was the late John Lawrence Barnett. John’s enduring interest in animal biology and how animals deal with challenges, together with his rigorous approach to research and review, have been signifi cant contribu-tions to developments in understanding human–livestock interactions and their impact on the animal. Strategic support and funding by many Australian research organizations, such as Australian Pork Ltd, Australian Poultry Cooperative Research Centre, Australian Egg Corporation Ltd, Dairy Australia, Australian Research Council, Rural Industry Research Corporation (Chicken Meat) and Meat Livestock Australia as well as the Department of Primary Industries (Victoria), have been essential in understanding and consequently improving human–animal relationships in the livestock industries.

Paul H. Hemsworth and

Grahame J. Coleman

This page intentionally left blank

© CAB International 2011. Human–Livestock Interactions, Second Edition (P.H. Hemsworth and G.J. Coleman) 1

1 Introduction: the Stockperson as a Professional – Skills, Knowledge and Status

Human–animal interactions are a key feature of modern livestock production and research has shown that the quality of the relationship that is developed between stockpeople and their animals can have substantial effects on both the animals and the stockpeople. For example, there is good evidence based on han-dling studies and observations in the livestock industries that human–animal inter-actions may markedly affect the productivity and welfare of farm animals. By infl uencing the behavioural response of animals to humans, and in particular the ease with which animals can be observed, handled and managed, human–animal interactions may also have implications for a number of work-related characteristics of the stockperson, such as job satisfaction, which may, in turn affect job retention and thus may have a substantial impact on the stockperson. Since the fi rst edition of this book in which our research fi ndings were heavily utilized to review the infl u-ence of human–animal interactions in livestock production, there has been a sub-stantial amount of research that has extended these fi ndings. The original research was mainly conducted in Australia and primarily on pigs, but data are now available from New Zealand, Europe and the USA on pigs and other livestock species. The fi rst objective of this book is to review human–animal interactions in livestock pro-duction and, in so doing, examine their implications for both the farm animal and the stockperson. There is still a tendency in the livestock industries for stockpeople not to be treated as professionals. Most stockperson training targets codes of prac-tice, regulations and husbandry competencies rather than attitudes and behaviour towards animals, work attitudes and job satisfaction.

A number of human and animal characteristics infl uence human–animal interactions, which in turn may have marked effects on both partners. Thus an understanding of these key human and animal characteristics and an ability to manipulate at least some of these may offer the livestock industries opportunities to provide benefi ts for both their animals and their stockpeople in order to improve industry economics and sustainability. The second objective of this book is to explore the key human characteristics that infl uence animal behaviour,

2 Chapter 1

performance and welfare. If the characteristics of the stockperson are important determinants of human–animal interactions, opportunities may exist to improve animal performance and welfare in those situations in which the human–animal relationship is poor. Thus the third objective of the book is to examine the oppor-tunities for industry to improve these key human characteristics through stock-person training and selection. The fi rst half of this book contains a detailed review of our empirical knowledge of human–animal interactions (Chapters 3 to 5), which, in the second half of the book (Chapters 6 to 8), leads to the discus-sion of the theory underlying this empirical research and an examination of the opportunities to manipulate these human–animal interactions.

A major motivation in writing this book is to provide a consolidated account of the role of stockpeople or animal carers in the livestock industries so that the contribution of the stockperson to farm animal welfare and productivity can be seen to have an importance similar to many of the factors that are traditionally considered in animal husbandry. It is the contention of this book that human factors contribute to farm animal welfare and productivity to an extent similar to that of many other factors, such as housing, and that human factors need to be recognized as a routine component of animal husbandry. To some this may be obvious and not worth saying, but for others it remains less obvious or even contentious. More importantly, even if there is broad agreement about the importance of the stockperson, the relevant human factors and their specifi city or generality across different livestock species constitute an empirical issue that may not be intuitively obvious. It is the purpose of this book to establish the credibility of this point of view.

This fi rst chapter together with Chapter 2 attempts to set the stage by exam-ining the role of the stockperson in terms of both the stockperson’s role in live-stock production and the ethical and welfare issues relating to farming of livestock. In particular, the fi rst chapter considers the role of the stockperson, focusing on his or her skills and knowledge that are required to achieve high animal performance and welfare.

1.1 The Role of Stockpeople

Any reasonable assessment of the role of stockpeople in modern agriculture indicates that they are professional managers of animals who are integral to determining animal performance and welfare. Yet there appears to be a general lack of appreciation of this by people within livestock production, including stockpeople themselves. It is one of the important contentions of this book that the recognition of stockpeople as key professional managers of livestock is an important cultural change that is required within livestock production; such a change is likely to have implications for the image and self-esteem of stockpeople and the opportunities for training stockpeople, which in turn are likely to be highly infl uential in affecting animal performance and welfare.

In some sectors of livestock production, stockpeople are recognized as important resources and consequently policies on the development of this human resource have been introduced. Owner-operators of farms also may

Introduction 3

undervalue their contribution as livestock managers to animal performance and welfare. There needs to be a widespread recognition and appreciation of the important role of the stockperson in livestock production; such a cultural change will; in itself; facilitate the appropriate management of this important human resource. For example, appropriate staff selection and training policies and other strategies to select, retain and further develop stockpeople are likely to become increasingly widespread as the livestock industries recognize the impact of stock-people on animal performance and welfare, and thus on industry profi tability and sustainability. These developments in the management of human resources should not necessarily remain in the domain of the large corporate enterprises, because commercial and government services have the opportunity to provide such support to smaller enterprises in the interests of industry economics and sustainability. Since the fi rst edition of this book, there has been relatively little research on staff selection (see Coleman, 2004), despite the evidence that there are several personal characteristics that predict good stockperson performance (Carless et al., 2007). In contrast, there have been substantial developments in the training of stockpeople.

One of the most famous pronouncements on the role of the stockperson in livestock production was contained in the British Codes of Recommendations for the Welfare of Farm Livestock (Ministry of Agriculture, Fisheries and Food, 1983): ‘Stockmanship is a key factor because, no matter how otherwise acceptable a system may be in principle, without competent, diligent stockmanship, the wel-fare of animals cannot be adequately catered for’. Unfortunately, it is debatable whether these sentiments have been fully accepted by the livestock industries and others.

Few studies have attempted to quantify the contribution of the stockperson to animal productivity and welfare. This is partly owing to the diffi culty of such research but also because of the industry’s focus on technological developments in livestock production. Some of the research documenting the stockperson’s contribution will be discussed later in this book, but it is useful at this point to consider a study at our laboratory that provides an indication of the contribution of the stockperson to the productivity and welfare of farm animals.

A study by Pedersen et al. (1998) provides limited evidence that positive handling by stockpeople may ameliorate the chronic stress response associated with an aversive housing system. Research on pigs has consistently shown that pregnant sows housed in tether stalls of a specifi c design will experience a sus-tained elevation in the basal plasma concentrations of the stress hormone corti-sol, which is indicative of a chronic stress response (Barnett et al., 1989, 1991). In this study by Pedersen and colleagues, 24 pregnant sows housed in stalls with neck tethers were randomly assigned to one of three handling treatments: posi-tive handling in which pigs were patted or stroked whenever they approached; minimal human contact; and negative handling, in which pigs were briefl y shocked with a battery-operated goad or prodder whenever they closely approached. The positive and negative handling treatments were imposed daily for 3 min and the experimenter squatted in front of each pig’s stall to impose the appropriate treatment if the animal closely approached. Daytime profi les of plasma free cortisol were signifi cantly lower in the pigs in the positive handling

4 Chapter 1

treatment than those in the minimal and negative handling treatments (Table 1.1). The cortisol concentrations were measured in isolation from humans by the use of extensions to indwelling catheters so that blood samples could be collected by the experimenter in visual isolation from the pigs. These results demonstrate the importance of human factors in pig welfare and highlight, in this situation, the importance of human factors for animal welfare and productivity. Other handling studies will be discussed in Chapter 3 (Sections 3.5 and 3.6) that highlight the importance of human factors in farm animal performance and welfare.

This chapter will consider the factors that need to be addressed when defi n-ing the role of the stockperson. The underlying principles drawn from industrial/organizational psychology will be discussed fi rst, followed by an overview of skills, knowledge and status of stockpeople.

1.2 Characteristics of the Stockperson’s Job

In general, farm personnel in animal agriculture are regarded as itinerant and unskilled workers. Farm managers often appear to be reluctant to invest too much effort in training stockpeople because of the high rate of turnover. The role of the stockperson as the key person responsible for the day-to-day welfare and pro-ductivity of the animals under his or her care has not received due acknowledge-ment. However, recognition of this role of farm personnel in livestock production leads to the recognition of these personnel as an important human resource that needs to be selected, trained and managed in a way similar to current practice in a wide range of white-collar industries. In fact, stockpeople are professional managers of livestock. Farm owners and the general community (through governments) entrust the welfare and performance of large numbers of animals to the care of stockpeople. A duty statement for a modern stockperson may be presented as:

1. A good general knowledge of the nutritional, climatic, social and health requirements of the farm animal.2. Practical experience in the care and maintenance of the animal.3. Ability to quickly identify any departures in the behaviour, health or perfor-mance of the animal and promptly provide or seek appropriate support to address these departures.4. Ability to work effectively independently and/or in teams, under general supervision, with daily responsibility for the care and maintenance of large numbers of animals.

Table 1.1. Cortisol concentrations in 24 pregnant sows housed on tethers receiving either positive, negative or minimal handling, 3 min/day for 4 weeks (from Pedersen et al., 1998).

Handling treatment

Dependent variable Positive Negative Minimal

Daytime mean cortisol concentrations (nmol/l) 2.9 7.4 6.1

Introduction 5

The stockperson is, therefore, required to possess a basic knowledge of both the behaviour of the animal and its nutritional, climatic, housing, health, social and reproductive requirements, together with a range of highly developed husbandry and management skills to effectively care for and manage farm animals. For example, stockpeople may have knowledge and skills in a number of diverse management and husbandry tasks, such as: oestrus detection (Fig. 1.1) and mat-ing assistance; semen collection, semen preparation and artifi cial insemination; pregnancy diagnosis with real-time ultrasound; artifi cial rearing of early weaned animals; milk harvesting; controlling and monitoring of feed intake to optimize growth, body composition, milk production and reproductive performance; pas-ture management to optimize pasture production; routine health checks; moni-toring and adjusting climatic conditions in indoor units; administering antibiotics and vaccines; shearing and crutching of sheep; teeth and tail clipping of pigs; castration of males; and effective and safe animal handling. These are highly skilful tasks and stockpeople are required to be competent in many of them. Clearly, the training of stockpeople to develop these competencies should be a systematically and soundly implemented process in which the requirements of both the stockperson and the industry are addressed.

The conditions in which stockpeople are required to work may differ within and between livestock industries and also from those encountered in non- agricultural industries. Stockpeople are often required to work unconventional and unsocial hours and, at times, under unpleasant conditions. Climatic condi-tions may vary markedly in extensive livestock production and stockpeople in these industries are required to work outdoors in extreme weather conditions.

Fig. 1.1. Stockpeople in piggeries are required to detect oestrus in female pigs in order to accurately time either artifi cial insemination or natural mating and identify returns to oestrus.

6 Chapter 1

The standard of workplace amenities for both animals and workers may also vary considerably in each type of production system within this continuum. For example, in some indoor units effl uent odours and dust levels may be offensive.

The requirements of the job and workplace conditions are onerous and demanding, and often not widely and fully recognized by the animal industries and others.

1.3 The Stockperson: Image and Self-Esteem

In contrast to the views discussed earlier and contained in the British Codes of Recommendations for the Welfare of Farm Livestock (Ministry of Agriculture, Fisheries and Food, 1983) and the implications of studies such as that by Pedersen and colleagues (Section 1.1), surveys on the self-esteem of stockpeople and their perceptions of their image outside agriculture often portray a disturbing picture. For example, Beynon (1991) reported that the families of pig stockpeople regarded employment in the pig industry as having a low status. Industry leaders commonly quote the poor image of stockpeople as a factor contributing to the problem of attracting people to and retaining staff in the livestock industries. In a compassionate plea for recognition of the role of stockpeople, English et al.(1992) suggest that stockpeople are the world’s most undervalued profession.

1.4 Fitting the Stockperson to the Job

There is a clear need to identify the attributes that best allow a person to meet the job requirements for a stockperson. There is a wide range of factors that infl u-ence an individual’s performance in the workplace. Seabrook (1982) attempted to identify these in agricultural workers, but in the absence of any quantitative data was only able to identify some generic descriptive traits that related to the degree of control and level of responsibility with which a stockperson was com-fortable. Seabrook concluded that ‘one can come to no generalized conclusions explaining why (stockpeople) behave in the way that they do’ (Seabrook, 1982, p. 69). This conclusion demands closer scrutiny. There is a need for specifi city in identifying the characteristics relevant to stockperson performance. First, there are characteristics of the individual. On the one hand, a range of dispositional factors, that is, relatively stable characteristics of the person which initiate and mediate or moderate behaviour, provide a basic framework within which the individual interacts with his or her environment; on the other hand, there is a range of learned factors, including not only skills and knowledge but also learned motivations, which affect behaviour. Second, there is a range of environmental factors that provide physical constraints within which the person works. Finally, there is a range of demographic factors such as family size, distance from work, and so on.

Stockperson characteristics will be a combination of dispositions (personal-ity, empathy, etc.) and learned factors (skills, knowledge, etc.). In general, the

Introduction 7

small amount of research that has been done on stockperson characteristics has adopted one of two broad approaches. On the one hand, some studies have attempted to identify the characteristics of stockpeople currently employed in livestock industries on the assumption that there will be a match between the job requirements and the person. On the other hand, a few studies have attempted to identify those characteristics that are associated with stockpeople that achieve good performance in livestock industries. The implications that may be drawn from these two kinds of studies are different. The fact that a stockperson is employed in an agricultural system may be a result of a multitude of factors, partly geographical and fi nancial, and partly related to the characteristics of the person. Certainly, the fact that a stockperson is currently employed in agriculture does not imply that the person is necessarily the best for the welfare or productiv-ity of the animals under his or her care.

The personal characteristics of the stockperson will be discussed in this chap-ter from a largely theoretical perspective, with a more detailed review of current empirical research in later chapters.

1.4.1 Skills and knowledge

The single most important factor in job performance is the skills that the person brings to the job. Knowing and being skilled at the techniques that must be used to accomplish the task are clearly prerequisites to being able to perform that task. For example, in the pig industry, a stockperson working with the breeding herd must be good at oestrus detection and conducting artifi cial insemination or assisting mat-ing and, if he or she does not have these skills, then production will be severely impaired. The stockperson must also be able to recognize changes in the behaviour and physical condition of livestock that may be indicative of welfare, health and productivity problems. Less obvious, and also the subject of limited research, is the impact of stockperson skills in handling and interacting with intensively and exten-sively farmed animals. Research has shown that many stockpeople in the pig and dairy industries do not know what aspects of routine handling farm animals fi nd aversive, despite the fact that it has been shown that aversive handling has marked negative consequences for the animal and its productivity and welfare (Hemsworth et al., 1993). This will be the subject of extensive review in Chapter 3.

1.4.2 Personality

The principal dispositional characteristic that is invoked to account for a range of human behaviours is personality. Although there is some disagreement among psychologists, it is reasonably well accepted that a personality trait is a relatively enduring characteristic which exerts a general effect on that person’s behaviour and which we cannot observe directly, but that we can infer from the person’s behaviour. Gordon Allport defi ned personality as ‘the dynamic organization within the individual of those psychophysical systems that determine his unique adjustments to his environment’(Allport, 1937, p. 48).

8 Chapter 1

Today most researchers agree that personality can be characterized in terms of fi ve dimensions (the so-called ‘big fi ve’): (i) extraversion/introversion; (ii) emo-tional stability; (iii) agreeableness; (iv) conscientiousness; and (v) intellect (Costa and McCrae, 1992). Extraversion is associated with sociability, assertiveness and an outgoing nature. Emotional stability refers to a trait similar to Eysenck’s (1966) neuroticism and includes such things as anxiety, embarrassment and insecurity. Agreeableness is associated with cooperation, good nature and toler-ance. Conscientiousness is characterized by dependability, hard work and perse-verance. Intellect includes being imaginative, cultured and original (Barrick and Mount, 1991). These personality factors appear to be useful in matching people to some kinds of jobs. Barrick and Mount (1991) found that conscientiousness predicted job success across a range of job categories and was the most signifi -cant personality characteristic associated with sales performance. There was also a less-strong relationship between extraversion and sales performance.

Another measure of personality type, the Myers-Briggs Type Indicator (MBTI), identifi es the basic characteristics of people in relation to how they use their perceptions and judgement and, therefore, how they differ in their reactions, values, motivations, skills and interests. The MBTI has four bipolar dimensions: extraversion–introversion, thinking judgement–feeling judgement, sensing perception–intuitive perception, and judgement–perception.

As will be discussed in more detail later in this book (Section 4.3.1), there is little evidence relating personality directly to work performance in the livestock industries. Seabrook (1972a,b) reported that the stockperson’s personality was related to the behaviour of the cows and milk yield of the herd. Beveridge (1996) investigated the relationship between personality types, as measured by the MBTI, and the behaviour of dairy stockpeople toward cows. The MBTI corre-lated more strongly with measures of stockperson attitude than with stockperson behaviour, but showed no correlations with milk yield. In a study by Waiblinger et al. (2002), the personal characteristics of stockpeople, based on the measures used by Seabrook (1972a,b), also did not correlate signifi cantly with milk yield, but did correlate with the attitudes of stockpeople. This relationship between personality and attitude is consistent with Fishbein and Ajzen’s Theory of Rea-soned Action (Ajzen and Fishbein, 1980), which is considered in Chapter 4 (Section 4.2). These authors argue that personality is one set of factors that underlie attitude formation rather than being a direct determinant of behaviour.

Signifi cant relationships have been found in the pig industry between per-sonality types of stockpeople and productivity in farrowing units. Seabrook (1996) reported that pig performance, measured by litter size, was associated with aspects of stockperson personality. Ravel et al. (1996) found that some per-sonality attributes were associated with piglet survival at independent owner-operated farms, and higher piglet mortality at large integrated farms.

It is diffi cult to extract a pattern from these varied results. The fact that sev-eral researchers, in different contexts and using different measures of personality, have been able to fi nd direct or indirect relationships between stockperson per-sonality and production outcomes does suggest that personality may well be a relevant factor in animal systems. The perennial problem of different measures of personality with variable validity coupled with a variety of outcome measures

Introduction 9

serves to obscure the picture. Also, the fact that the dependent variables often are farm outcomes rather than the performance of individual stockpeople means that, because many factors can intervene between stockperson characteristics and the productivity and welfare of the animals under his or her care, it may be diffi cult to determine the causal sequence between stockperson characteristics and productivity.

Apart from personality measures, it may be the case that degree of empathy predisposes people to be good stockpeople. Certainly, the idea that stockpeople will perform best if they have good insight into the emotional responses of the animals under their care has strong intuitive appeal and is consistent with our fi ndings that stockperson behaviour is an important determinant of farm animal productivity. Empathy can be described as the capacity to put oneself in the place of another. It has two basic components, an attributional and an experiential element. The attribu-tional element relates to people’s capacity to recognize the emotional state of the animal, while the experiential element refers to people’s capacity to experience an emotional state following its observation in an animal. These elements may take the form of vicarious experience of another’s emotions or may simply be a capacity for role taking. Chlopan et al. (1985) have concluded that the most appropriate way of considering empathy is to regard it as being multifaceted and containing both role taking and vicarious experience components. While empathy is a dispositional characteristic, unlike personality, there is some argument about whether empathy is innate or learned. In a recent review, Duan and Hill (1996) distinguished between a trait approach, which is widely adopted by psychotherapists and others, and a situation-specifi c social learning approach, which is amenable to training.

In the agricultural literature, the term empathy has been used to describe the bond that exists between humans and animals under their care (English et al.,1992). In fact, an empathic bond may exist between stockpeople and their ani-mals; however, empathy does not refer to the bond itself, which may have its origins in a number of factors of which empathy is one. Empathy refers to the way in which stockpeople may feel a bond with their animals because of being able to put themselves in the animal’s position or to understand the way in which the animals are reacting.

Only limited empirical data from livestock production are available (Beveridge,1996; Coleman et al., 1998). The limited research on empathy will be discussed in detail later (Section 4.3.1), but in general, empathy appears to be correlated with attitudes rather than with specifi c stockperson behaviours.

In summary, data from the studies by Beveridge (1996), Coleman et al. (1998) and Waiblinger et al. (2002) provide evidence in support of the proposed relation-ship between personality variables and attitudes on the one hand, and between attitudes and stockperson behaviour on the other. These results are consistent with the Ajzen and Fishbein (1980) approach, which suggests that personality may infl uence the development of attitudes but not directly affect behaviour.

1.4.3 Work motivation

The extent to which a person applies him- or herself to a task will depend, in part, on the extent to which the person ‘wishes’ to achieve the task. In other

10 Chapter 1

words, a good stockperson is one who is motivated to apply skills and knowledge to the management of the animals under his or her care. What this means, of course, is that the person must be motivated. In general, it is accepted that moti-vation alone is insuffi cient for good work performance; if a person does not have the knowledge, skills or opportunity to perform a job, then motivation will not make any difference. However, if a person does have the knowledge, skills and opportunity to perform the task, what is the role of motivation in professional stockperson behaviour?

Motivation refers to the underlying forces that direct behaviour. Motivation cannot be directly observed, it is inferred from observed behaviour, and in Chapters 3 (Section 3.4.1) and 5 (Section 5.2.1), we discuss motivation in terms of the underlying state of the organism, particularly in terms of hunger or thirst or some hormonal state that directs behaviour. In the case of humans, there is con-siderable argument about the nature of motivation. At one extreme, motivation can be seen to result from the rewards and punishments that a particular behav-iour has produced. In other words, motives develop through learning and depend on the history of rewards and punishments a person has experienced. How behav-iours develop is discussed in some detail in Chapter 5. This approach to motiva-tion has its origins in the work by B.F. Skinner (1969). Skinner’s contribution to our understanding was his discovery of operant (instrumental) conditioning. Operant conditioning is considered in more detail later in this book (e.g. Sections 2.5.1.2 and 5.2.2), but it occurs when a person (or animal) responds naturally to an event and the response is reinforced (rewarded or punished). The person may make several natural or previously learned responses to an event simultaneously, but it is the one that is reinforced that becomes established (following reward) or extinguished (fol-lowing punishment). In the context of a job, rewards can include learned rewards such as feelings of pride or accomplishment. At the other extreme, Maslow, a clinical psychologist, proposed that there is a hierarchy of innate needs, which directs behaviour (Maslow, 1970). These needs, in sequence, were physiological needs (air, food, etc.), safety needs, social needs (presence of others, etc.), esteem needs (recognition by others) and self-actualization (self-fulfi lment). The idea is that people will be motivated by, for example, a need for food or water but, if that need is met, they will be motivated by a need for security. If both of these are met, the person will be motivated by a need for social contact, and so on. Maslow’s hierarchy of needs was based on a number of principles, in particular, the idea that the most basic needs were most fundamental to survival of the organism and that higher order needs could be postponed without detriment to the organism (Fig. 1.2).

In some respects, it is not so important for us to settle on a particular theory of motivation so much as to consider the relevance of motivation, whatever its origin, to st ockperson performanc e in a chieving high anim al wel fare and pro-ductivity. I n fact , there is little sy stematic stu dy of the effect of motivation on stockperson perf ormance. How ever, a st udy carrie d out in India (Singh, 1 983)showed that product ivity, as meas ured by progressiv e farm b ehaviour, was asso-ciated with career interest, u pward strivi ng, at titude towards maki ng money on the fa rm, intelligence, to lerance for wo rk pressure an d punctuality. This suggests that motivational factors, such as upward striving and the need to make money, can contribute to productivity on a farm.

Introduction 11

1.4.4 Absenteeism, job turnover and job commitment

Absenteeism is usually defi ned in terms of number of days absent and frequency of absences. However, it is desirable to distinguish between excused and unex-cused absences. Absenteeism is one of the major sources of disruption and cost to industry, and agricultural industries are no exception. Anecdotal evidence from the pig industry in Australia indicates that turnover rates of 50% per annum for new stockpeople are not uncommon. In the USA, Johns (1987) estimated that absenteeism costs to the US industry at that time could be as high as US$30 billion with about 2–4% of the labour force being away from work on any given day. Despite the fact that most people mention some medical problem as the reason for absenteeism, it is unlikely that this is always the real reason. To use an Australian vernacular expression, people often take ‘sickies’, that is, sick days that are really only days off work when the person is not sick. There is a large number of factors that contribute to absenteeism, but the most consistent factor is level of job dissatisfaction.

Although the relationship between job dissatisfaction and absenteeism is consistent, it is not a particularly strong relat ionship. Jew ell and Si egall (199 0)have re viewed factors associated with a bsenteeism. The only persona l factor t hatis con sistently related to abs enteeism is gend er. Women are absent from workmore th an men. Th e explan ation for this comes i n part from the fact t hat womenhave f amily responsibili ties that may requir e them to be absent. Another contrib-uting factor m ay be that women generally occupy lower-status jobs than men with lower levels of job satisfaction. Organizational variables, such as how boring the job is, the size of the organization and, perhaps, other characteristics of the work situation, may all contribute to absenteeism.

Turnover usually refers to workers leaving the organization and being replaced by others. However, it is sometimes the case that the organization may encourage the de parture of a worker because of organizational change or because th e perf ormance of t he worker is n ot satisfactory in some resp ect.Campion (1991) has defi ned turnover as individual motivated choice behaviour. He argues that there may be a number of motivations underlying employee departure and, in order to satisfactorily analyse such data, it is essential that detailed records of employees’ reasons for departure are obtained. Because

Basic needs

Higher-order needs

Basic physiological needs

Safety needs

Social needs

Ego (esteem) needs

Self-actualization(fulfilment) needs

Fig. 1.2. Maslow’s (1970) hierarchy of needs.

12 Chapter 1

good records of workers’ reasons for departure are often not kept, it can be diffi cult to decide whether the departure was voluntary or involuntary. Job turnover is regarded as independent of absenteeism. In general, correlations between absenteeism and turnover are close to zero. Rusbult et al. ( 1988) sug-gest that decline in job satisfaction is associated with staff turnover. They argue that people with high levels of job satisfaction before it declined, would be more likely to stay in the job and try to change the situation. Those who entered the job with low levels of satisfaction would be more likely to leave if satisfaction declined.

Most research has invest igated the relationship between job satisfaction and turnover, and other variables have not received much attention. However, there is some evidence to suggest that i t is the better perfor mers w ho stay in the orga-nization. Jewell an d Sieg all (1990) have reviewed the r esearch in t his ar ea. They concluded that th ere were two key components that contributed to the intention of a worker to seek a new job. The fi rst of these is the extent to which a worker is able to move to a new job. This capacity for movement is determined by the worker’s task-relevant abilities, self-esteem and recent job-seeking experience. The second component is the worker’s desire for movement to a new job. This is determined in part by the nature of the current job, the opportunity for pro-gression in the organization and the rewards associated with the job. There appear to be no data on the causes of job turnover that specifi cally relate to agricultural industries.

Job commitment refers to the behaviour opposite to that of leaving the orga-nization. Job commitment is a product of personal variables and the characteris-tics of the work situation. There is evidence to suggest that people with a strong work ethic, that is, those who have a strong degree of job commitment, are also those who have a high degree of job satisfaction. Chusmir (1982) has proposed a model of job commitment in which personal factors such as gender, age, edu-cational level, work attitudes and needs all contribute to job commitment (Fig. 1.3). These personal variables are moderated by family characteristics, including marital status, spouse support and earnings, and satisfaction with family life. They are also moderated by job circumstances that include job satisfaction, meaningfulness of work, utilization of skills, psychological factors and other non-motivational job factors. All of these contribute, in turn, to the person’s attitude towards the job, satisfaction of needs and work ethic, which together contribute to job commitment. It is clear, therefore, that there is a complex set of factors that contribute to job commitment. There has been no systematic investigation of factors associated with job commitment in agricultural industries.

1.4.5 Job satisfaction

Job satisfaction is an attitude towards work that is based on how the person evaluates the work. In other words, job satisfaction refers to the extent to which a person reacts favourably or unfavourably to his or her work. In general, job satisfaction is thought to derive from the extent to which a person’s needs or expectations are being met by the job. If job satisfaction is related to work

Introduction 13

performance, then this provides a clear role for motivation in contributing, indirectly, to work performance.

Unfortunately, despite the widespread use of the construct in industrial/orga-nizational psychology, well-substantiated theories of job satisfaction are diffi cult to fi nd. Nevertheless, research shows that absenteeism and job turnover are asso-ciated with poor job satisfaction. In a review of studies from the Soviet Union, Phillips and Benson (1983) reported that there were numerous reports, both from the Soviet Union and Western countries, showing a clear correlation between work dissatisfaction and the probability of switching jobs. They reported, in some detail, research from Ivanov and Patrushev (1976) into the ‘social fac-tors of increasing the productivity of labor in agriculture’. In general, 74% of farm workers reported that they were satisfi ed with their work, and job satisfaction varied directly with working conditions and skill level. This research showed that in the Moscow region about 30% of workers had decided to or were nearly resolved to switch jobs; 44% of young people in general were similarly disposed, as were 61% of the young people in animal husbandry. This research also showed a strong relationship between degree of work satisfaction and the desire to achieve better results. A strong inverse relationship was found between num-ber of days missed and work attitude. A satisfi ed worker missed an average of

Sex

Background Family characteristics

Perceived rolebehaviour andattitude

Job commitment

Job circumstances

Age Marital status

Personalinfluence

Externalmoderating influences

Moderatedperceptions

Jobcommitment

Child responsibilitySupportiveness ofspouseSpouse’s earnings

Job satisfaction

Sex–role conflict

Propensity to staywith jobPropensity to becomedeeply involvedwith job

Satisfaction ofneedsWork commitmentMeaningfulness of work

Utilization of skillsPsychological jobfactorsNon-motivational jobfactors

Satisfaction with familylife

Education levelBirth orderParents’ socialclass

Centrality of work

High-ordergrowth needs

Extrinsic needsPersonal values

Attitudes andvalues

Intrinsic needs

Fig. 1.3. A model of job commitment (adapted from Chusmir, 1982, with permission from the publishers).

14 Chapter 1

4.3 days each year, while a dissatisfi ed worker missed an average of 10.3 days annually.

1.5 Evaluating Job Performance

Without doubt, the key feature of any employee is how well he or she does the job. There are two aspects to work performance, quality and quantity. Blumberg and Pringle (1982) have proposed a model of work performance that identifi es three classes of contributing factors (Fig. 1.4).

The fi rst factor is capacity to do the job, the second is opportunity and the third is willingness. Capacity includes variables such as skills, health, ability and knowledge, while willingness includes motivation, job satisfaction and work atti-tude, and opportunity includes working conditions, actions of co-workers and organizational policies and rules. This is not an exhaustive list of variables in Blumberg and Pringle’s model, but it should be evident that all of the variables we have discussed so far contribute to work performance under the model.

Many of the factors just discussed were addressed in a study which identifi ed the characteristics of pig stockpeople that predict work performance (Coleman,

Capacity

Performance

Willingness Opportunity

Ability, age, health,knowledge, skills,intelligence, level ofeducation, endurance,stamina, energy level,motor skills

Motivation, job satisfaction,job status, anxiety, legitimacyof participation, attitude,perceived task characteristics,job involvement, ego involvement,self-image, personality, norms,values, perceived role expectations,feelings of equity

Tools, equipment, materialsand supplies; workingconditions; actions of co-workers; leader behaviour;mentorism; organizationalpolicies, rules and procedures;information; time; pay

Fig. 1.4. A model of work performance (adapted from Blumberg and Pringle, 1982, with permission from the publishers).

Introduction 15

2001; Carless et al., 2007). A total of 144 inexperienced stockpeople partici-pated in this study in which stockpeople, at the commencement of employment, completed a set of computerized questionnaires, which included measures of personality, motivation, turnover potential, performance potential, attitudes and empathy toward pigs. Those stockpeople who remained at the piggery for 6 months were studied. Stockperson performance was assessed at this time using ratings of stockperson behaviour towards their pigs, technical knowledge, consci-entiousness, satisfaction and intention to leave the job soon (intention to turn over) (Table 1.2). Stockperson behaviour towards their pigs, technical knowledge and work ethic were directly assessed by an independent observer. A supervisor report of satisfaction and conscientiousness was used to measure these aspects of stockperson performance. Intention to turn over was assessed using the Michigan Organizational Assessment Questionnaire (Cammann et al., 1979).

Females were rated as more conscientious by their supervisors, and were independently rated as having greater technical knowledge and a better work ethic compared with males. The self-reported PDI Employment Inventory (PDI-EI) Performance measure (Personnel Decisions Inc., 1996), assessed on com-mencement of employment, was related to the independently observed behaviour of stockpeople towards pigs. A person scoring high on this measure is likely to adhere to rules, show stability of behaviour, take care while performing tasks and take responsibility. In contrast, the PDI-EI Tenure measure correlated

Table 1.2. Predictors of pig stockperson performance (from Coleman, 2001).

Stockperson performance variableb

Supervisor rating Independent observer rating Stockperson

Predictor variablesaConscien-tiousness Satisfaction

Technical knowledge

Behaviour towards pigs

Work ethic

Intentionto turn over

Sex 0.29* 0.10 0.45** −0.03 0.29* −0.06PDI-EI

performance0.32* 0.39** 0.27 0.29* 0.22 −0.35**

PDI-EI tenure 0.03 0.18 0.13 0.15 0.05 −0.28*Positive attitude 0.27* 0.26 0.23 0.23 0.11 −0.37**Empathy affect 0.11 0.12 0.27 0.37** 0.19 −0.41**Empathy attribution 0.13 0.05 0.33* 0.30* 0.17 −0.39**

Signifi cant correlations (* = P < 0.05 and ** = P < 0.01) indicate associations between the two variables; n = 50–64 stockpeople.aPredictor variables are as follows. Sex: 1 = male, 2 = female; PDI-EI performance: expected job performance; PDI-EI tenure: likelihood of remaining in a job for at least 3 months; Positive attitude: positive attitude toward pigs; Empathy effect: concern about animals’ feelings; Empathy attribution: belief that animals are like humans.bStockperson performance variables are as follows. Conscientiousness: how conscientious supervisors judge stockpeople to be; Satisfaction: how satisfi ed supervisors judge stockpeople to be; Technical knowledge: independent observer’s rating of stockperson’s technical knowledge; Behaviour towards pigs: independent observer’s rating of stockperson’s handling of pigs; Work ethic: independent observer’s rating of stockperson’s work ethic; Intention to turn over: likelihood of seeking a new job in the next year.

16 Chapter 1

with intention to remain in employment over the next year. A positive attitude towards the characteristics of pigs correlated signifi cantly with conscientiousness and intention to remain in the job. Finally, empathy towards animals correlated with technical knowledge, behaviour towards pigs and intention to remain in the job. Another fi nding was that 50% of new stockpeople left their jobs within 6 months of employment.

One of the important features of this study is that all the measures of stock-person characteristics were taken in an initial interview and the performance measures 6 months later. Therefore, these measures of stockperson performance have the potential to be used to identify potentially good stockpeople.

As is always the case in research, this study leaves some tantalizing ques-tions. Why do women appear to perform better than men? Is it because they self-select in animal care jobs? Do (male) raters rate females higher in gender-stereotyped areas of behaviour? Are females better stockpeople and are they more conscientious? Notably, being female was not associated with better observed behaviour towards pigs.

Although a ‘big fi ve’ measure of personality was used in this study, no consistent relationships between personality and stockperson performance were observed. However, neuroticism was associated with observer rating of work ethic (correlation coeffi cient of 0.39, P < 0.01) and a trait unique to the Kline and Lapham (1991a,b) Professional Personality Questionnaire, tender mindedness, was related to intention to turnover (correlation coeffi cient of 0.28,P < 0.05).

As we have discussed in this chapter, there are some data in support of the idea that personal characteristics are related to work performance in the agricul-tural industries, and there will be further review of relevant studies later in this book. Nevertheless, it should be clear by now that it is possible to characterize a stockperson in terms of the person’s motivation, needs, work ethic, job satisfac-tion, abilities, skills and knowledge. Later on, a comprehensive model of factors contributing to the stockperson’s performance on animal welfare and productiv-ity will also be outlined.

1.6 The Stockperson’s Role: Preparing the Stockperson for the Task

1.6.1 Training of stockpeople

If the question was asked of industry personnel about the role of the stockperson in livestock production, it would probably be generally stated that stockpeople have a critical role. However, just how seriously and extensively this subject is taken and thus acted on are questionable. Stockpeople often underestimate their value and contribution to livestock production (English et al., 1992), and it appears that supervisors, managers, farm owners and industry leaders may also undervalue the contribution of stockpeople. Research and development in live-stock production have focused on technological innovation, especially in areas such as housing, nutrition, breeding and health, and most of the industry training

Introduction 17

has generally targeted training supervisors and managers in these new technolo-gies. In comparison, training of stockpeople has often been neglected, perhaps refl ecting to some extent the attitude of senior industry personnel to stockperson training. The attitude of stockpeople to training, and the high staff turnover rates in some livestock industries, may also be considered by some industry personnel to reduce the imperative for training.

The less than satisfactory opportunities for training stockpeople and the poor attitudes of many industry personnel to staff training also refl ect a lack of appreciation of the importance of training in the industry. For example, while well over half of the employers and employees surveyed in a study of the Canadian pig industry recognized that specifi c skills were required to effectively conduct a number of routine tasks in pig production, surprisingly less than half of them generally believed that training was required to develop these skills (survey results quoted by English et al., 1992).

Several authors have commented on the parlous state of stockperson train-ing in a number of livestock industries. Lloyd (1974) reported that while 86% of poultry stockpeople had no previous experience with poultry, only 4% of stock-people had attended any form of training courses in the previous 2 years. Segundo (1989) found that 87% of 15 piggeries in Scotland had no off-site train-ing available and 53% had no on-site training for stockpeople. Only 13% of Australian pig stockpeople surveyed by Kondos (1983) had received some orga-nized technical training.

With the rapid technological development in livestock production, it is imperative that technological training of its staff occurs. This training, which should contain both the appropriate theoretical components and practical train-ing, should target all farm staff, not just senior staff, and should be an ongoing process. For stockpeople, this training obviously needs to focus on technical skills and knowledge, occupational, health and safety, and equipment use and main-tenance, and perhaps, where appropriate, on interpersonal skills such as working effectively in teams. The widespread availability and the effectiveness of both on-site and off-site training of stockpeople are highly variable.

Where available, agricultural training for stockpeople has been traditionally college based, with the focus on classroom teaching and often less than ideal opportunities for experience in commercial-like conditions. One could also ques-tion the relevance of some of these programmes in addressing the requirements of industry and stockpeople themselves. Many of these courses are particularly aimed at future supervisors and managers, with little relevance to base-grade stockpeo-ple. Apprenticeship schemes offer both theoretical and practical training, but at times the focus here is also on training future supervisors and managers. Many livestock industries in many countries are well serviced by industry days and semi-nars offered by government or university agricultural agencies and agricultural product companies, but these programmes often target senior farm staff.

There appears to be a growing international trend towards providing more training for stockpeople on site and less in agricultural colleges or technical education centres (Fig. 1.5). The costs and logistics involved in replacing staff and releasing staff to travel to training venues may certainly infl uence industry’s support for this development. On-site training may also be more attractive to

18 Chapter 1

industry because it occurs in a commercial setting, in which relevant practical sessions can be offered, and because the training can occur in a familiar less formal training setting that is often in contrast to the more formal classroom set-ting. This may be particularly relevant as the latter setting may be more threaten-ing and intimidating for many base-grade stockpeople who may have previously had diffi culty with a formal educational system.

Organized discussions between experienced and inexperienced staff have been and continue to be a common on-site training activity (Cleary, 1990). However, large corporate farming companies appear to be becoming increas-ingly active in developing and introducing local outcome-oriented training for their stockpeople. The on-site training may vary from simple training activities in which the more experienced staff, not necessarily with any formal instruction on teaching techniques, act as mentors for inexperienced staff, to the situation where structured in-house training programmes, conducted either by experienced staff or trained human resource management staff, are undertaken and these training activities are linked with competency assessments and remuneration (Miller, 1995). Indeed, some of these in-house training schemes are very progressive with well-developed syllabuses and involve both classroom instruction, by either local or external trainers, and skills training under commercial conditions.

Low levels of training skills by senior farm staff can inhibit learning in train-ing programmes conducted on site, and a lack of suitable reference material for in-house training appears to be a limitation for many livestock industries. Staff cultural attitudes about education may also inhibit learning in both on-site and off-site settings. Multimedia training packages, in which information is presented in several forms, and which allow stockpeople to individually interact at their

Fig. 1.5. On-site training of new staff is becoming increasingly common in the livestock industries.

Introduction 19

own pace with the teaching programme in an informal non-threatening manner, may be more conducive for learning (Hemsworth and Coleman, 2009). This approach also has the potential to relieve managers or senior stockpeople of formal didactic teaching sessions and allow them to be used more effectively as facilitators in group sessions where experiences can be shared and views devel-oped in an advantageous setting. Furthermore, these interactive multimedia pro-grammes allow staff to be released for training without major disruptions to work, and to progressively complete training modules and be assessed at times that are convenient for management. Support by government and commercial services in the provision of interactive user-friendly multimedia packages on key technical issues, and the provision of trainers or opportunities to train in-house trainers to facilitate the training and assess staff undergoing this training, will make training more accessible to industry.

A series of multimedia programmes using the model developed in this book have appeared. These are the ProHand packages that are designed for stock-people on farm and in abattoirs in the pig, dairy and red-meat industries, and a similar set of packages developed in Europe as part of the EU Sixth Framework Programme for stockpeople in the pig, poultry and cattle industries. While there is a limited number of reports on the effectiveness of these packages in the livestock industries, they were rigorously evaluated during development (Hemsworth et al., 1994a, 2002; Coleman et al., 2000), and the EU packages have been subjected to a preliminary fi eld evaluation with promising results (Windschnurer et al., 2009b; Ruis et al., 2010). These are discussed in more detail in Chapter 7 (Section 7.4.3).

1.6.2 Selection of stockpeople

Selection is the process of matching a person to a particular job. This matching process can involve assessing applicants with respect to many of the variables discussed earlier in this chapter. For example, because work preference, skills, knowledge, work ethic and, perhaps, personality factors may all contribute to worker performance, high job satisfaction and capacity for development in the job, it may be appropriate to measure all of these factors as part of a selection process. However, in general, demographic data and previous experience may also be useful in screening out potential employees with a low chance of success. Usually a face-to-face interview with an applicant can also provide useful data if the interviewer is experienced in selecting people for a particular industry or profession. As an example of the advantages of a selection procedure, Borofsky and Smith (1993) found that the use of such a procedure resulted in a reduction in turnover, accidents and absenteeism.

It is important that the person carrying out the selection process should behave ethically towards job applicants. There are three basic facets to this responsibility: (i) the selection offi cer should ensure that the selection processes that he or she uses are as up to date and reliable as possible; (ii) information provided by the applicant should be kept confi dential and should not be used for any other purpose without explicit permission from the applicant; and

20 Chapter 1

(iii) the selection offi cer should evaluate the effi cacy of the selection procedure and ensure that the organization is able to accurately determine the costs and benefi ts of the procedure.

Despite the fact that we can identify some characteristics that predict good stockperson performance in intensive farming industries (Coleman, 2001, 2004; Carless et al., 2007), this knowledge has not been incorporated into a selection procedure that has been fully fi eld tested. Some suggestions to indicate how this might be done will be discussed later in the book (Section 8.5).

1.6.3 Human resource issues

In non-agricultural areas, the need to retain desirable staff and to foster high standards of work ethic are generally well recognized and addressed. The man-agement of the human resources in livestock production at the level of the stock-person appears to be highly variable. The earlier comments about lack of recognition of the stockperson’s role in animal performance and welfare, the small-scale operations that are often prominent in livestock production and the lack of business management skills seen at times in livestock production may contribute to this weakness.

One of the minor aims of this book is to foster recognition of the role that the stockperson plays in livestock production. Recognition of this contribution should facilitate improvements in stockperson selection and training, and appropriate fi nancial and personal rewards for stockpeople, all of which should contribute to higher job performance and staff retention rates. The opportunities for staff selection and training are explored in detail in the latter half of this book, after the interrelationships between the characteristics of stockpeople and the behaviour, performance and welfare of farm animals are reviewed.

1.6.4 The stockperson and animal welfare

Because of the important infl uence of the stockperson on the welfare of farm animals and the widespread community interest in animal welfare, its measure-ment and its consequences, Chapter 2 discusses some of these welfare issues that are critical to the sustainability of livestock production. Indeed, general com-munity concern about animal welfare problems in a livestock industry, through consumer boycotts, requirements of food processors and retailers, or govern-ment regulation, may infl uence the ability of the industry to produce or sell its animal products. It is important to discuss these issues in a book about stock-people because the stockperson plays an integral role in safeguarding the welfare of both intensively and extensively farmed animals. The importance of this topic should not be underestimated because of its implications for not only the animal but also the sustainability and economics of the livestock industries.

© CAB International 2011. Human–Livestock Interactions, Second Edition (P.H. Hemsworth and G.J. Coleman) 21

2 Farm Animal Welfare: Assessment, Issues and Implications

2.1 Introduction

There is considerable public interest in animal welfare and most people believe that animals, including farm animals, should be not be subjected to pain or severe discomfort (Fraser and Broom, 1990; Coleman, 2008). The livestock industries are also sensitive to the issue of the welfare of farm animals. This arises, in part, because a deterioration in the welfare of animals is often associ-ated with reductions in individual animal performance. Furthermore, local and international agricultural markets are fi ercely competitive, and in addition to the necessity of technological improvements, development of new products and marketing expertise to maintain competitiveness and increase sales, livestock industries need to project a welfare-friendly image of their products to maximize their marketing advantage. Food processors and retailers may act on consumer and public concerns to restrict or eliminate contentious welfare issues in farms that supply their animal products. Concerns about the welfare of farm animals in a particular industry may also infl uence the buying behaviour of current or potential consumers of the product from that industry. Animal welfare is a social issue often discussed in the public domain and thus the public may be infl uential, via government decisions, in determining animal welfare standards (Coleman, 2008). Codes of practice or government regulations may restrict specifi c practices in a particular livestock industry that the general community fi nds objectionable on welfare grounds. The image of a welfare-friendly product requires farming prac-tices that minimize the risk to animal welfare and the provision of objective infor-mation that positively infl uences the public’s and consumers’ beliefs about the welfare implications of the farming practices that produce the product concerned. Indeed, the results of welfare research on farming practices will infl uence both industry practices and the public’s and consumers’ perception of the product.

While not always well and widely recognized, stockpeople have a critical role in ensuring that the welfare of their livestock is not compromised. The general

22 Chapter 2

public and, indeed, many within the livestock industries would contend that housing systems pose the main risk to animal welfare in intensive livestock sys-tems. However, in contrast, the contribution of stockpeople to animal welfare in both intensive and extensive systems is poorly recognized. This chapter intro-duces the topic of the human–animal relationship in livestock production and its consequences for the welfare of farm animals. Because of its implications for the animal, the stockperson, the livestock industries and the general public, a detailed discussion of animal welfare is presented in the chapter. This discussion of animal welfare focuses on its defi nition, its assessment and its consequences for the animal.

2.2 Human–Animal Relationships in Livestock Production

Humans working closely with their livestock develop relationships with their animals often not dissimilar from those that develop between humans and com-panion animals. The similarity of the two relationships, particularly in terms of their strength and quality, may surprise many people. The fact that most of us view our pets as companions, often with considerable affection, is well recog-nized. However, while the symbiosis or interdependence of this relationship between humans and animals in the livestock industries is recognized to some extent (Hemsworth, 2007a), the general public probably does not recognize the general fondness and friendship that many farmers or stockpeople have for their animals. The general public’s romantic view of livestock farming involving the shepherd looking after the ‘fl ock’ does not extend to modern farming systems and, in fact, the human–animal relationship in modern farming is often viewed by the general public as an exploitative one by humans in which little or no regard is afforded to the welfare of these farm animals. This is a simplistic and inaccurate way of describing the human–animal relationship in modern livestock production because it fails to recognize the interdependence between stock-people and their animals.

As in the early phases of domestication, the relationship between humans and animals in modern livestock production has components of symbiosis in that, in return for the animal products that they provide for humans, the animals are maintained and cared for by humans. Undoubtedly, the satisfaction of eco-nomic demands is the main feature of this human–farm animal relationship today. A similar, needs-based, relationship often exists between humans and pets in the general community. For example, caged birds and fi sh in aquariums in many households are predominantly kept for entertainment and aesthetics. Despite the fact that reciprocal need has been, and still is, a feature of the rela-tionship between humans and domestic animals, our ethics concerning the care and use of animals in general have greatly improved in recent times.

The popular view of modern farm animals being exploited has probably arisen for several reasons. Economic pressures have led to the intensifi cation of farming, with large numbers of animals, often under tightly controlled environ-mental conditions, and managed by few stockpeople who often may be employed with limited animal husbandry experience or training. This intensifi cation of

Farm Animal Welfare 23

animal farming was developed and introduced to optimize animal growth, repro-duction and health, the trend towards intensifi cation has probably been infl uen-tial in shaping public opinion. Unfortunately, at times, these management innovations, particularly the housing developments, were introduced without adequate knowledge of or regard for the animal’s ability to adapt to the changes and, consequently, animal welfare has sometimes been compromised. Further-more, the general public’s requirement for cheaper and safe farming products has dictated the search by agricultural scientists, veterinary scientists and farmers for maximum animal output at the minimum cost, which invariably has pro-moted a view of exploitation of animals. The enormous difference today in the lifestyles and experiences of people in urban communities compared with those in rural communities does not assist the general public’s awareness and under-standing of what has happened and is happening in livestock production, and this, in turn, often creates suspicions about conditions for farm animals. The farming communities around the world can certainly improve their efforts in educating the public about modern farming practices and the reasons for these practices (Coleman, 2010).

Most stockpeople recognize that deterioration in the welfare of their animals may result in depressions in the productivity and health of individual animals, with potential adverse consequences for profi tability. Stockpeople obviously consider their animals as resources, but have long treated and viewed their animals with affection as companions (Fig. 2.1) (Fraser and Broom, 1990). This opinion may be diffi cult for some outside livestock production to accept, and we

Fig. 2.1. Stockpeople have long treated and viewed their animals with affection and have often considered them as companions (photograph courtesy of Wageningen UR Communication Services).

24 Chapter 2

will return to this point later in the book. The quality of these human–animal relationships, measured in terms of stockpeople’s beliefs about their animals and their behaviour towards these animals, as well as the behaviour of these farm animals in the presence of humans, are considered in detail in a number of animal industries.

It is indisputable that there have been cases of animal abuse and cruelty in livestock production, just as there have been cases involving companion ani-mals, and these are likely to continue. However, recent developments, particu-larly in terms of awareness and recognition of the subject, continuing research, the introduction of improved codes of practice, legislative change incorporating animal welfare standards and animal welfare auditing schemes, indicate the promise of ever-increasing improvements in farm animal welfare. A similar situa-tion exists with the welfare of companion animals, although one could argue that, in general, the developments have been less. Improvements in a range of areas, such as education of the community in terms of its responsibility, suitability of par-ticular species as pets and the proper care and maintenance of pets, are likely to reduce the incidence and magnitude of abuses that have occurred in the past. Community pressure on the welfare of laboratory animals has also resulted and is likely to continue to result in improvements in the care and use of these animals.

While physical, social, nutritional, disease and climatic factors may infl uence the welfare of farm animals, the competency and motivation of the stockperson in the care and management of the animals are critical ingredients in determining their welfare. Ultimately, it is the stockperson that is charged with ensuring that a particular farming system is operated properly and diligently. Weaknesses in the motivation of the stockperson to follow farm protocols for animal care and main-tenance and to monitor and promptly address welfare issues arising in his or her area of responsibility will place animal welfare and productivity at risk. Media reports of poor welfare in the livestock industries often relate to bad behaviour by stockpeople (arising from negligence or lack of care) as well as poor animal facilities. Furthermore, the large number of animals under the care of an indi-vidual stockperson in modern farming systems, together with any weaknesses in the competency and diligence of an individual, may have serious effects on animal welfare and productivity. This arises not only because of the number of animals involved but also because of the almost total dependence of animals on humans in some of these systems. We will contend throughout this book that while the attitude of the stockperson towards his or her animals is highly infl uential in deter-mining animal welfare and productivity in modern livestock production, extensive research on this subject has not been translated widely into industry practice. As the dependence of animals on the stockperson for their care and maintenance increases, the infl uence of the attitude of the stockperson on the welfare and productivity of modern livestock increases correspondingly.

2.3 Animal Welfare and the Debate

Animal welfare is a highly emotive subject and most of us are not spared the emotions that the topic can create. Irrespective of our lifestyle, most people in

Farm Animal Welfare 25

modern society are regularly confronted with cases of animal suffering. Images in wildlife documentaries portraying the survival of the fi ttest individuals and the death of the weakest generally strike an emotive chord in most of us. The news media are not reticent in extensively covering animal welfare abuses because of the public interest that these stories generate. Most people experience fi rst hand, at some stage in their life, the suffering of an ageing or ill pet and the diffi culty in assessing the pain or suffering of a loved animal. The emotional effects of our possible loss may accentuate our concern. In general, we feel a degree of empa-thy for our pets; we tend to think of their suffering in human terms and try to deal with the situation much as we would with another human. In fact, there is abun-dant evidence in the literature for pets being treated as family members (Voith, 1985; Cain, 1991; Bodsworth and Coleman, 2001; Walsh, 2009), even to the extent that jealousies and confl ict arise because a family pet competes for our affections (Cox, 1993).

Nevertheless, animal welfare is a controversial subject because of its subjec-tivity. A range of views on the subject, often based predominantly on value judgements, exist within the general community leading to marked and often extreme attitudes on animal welfare issues. The action of some welfare groups in lobbying and boycotting specifi c animal industries or practices is indicative of their views and the strength of these views. Equally, there are strong community views about the adverse effects of domestic animals on humans. Dog bites receive wide publicity and, in Australia, so does the predation on native wildlife by domestic cats. This, in turn, leads to legislative change restricting the move-ments of these animals, with possible welfare implications. Decisions about ani-mal welfare are obviously morally and politically important. Failure to address the community’s views may result in the general community, via governments, restricting people’s use of and access to animals.

The public interest in the welfare of farm, laboratory and companion animals has rapidly increased in Western society during the last 40 years. The writings of a number of authors (e.g. Harrison, 1964; Singer, 1975) were impor-tant in raising this as a topic in the general domain, and the ensuing interest and debate clearly focused, for example, the farming community and, in turn, the sci-entifi c community on the topic of animal welfare. Most people, irrespective of their involvement in livestock production, would probably agree that the debate hashad positive effects on the welfare of farm animals and, indeed, animal welfare remains a topical subject for those with interests in livestock production.

2.4 The Community’s Views on Animal Welfare Issues

Most people accept that humans have a moral obligation towards farm, companion and laboratory animals. In addition to the undeniable benefi ts that these animals provide to humans, their domestication has increased their depen-dence on humans and thus necessitates this obligation. However, what is at question for most people is the extent of this obligation, particularly in relation to the standards of welfare that society should provide to these animals. A consensus on this aspect of fair and humane treatment is diffi cult to achieve

26 Chapter 2

when such diverse views exist in the general community. In many Western coun-tries there are several mechanisms used to protect the welfare of farm animals (Barnett and Hemsworth, 2009). In some countries there is animal welfare legis-lation, which may be limited to the protection of animals from cruelty or may prescribe mandatory standards. Furthermore, many Western countries have introduced codes of practice, and while these codes generally rely on voluntary compliance, there are some examples where the codes or some code provisions are incorporated into legislation. In other Western countries, livestock industries provide and recommend animal care standards. There is also a range of animal welfare monitoring schemes in many countries and although the majority of these schemes are voluntary rather than legislated, some are nevertheless enforced through a number of customer requirements that industries must meet to gain access to markets. However, in many other countries such protection does not exist.

An individual’s decision on the acceptability or otherwise of a specifi c animal use can be a diffi cult and complex choice. Decisions on specifi c animal use are affected by a number of considerations, including scientifi c information on the harms and benefi ts to the animal caused by the animal use (Hemsworth et al., 2007). Furthermore, while attitudes to animals at the individual level infl u-ence how people behave towards animals, attitudes to animals at the community level can also infl uence the development of animal-related policy and legislation. Attitudes to animals appear to be particularly affected by people’s affective or emotional responses to animals, as well as their perception of the utility or instru-mental value of the animal (Serpell, 2004). Individuals may also judge a decision or choice of an animal use on the basis of its adherence to reasons or moral principles such as: enhancement of personal character (virtue ethics); support for self-interest (egoism); consequences of the choice for everyone, not just them-selves (equal consideration or utilitarianism); or that good results do not justify using evil means to violate an animal’s rights (rights-based justifi cations of equal consideration). For many, modern animal ethics embraces not only our duties and obligations to animals, but also our duties and obligations to animal users and society in general (Levy, 2004). Thus, the impact of the animal use on the animal owner, the environment and the economy may also affect these deci-sions. Nevertheless, through broad stakeholder consultation, governments and others set farm animal welfare recommendations and standards based, to varying degrees, on these considerations.

Science, therefore, has a critical role in underpinning our decisions on animal use and the attendant conditions and compromises. In consequence, biologists are charged with the responsibility of establishing the facts on how animals biologically respond to the practices under question, whether they relate to farming, laboratory or general community uses of animals. Gaining a consen-sus on the welfare implications of a specifi c animal use would appear to be an easier task to achieve among scientists than within the general community. However, this is not always true. As will be discussed shortly, confl icts have arisen in scientifi c circles for several reasons: the defi nition of animal welfare and thus the methodology used to assess animal welfare varies among scientists; only limited animal welfare research has been conducted to date; and the results of

Farm Animal Welfare 27

this welfare research are often complicated and, at times, appear contradictory. This is a limitation, because one of the important steps in developing defensible policies on animal care and use is to assemble factual information on the ani-mal’s biological responses to the particular system or treatment. Without factual information, it is diffi cult for a society or groups to develop a consensus on a defensible policy on an animal welfare issue.

Although science has an important role in providing sound defensible infor-mation on how animals respond to a specifi c practice, ultimately it is an ethical decision by the general community that will determine acceptable welfare stan-dards for farm, companion and laboratory animals. The community, via govern-ments and, at times, food processors and retailers, provides the livestock industries with a ‘licence’ to farm in which standards, including animal welfare standards, are prescribed. However, the development of a clear consensus on an ethically and scientifi cally defensible philosophy on animal welfare is obviously diffi cult. A society’s attitudes to the use of and obligations to farm, companion and labora-tory animals are extremely disparate, infl uenced by demographic factors, religion and culture, and vary over time with economic and ideological changes.

As Teutsch (1987) reported, there are clear within- and between-country variations in the attitudes of people to their obligations towards animals. The Eurobarometer surveys (European Commission, 2007) provide a snapshot of attitudes to animal welfare across the European community. Notwithstanding these variations, Coleman (2008) concluded that the community generally con-siders farm animal welfare to be important, with laying hens seen to be at the greatest welfare risk, followed by pigs. Dairy cows are seen to be at lower risk and there appear to be no data available on sheep, beef cattle or some other farmed species, such as turkeys, goats, etc. Coleman also concluded that, although there is variability among the European Union countries, most countries throughout the Western world show similar patterns of attitudes to farm animal welfare.

The debate in the general community about the care and use of animals in livestock production, research and modern society has increased over time and there is continuing widespread pressure for legislation and regulation, particu-larly in developed countries, that will protect animals from pain and stress. The problem in developing such guidelines and regulations is to defi ne what consti-tutes good welfare or well-being for the animal. Without such a defi nition, legisla-tion concerning the care and use of farm, companion and laboratory animals will be developed based on emotions without serious regard for objective data and, notwithstanding the good intentions of most people involved in the process, could jeopardize animal welfare.

In addition, there is evidence that community knowledge of animal husbandry practices is limited and that there is both a need and an opportunity to address this properly, not only for the general community but also for specifi c stakeholders, such as producers, retailers and regulators. Public attitudes have a signifi cant role in determining how people behave, both as consumers and as citizens. This, in turn, affects the commercial viability and even the sustainability of animal industries (Coleman, 2010). Furthermore, public attitudes about ani-mal welfare are often based on limited knowledge, and the public’s beliefs are largely acquired from the mass media, perhaps fi ltered by opinion leaders.

28 Chapter 2

Therefore, there is a need to ensure that the community is well informed because community views affect decision makers at the political, regulatory and retail levels, with considerable consequences for all. The process of informing the com-munity necessarily involves changing beliefs and, to this extent, involves persua-sion. Target groups include stockpeople, animal handlers post-farm, including transport drivers and abattoir workers, animal laboratory technicians, owners of companion animals, legislators and regulators, retailers and the general commu-nity. Mass communication strategies tend to produce change over a long time frame. However, targeted education strategies, in particular to stakeholder groups, may be tailored to the individual and therefore produce change over a relatively short time frame. In addition, government has a role in disseminating the values that should govern animal welfare, and this should be a refl ection of both community views and current thinking among ethicists. Probably the only target group that can be educated in the sense of knowledge transfer are stu-dents. Schoolchildren, particularly those living in urban areas, have little expo-sure to farm animals, and there is a need to provide them with the basic facts about where their food comes from. Such knowledge should make children less susceptible to extreme views about animal welfare in farm animals because the knowledge they have acquired will tend to make them more or less resistant to inappropriate persuasion (Coleman, 2010).

As discussed earlier in this section, there is considerable public interest in animal welfare and so it is useful to briefl y review some of the most controversial welfare issues in livestock production. Confi nement of animals appears conten-tious for many people, possibly because of concerns for reduced animal welfare with restrictions in space, social contact and choice of stimuli for interaction. The public appears to be more concerned about the welfare of pigs and poultry than other farm animals (European Commission, 2007), presumably because their housing is viewed as confi nement. However, the design features of group hous-ing systems, such as space, group size, feeding system and fl ooring, are important and markedly affect animal welfare (Barnett et al., 2001; Barnett and Hemsworth, 2003). There is no doubt that sectors of the general public are increasingly ques-tioning the welfare impact of and the need for some husbandry procedures, particularly surgical interventions that are likely to cause pain (Coleman, 2008). While transport is always likely to be stressful, it can generally be managed by ensuring that animals are ‘fi t’ for the journey, and particularly by ensuring that young animals or animals in poor (e.g. drought) condition are handled appropri-ately, only travel the minimal distance possible and are appropriately scheduled for slaughter if transported to abattoirs (Barnett et al., 2008). The impact of human factors has received less attention from a welfare perspective than hous-ing and husbandry, even though as reviewed in this book, stockpeople have a major impact on the welfare of their livestock.

2.5 Defi ning and Measuring Animal Welfare

Because of the widespread use of the term welfare in a number of scientifi c disciplines, in philosophy and in the general community, defi nitions of welfare

Farm Animal Welfare 29

vary considerably. Even within science there is considerable uncertainty sur-rounding the concept of animal welfare (Fraser, 2003; Sandøe et al., 2004) and, consequently, around the methodology used to assess or judge the welfare of animals. For example, many scientists defi ne and measure animal welfare on the basis of normal biological functioning, while for others animal welfare can be measured on the basis of animals’ preferences as these preferences are either infl uenced by the animal’s emotions, which have evolved to motivate behaviour to both avoid harm and facilitate survival, growth and reproduction, or refl ect important biological requirements of the animal. Furthermore, the term is often used to avoid being too specifi c about the nature of the particular issue (Broom and Johnson, 1993).

This uncertainty surrounding the concept of animal welfare and how it should be judged does not diminish the rigour of the research utilizing method-ologies or measurements arising from the various concepts of animal welfare. Because a key focus of this book is the effects of human–animal interactions on farm animal welfare, it is useful to review the main welfare methodologies and their rationale in order to understand their value and limitations in judging ani-mal welfare. This has been done elsewhere (e.g. Broom and Johnson, 1993; Duncan and Fraser, 1997; Fraser, 2003, 2008 ; Barnett and Hemsworth, 2009), but nevertheless, it is important to understand the science on which welfare decisions may be made.

2.5.1 Welfare assessment

There are basically three prominent concepts of animal welfare in the literature: the welfare of animals is judged on the basis of: (i) how well the animal is per-forming from a biological functioning perspective; (ii) affective states, such as suffering, pain and other feelings or emotions; and (iii) the expression of normal or ‘natural’ behaviours. In addition, there are elements of these three concepts in the so-called ‘fi ve freedoms’, which were proposed by the UK Farm Animal Welfare Council (1993) to protect the welfare of animals. These concepts of animal welfare and the resulting methodologies will be discussed here.

2.5.1.1 Biological functioningThe rationale underpinning this fi rst concept is that diffi cult or inadequate adap-tation will generate welfare problems for animals. Broom (1986) defi nes the wel-fare of an animal as ‘its state as regards its attempts to cope with its environment’. The ‘state as regards its attempts to cope’ refers to: (i) how much has to be done in order to cope with the environment and includes biological responses such as the functioning of body repair systems, immunological defences, physiological stress responses and a variety of behavioural responses; and (ii) the extent to which these coping attempts are succeeding, which includes the lack of bio-logical costs to the animal, such as deterioration in growth effi ciency, reproduc-tion, health and freedom from injury. Thus using this concept, the risks to animal welfare are assessed at two levels: the magnitude of the behavioural and physi-ological responses and the cost(s) to biological fi tness of utilizing these responses.

30 Chapter 2

The concept of biological fi tness generally applies to natural populations and refers to ‘fi tter’ animals having a greater genetic contribution to subsequent gen-erations (Pianka, 1974) as a consequence of their ability to successfully survive, grow and reproduce. Although the last attribute may not always apply to indi-vidual animals on the farm because reproduction is either controlled or absent for many farm animals, the ability to grow, survive and reproduce could be con-sidered measurements of fi tness within the limits of the management system (Barnett et al., 2001).

Broom’s (1986) defi nition is similar to the one recently endorsed by the 172 member countries of the World Organization for Animal Health (OIE), which has an animal welfare mandate that covers livestock (OIE, 2008):

Animal welfare means how an animal is coping with the conditions in which it lives. An animal is in a good state of welfare if (as indicated by scientifi c evidence) it is healthy, comfortable, well nourished, safe, able to express innate behavior, and if it is not suffering from unpleasant states such as pain, fear, and distress. Good animal welfare requires disease prevention and veterinary treatment, appropriate shelter, management, nutrition, humane handling and humane slaughter/killing. Animal welfare refers to the state of the animal; the treatment that an animal receives is covered by other terms such as animal care, animal husbandry, and humane treatment.

Thus, a key precept in this concept is that animals use a range of behavioural and physiological responses to assist them in coping with environmental condi-tions, and while biological regulation in response to environmental change is constantly occurring, adaptation is not always possible. When homeostasis (i.e. constancy of the internal environment, which varies only within tolerable limits) fails, there is damage, disease or even death (Broom and Johnson, 1993). Therefore, diffi cult or inadequate adaptation generates animal welfare problems. While this concept has support within animal welfare science, the rationale for it requires further elaboration.

In examining this concept, we fi rst need to consider the two key components of Broom’s (1986) defi nition of welfare, that is, how much has to be done in order to cope with the environment and the success of the coping attempts. Ani-mals use a wide range of biological responses, behavioural and physiological, to both regulate their lives and deal with diffi culties. These behavioural and physi-ological responses are adaptive responses that may help an individual to cope with its environment. While failure to adapt may ultimately result in death, less severe challenges can result in less serious biological costs, such as impaired growth, reproduction and health, and so both sets of consequences demonstrate that diffi cult or inadequate adaptation will generate welfare problems for ani-mals. The development of this discussion has been substantially aided by the lucid accounts presented by Broom and Johnson (1993) and Moberg (2000).

The stress response commences as soon as the central nervous system perceives a potential challenge (stressor) to homeostasis, followed by the devel-opment of a biological response or defence that consists of some combination of the four general biological defence responses: behavioural responses, responses of the autonomic nervous system, responses of the neuroendocrine system andresponses of the immune system. For many stressors, the fi rst and, at times, the

Farm Animal Welfare 31

most biologically economical and effective response is a behavioural one. In concert with the behavioural responses, the physiological responses that can be used by the animal are elicited basically in three series of events, with the full elicitation of these dependent on the time of exposure to the stressor and the suc-cess of the biological responses in coping with the challenge. Two key physiologi-cal responses that involve both neural and hormonal systems are the activation of the sympathetic–adrenal–medullary (SAM) and the hypothalamic–pituitary–adrenal (HPA) axes. A diagrammatic representation of the SAM and HPA axes and their hormones is presented in Fig. 2.2. Together, the responses of the SAM and HPA axes result in what is commonly termed the stress response, which encompasses one of the body’s major coping mechanisms to environ-mental disturbance.

Perceivedthreat

Pituitary gland

CRHAVP

ACTH

Adrenal cortex

Adrenal medulla

Spinal cord

Adrenalinenoradrenaline

Hypothalamus

Brain

Corticosteroids

Fig. 2.2. A diagrammatic representation of the sympathetic–adrenal–medullary (SAM) and the hypothalamic–pituitary–adrenal (HPA) axes and their hormones (courtesy of L.E. Edwards). ACTH = adrenocorticotrophic hormone; AVP = arginine vasopressin; CRH = corticotrophin releasing hormone.

32 Chapter 2

The SAM axis response is the fi rst series of physiological events and is char-acterized by a rapid, specifi c response of the autonomic nervous system and consequent secretions of catecholamines (adrenaline and noradrenaline, also called epinephrine and norepinephrine). These physiological adjustments are the immediate or ‘emergency’ response proposed by Walter Cannon (Cannon, 1914) as the ‘fi ght or fl ight’ response. The autonomic nervous system controls functions such as the digestive and cardiovascular systems, respiration and ther-moregulation, and consists of the sympathetic and parasympathetic nervous sys-tems, which have opposing effects. For example, the activation of the sympathetic nervous system, in which noradrenaline is a synaptic transmitter, inhibits diges-tive functions and stimulates cardiac output, whereas activation of the parasympa-thetic nervous system does the opposite. A major and important function of adrenaline is to quickly provide energy in the form of glucose from liver and mus-cle glycogen, a process known as glycogenolysis, and free fatty acids from lipolysis of adipose tissue (Murray et al., 2003). Thus the SAM axis response with the secre-tions of catecholamines (adrenaline and noradrenaline) is the principal regulatory mechanism that allows the animal to immediately meet physical or emotional challenges by its effects on metabolic rate, cardiac function, blood pressure, periph-eral circulation, respiration, visual acuity and energy availability and use.

If the responses of the SAM axis to a stressor are insuffi cient, there is another series of events involving the HPA axis and corticosteroid hormones. This starts with the brain perceiving a stressor and the hypothalamus producing corticotro-phin releasing hormone (CRH) in response to the stimulus. This results in the release of adrenocorticotrophic hormone (ACTH) from the adenohypophysis (anterior pituitary). ACTH is transported in the blood to the adrenal cortex where corticosteroid synthesis and release occurs. In some species, arginine vasopres-sin (AVP), another hormone from the hypothalamus, has a role in stimulating ACTH secretion (Matteri et al., 2000).

Corticosteroids are not stored in the adrenal glands and have to be synthe-sized. There are two predominant corticosteroids: cortisol is the predominant corticosteroid of most mammals, including humans, and of bony fi sh, and corti-costerone is the major corticosteroid in rodents, birds, reptiles, amphibians and cartilaginous fi sh (Chester Jones and Henderson, 1976). The half-life of cortisol (i.e. the time that it remains in a biologically active state before it is degraded) is about 1.5–2 h but its mode of transport in the blood provides some protection against its degradation and hence provides a ‘stor e’ of cortisol (Barnett, 2003). It is partitioned in the blood in three ways. Between 60% and 80% of the total cortisol is bound to albumin, which has a low affi nity but a high capacity, mean-ing that while it is weakly bound, a considerable amount can be bound. From 10% to 20% is bound to transcortin or corticosteroid binding globulin, which has a high affi nity but low capacity, meaning that it binds a small amount but strongly(Westphal, 1971). This tightly bound form provides a reservoir in which there is only a slow turnover. Only about 10% of total cortisol is free in the blood and this is the biologically active component. The concept of biologically active cortisol (or corticosterone) is important as it has implications for determining the magnitude of a stress response and its welfare implications; this important consideration will be discussed at the end of this section.

Farm Animal Welfare 33

The second series of responses, called the acute stress response (Selye, 1946, 1976), is a corticosteroid-dependent mechanism and thus the HPA axis is central to its function. The adrenal cortex and, in particular, the cortical cells that secrete the corticosteroids are controlled by higher centres of the hormonal system: the pituitary gland which, in turn, is controlled by the hypothalamus at the base of the brain. The acute response may last from minutes to hours and has the major func-tion of providing glucose from food or muscle protein (gluconeogenesis) for the required increased metabolic performance. Therefore, during this stage a steady state is achieved in which the increased demand for energy is met by increased metabolic performance. This physiological state of stress disappears on removal of the stressor with generally no ill effects other than a depletion of energy reserves.

The activation of the SAM and HPA axes is obviously an effective mecha-nism to assist the animal in adapting to changes in its environment. The physi-ological outcomes include adjustments in metabolic rate, cardiac function, blood pressure, peripheral circulation, respiration, visual acuity, and energy availability and use that allow the animal to meet physical and/or emotional challenges. In the short term, corticosteroids also reduce some of the damaging effects of the immune response, such as repressing the infl ammatory response. There are also some behavioural adaptations as a consequence of the short-term activation of the SAM and HPA axes, such as increased arousal and alertness, and increased cognition, vigilance and focused attention (Mendl, 1999; Kaltas and Chrousos, 2007), that should assist the animal to search, scrutinize and remember threaten-ing or rewarding situations. Some of the behavioural and physical adaptations that occur during acute stress are listed in Table 2.1.

If the stressor continues, the response proceeds to the third series of events, which is the chronic stress response, and it is this series of events that can have serious consequences for the animal. The chronic stress response is also a corticosteroid-dependent mechanism, but while in the acute phase the effects are potentially benefi cial, this chronic activation of the HPA axis comes at a physiological cost to the animal, such as a decreased metabolic effi ciency,

Table 2.1. Behavioural and physical adaptation during acute stress (modifi ed from Kaltas and Chrousos, 2007).

Behavioural adaptation: adaptive redirection of behaviour

Physical adaptation: adaptive redirection of energy

Increased arousal and alertness Oxygen and nutrients to the central nervous system and stressed body sites

Increased cognition, vigilance and focused attention

Altered cardiovascular tone and increased blood pressure and heart rate

Heightened analgesia Increased respiratory rateEuphoria (highly pleasant mood) or

dysphoria (highly unpleasant mood)Increased glycogenolysis, gluconeogenesis

and lipolysisIncreased temperature Inhibition of growth and reproductionSuppression of appetite and feeding

behaviourContainment of the infl ammatory/immune

response

34 Chapter 2

impaired immunity and reduced reproductive performance. Therefore, the long-term activation of the HPA axis can have marked effects on the effi ciency of growth with, for example, the breakdown of muscle protein under the catabolic effects of ACTH and corticosteroids (Elsasser et al., 2000). Corticosteroids also support the synthesis and action of adrenaline in stimulating glycogenolysis (i.e. provision of glucose from liver and muscle glycogen for the required increased metabolic performance) and lipolysis (provision of energy in the form of free fatty acids from the breakdown of adipose tissue) (Matteri et al., 2000). Stress-induced changes in the secretion of pituitary hormones have also been implicated in failed reproduction (Clarke et al., 1992; Tilbrook et al., 2000) and immune com-petency (Blecha, 2000). How serious these costs are depends on how long the animal is required to divert physiological resources to maintain homeostasis.

As discussed earlier, the concept of biologically active cortisol is important because of its implications for determining the magnitude of a stress response and its consequences. The earlier discussion, together with detailed reviews by Moberg (2000) and Barnett (2003), describe the typical physiological responses in animals under chronic stress. A predominant feature of a chronic activation of the HPA axis is increased basal secretion of corticosteroids with a loss in diurnal regulation of the axis (Harbuz and Lightman, 1992). It is this sustained elevation in free corticosteroids, together with changes in other hormones, cardiovascular function, metabolism and the immune system, that has broad, long-lasting effects on the body, such as decreased metabolic effi ciency, impaired immunity and reduced reproductive performance. In other words, increased basal secretion of corticosteroids has signifi cant fi tness consequences for the animal, and it is these biological and fi tness effects that refl ect both the magnitude of the stress response and the welfare implications.

While the role and actions of corticosteroids in acute and chronic stress responses are well known, this is not to imply that the HPA axis is the only neu-roendocrine axis affected by stressors. There is also involvement of the somato-trophic and thyroid axes and of other hypothalamic and pituitary hormones such as arginine vasopressin and prolactin, respectively. There is also a direct involvement of the immune system. While corticosteroids can suppress the immune system (Blecha, 2000; Kaltas and Chrousos, 2007), studies using vari-ous stress models show that factors other than corticosteroids may also be involved in the stress-induced immunosuppression observed in animals that are transported, restrained or isolated (Blecha, 2000).

It is clear that the hormones secreted from the HPA axis have broad, long-lasting effects on the body, and thus challenges to homeostasis that result in such neuroendocrine responses clearly have implications for animal welfare. While some component of behaviour is likely to be involved in every stress response, behavioural responses may not be appropriate or effective for all situations. Indeed, redirected behaviours and stereotypies, as with long-term neuroendo-crine responses, may indicate diffi cult or inadequate adaptation. Therefore, measuring both these behavioural and physiological responses to a stressor (Fig. 2.3), which refl ect the challenge confronting the animal, as well as the fi tness consequences of these responses for the animal, clearly affords an insight into the risks to the animal’s welfare.

Farm Animal Welfare 35

This concept of animal welfare has been criticized on several grounds. One common criticism is that fundamental requirements do not adequately include emotions or feelings. However, this would only be valid if emotions are indepen-dent of other biological processes; this is unlikely, and others consider that the mental state of an animal is an integral component of its biological state (Dantzer and Mormède, 1983). Emotional responses are produced in the limbic system, which projects to several parts of the brain, including those involved in the initia-tion and maintenance of the stress response, thus explaining why emotional insults activate a stress response (Kaltas and Chrousos, 2007). Emotions are part of the body’s regulatory system, and together with a range of learning processes, function to assist animals in avoiding potentially harmful situations or recogniz-ing potentially benefi cial situations (Cabanac, 1979). Emotions have several components, including cognitive processes, the associated sensation of emotion and the visceral and behavioural responses (Dantzer, 1988), and thus emotions are expected to infl uence the animal’s behavioural and physiological responses to challenges. Indeed, emotions such as fear can have marked effects on these biological responses and animal fi tness. The effects of handling of farm animals, as discussed in Chapter 3, clearly indicate the profound effects of fear on avoid-ance behaviour, stress physiology and fi tness. A consistent fi nding in biological psychiatry is that the HPA axis physiology is altered in humans with major depression (see Parker et al., 2003), for example, a sustained elevation in basal cortisol concentrations. Furthermore, emotions, or at least their physiological component, may modulate memory formation in several ways (Reisberg and

Fig. 2.3. Implanted catheters enable blood samples to be collected quickly and simply without infl uencing corticosteroid concentrations of the samples. The pigs in this photograph have indwelling catheters and the collars around the animals’ necks protect and store the external section of the catheters and their attached taps.

36 Chapter 2

Heuer, 1995). Studies principally on laboratory rodents have shown that acute stressors, presumably via their effects on hormones of the HPA and SAM axes, may enhance memory formation and recall (Mendl, 1999). As discussed earlier in this section, some of these effects can be viewed as having adaptive value in helping the animal to search, scrutinize and remember threatening stimuli or situ-ations. Thus, it is axiomatic that the biological responses that animals use when attempting to cope with diffi culties include emotional responses.

Early attempts in the 1960s and 1970s to study animal welfare focused on measuring animal stress. These attempts have been used to criticize the biologi-cal functioning-based concept of animal welfare (Duncan, 2004) without recog-nizing that this concept considers not only the magnitude of the stress responses but more importantly the consequences of these stress responses (Barnett and Hemsworth, 2003). While animals utilize the stress response to try to deal with diffi culties or challenges to homeostasis, it must be emphasized that in terms of welfare it is the consequences for the animal that are important. Failure to adapt to the stressor can adversely affect an animal’s fi tness because of adverse effects on nitrogen balance, reproduction, injury and health. The effects on health include ulcers, hypertension, arteriosclerosis and a suppression of the immune system, and some effects may be permanent even if the stressor is subsequently removed (Moberg, 2000). Furthermore, in addition to a broad range of physio-logical responses, including those of the nervous, immune, circulatory and endo-crine systems, and particularly those of the HPA axis, animals also utilize behavioural responses to try to deal with these challenges. Long-term behav-ioural responses such as stereotypies, demonstrated by bar biting in sows (Fig. 2.4), and pacing, weaving and wind sucking in stabled horses, are ‘harmful’ behavioural responses that have clear biological costs for animals such as physi-cal injury, digestive complications and the loss of production. Thus, the focus of

Fig. 2.4. A pregnant sow exhibiting the stereotypy of bar biting.

Farm Animal Welfare 37

the welfare concept of biological functioning extends to the consequences of the stress response (and other adaptive responses, such as behaviour) rather than the response per se.

Acute stress responses can occur to pleasant as well as unpleasant stimuli and this is often used to criticize the use of corticosteroids to measure stress. For example, mating in rats (Szechtman et al., 1974) and voluntary exercise in humans (Sutton and Casey, 1975) both result in increased plasma corticosteroid concentrations. In these cases, activation of the SAM and HPA axes is an appro-priate response in mobilizing the body’s reserves to support mating and exercise, and thus care needs to be taken not to interpret all acute stress responses as sig-nifying reduced welfare. The duration and intensity of the acute stress response can be used to address specifi c management procedures that minimize acute stress responses. For example, based on plasma cortisol concentrations in sheep, shearing is a more severe stressor than yarding, fast shearing is more stressful than slow shearing and herding by people is more stressful than using dogs (Kilgour and de Langen, 1970). Determination of the magnitude of acute stress responses is useful therefore in identifying improved or alternative management techniques, but acute stress responses are, by defi nition, short term and do not generally have long-term detrimental consequences. However, as indicated ear-lier in this section and emphasized by Moberg (2000), it is the biological cost of stress that is the key to understanding the welfare implications for the animal.

In conclusion, diffi cult or inadequate adaptation will affect the fi tness of the animal through a range of long-lasting behavioural and neuroendocrine responses; thus, the rationale underpinning this concept is that diffi cult or inad-equate adaptation generates welfare problems for animals. Using this concept of biological functioning, a broad examination of the behavioural, physiological, health and fi tness responses of animals to the condition of interest (i.e. under study) can be undertaken to assess the biological functioning of the animals. In other words, the risks to the welfare of an animal imposed by the condition of interest can be assessed at two levels: (i) the magnitude of the behavioural and physiological responses; and (ii) the biological cost of these responses. These behavioural and physiological responses include abnormal behaviours, such as stereotypies and redirected behaviours, and the stress response, respectively, while the biological cost includes adverse effects on the animal’s ability to grow, reproduce and remain healthy and free of injury.

This approach to welfare assessment has been used by scientists to examine the effects of housing, husbandry and handling. For example, studies examining surgical husbandry procedures have used a broad examination of the behav-ioural, physiological, health and fi tness responses to study animal welfare (Mellor and Stafford, 2000). A broad range of biological responses are also used in Chapters 3, 5 and 6, in which the effects of handling on animal welfare are reviewed and considered.

2.5.1.2 Affective states In general, this second concept defi nes animal welfare in terms of emotions and thus it emphasizes reductions in negative emotions, such as pain, fear and frustration, and increases in positive emotions, such as comfort and pleasure

38 Chapter 2

(Duncan and Fraser, 1997). The classic view is that emotions, both positive and negative, have as their antecedents some discrepancy or confl ict between the state of the world and the expectations of the individual. Furthermore, the view of emotions in both the animal behaviour and psychology literature highlights the linkage between visceral arousal and cognitive processes (Bolles, 1981; Mandler, 1981; Dantzer, 1988). Any discrepancy or any interruption of expecta-tions or of intended actions leads to the associated sensation of the emotion or feelings, which is either positive or negative, depending on the cognitive evalua-tion of this discrepancy or confl ict between the state of the world and the expec-tations of the individual. This associated sensation of emotion, in turn, is linked with undifferentiated visceral (autonomic) arousal and behavioural responses.

It is accepted that humans have a great variety of emotions; these can be positive (e.g. pleasure, contentment and relief) or negative (e.g. fear, pain, anger, frustration, boredom and discomfort) and are accompanied by a variety of behavioural and physiological signs (e.g. facial expressions and increased heart rate and adrenaline concentrations). Animal behaviourists generally consider that non-human animals are restricted to a few basic emotions such as anger, fear, joy and happiness (Bolles, 1981). This is predicated on the view that ani-mals probably only have emotions to deal with certain kinds of survival prob-lems for which there is some strong evolutionary benefi t. For example, while we might expect animals to show fear because of the adaptive value of being fright-ened in a dangerous situation, there is no reason to expect animals, for example, to show pity to other species because there would be no clear adaptive advan-tage if they did (Bolles, 1981). However, the question is still raised by some as to whether other non-human animals experience emotions or simply behave in an emotional fashion (Dawkins, 2006).

Duncan (2004) has described the argument that animal welfare ultimately concerns animal feelings or emotions as follows. All living organisms have cer-tain needs that have to be satisfi ed for the organism to survive, grow and repro-duce, and if these needs are not met, the organism will show symptoms of atrophy, ill health and stress, and may even die. Higher organisms (vertebrates and higher invertebrates) have evolved ‘feelings’ or subjective affective states that provide more fl exible means for motivating behaviour to meet these needs. Dawkins (1977) also suggests that it is reasonable to assume that subjective feel-ings evolved because animals that possessed them were fi tter than those that did not. Fraser and Duncan (1998) have suggested that negative feelings are infl u-ential when there is an immediate threat, while positive feelings motivate behav-iour when there is a long-term benefi t in performing the behaviour but no immediate need.

Although there do not appear to be relevant data, it is possible that emotions have reinforcing properties. Negative reinforcement strengthens a behaviour as a consequence of an aversive condition being stopped or avoided by the behav-iour, while positive reinforcement strengthens a behaviour as a result of the addi-tion of a rewarding stimulus immediately following the behaviour. Furthermore, punishment weakens a behaviour because a negative condition is experienced immediately following the behaviour. Increases in pleasurable emotions would be expected to be rewarding and the cessation of negative emotions would be

Farm Animal Welfare 39

expected to be negatively reinforcing. In an attempt to establish a paradigm to investigate positive emotional states in laying hens, Moe et al. (2009) produced anticipatory behaviours in hens by using food rewards; they have yet to demon-strate neuropharmacologically that this anticipation is a positive emotional state, but this paradigm offers a promising approach to the study of the behavioural concomitants of positive emotions.

Animal emotions have been considered inaccessible to scientifi c investiga-tion because they have been described as human subjective experiences or even as illusory concepts outside the realm of scientifi c inquiry (Panksepp, 1998). The diffi culties in studying emotions as though they were objective states of bodily arousal are well recognized in the literature (Cacioppo et al., 1993). While each emotion may refl ect a different pattern of arousal, the visceral response to many emotions is reasonably uniform in animals. Most animals react physiologically, at least in the short term, in essentially the same way whether the arousal is sexual, fear provoking or there is the anticipation of play or food. It is obviously a major challenge to study and understand emotions in animals, although there have been some promising recent developments in the comparative study of emotions which show that there are many homologous neural systems involved in similar emotional functions in both humans and other mammals, and perhaps in other vertebrates (LeDoux, 1996; Panksepp, 1998, 2005).

Over the past 30 years there has been growing interest in the measurement of the preferences of farm, laboratory and zoo animals in order to answer ques-tions about animal welfare. The rationale for using animal preference tests to study animal welfare is basically twofold. Some scientists argue that animal pref-erences can be utilized on the basis that these preferences are infl uenced by the animal’s emotions (or feelings), which are prime determinants of its welfare (Duncan and Petherick, 1991; Duncan, 2004). Thus, proponents of the concept of affective states advocate that because animal preferences are infl uenced by animal emotions and that preference tests can indirectly measure emotions, understanding animal preferences allows inferences to be drawn on animal wel-fare. While not contrary to this approach, others propose that preferences are likely to refl ect important biological requirements of the animal and thus opti-mize fi tness (Broom and Johnson, 1993). Behavioural decisions by animals will fl uctuate over time, owing to a complex interplay of internal and external factors, but the aim of these decisions is to basically maintain homeostasis and optimize fi tness. Thus by investigating preferences there is the opportunity to identify important biological requirements of the animal. It is clear that these two views on the value of preference testing are similar in that Duncan (2004) has argued that higher organisms have evolved feelings to motivate behaviour in order for the organism to survive, grow and reproduce.

The fi rst reports of animal preference testing appeared in the literature in the 1970s (e.g. Hughes and Black, 1973; Hughes, 1976; Dawkins, 1976) and many preferences studies, using different approaches, have been conducted since, par-ticularly on farm animals. The simplest preference study involves allowing the animal to make a choice between two situations in which the resource is varied. For example, Hughes (1975) found that laying hens preferred a spacious cage to a confi ned cage and that neither time of day nor strain of bird was infl uential in

40 Chapter 2

this choice. Preference tests have been used to measure both the choice for and the avoidance of options or environments. A Y-maze apparatus that allows a choice between access to two options has often been used to provide informa-tion about specifi c features in the environment, such as fl ooring on raceways (Hutson, 1981), restraint methods (e.g. Pollard et al., 1994), handling treatments (Rushen, 1986) and ramp design (Phillips et al., 1988). Furthermore, aversion learning techniques have been used to study the animal’s motivation to avoid husbandry and handling treatments. For example, Rushen (1986) studied the avoidance of sheep to electro-immobilization, a procedure in which a pulsed, low-voltage current can be used to immobilize the animal. Sheep were trained to associ-ate a location with a specifi c treatment and avoidance was assessed based on the effort required to move them repeatedly to the treatment location. It was found over repeated trials that sheep showed increasing avoidance of a location in which they were restrained with electro-immobilization rather than of a location in which they were restrained without it.

While the consistent choice or preference of one resource over another or others indicates the animal’s relative preference, some have argued that in addi-tion to establishing what an animal prefers, it is important to understand the strength of the preference (e.g. Dawkins, 1983; Matthews and Ladewig, 1994). To address the question of the strength of an animal’s preference, experiments have incorporated varying levels of cost (e.g. work effort, time and relinquishing a desirable resource) associated with gaining access to a resource or avoiding aversive stimulation.

For example, Dawkins (1983) varied the price paid for access to litter by increasing the duration of feed withdrawal before the test. She found that although hens preferred litter to wire fl oors, their preference was not strong enough to outweigh the attraction of food, and she concluded that in both experiments there was no evidence that hens regarded litter as a necessity.

Dawkins (1983) also suggested that quantitative measures of the importance of resources for animals can be derived from measures of demand elasticity. Consequently, ‘behavioural demand’ studies, using operant conditioning tech-niques in which the animal must learn to perform a response (such as pecking at a key or pushing through a weighted door, to gain access to a resource), have been used to study the animal’s level of motivation to access or to avoid the situ-ation being tested. For example, Matthews and Ladewig (1994) studied the behavioural demand functions of pigs for the resources of food, social contact and a stimulus change (door opening). The amount of work, in the form of push-ing a plate, required for access to each reinforcer (resource) was systematically varied. It was found that while the demand for opening the pen door was highly elastic (i.e. the willingness of the pigs to access the resource declined as the effort increased), the demand for food was inelastic and the demand for social contact was intermediate.

The initial preference studies stimulated considerable scientifi c debate relat-ing to conceptual and methodological diffi culties (Dawkins, 1977; Duncan, 1978), and Fraser and Matthews (1997) concluded that the usefulness of prefer-ence tests to answer questions about animal welfare is limited by three main issues: (i) preference tests should adequately refl ect the animal’s preference;

Farm Animal Welfare 41

(ii) they need to establish how strongly an animal prefers a chosen option, avoids a non-preferred one or is motivated to perform a behaviour; and (iii) preferences may not correspond to welfare if the choices fall outside the animal’s sensory, cognitive and affective capacity, or if the animal is required to chose between short- and long-term benefi ts. It is useful briefl y to consider these and other concerns here.

Preference tests measure an animal’s choice behaviour at a point in time and such measurements run the risk of failing to account for interactions of dif-ferent motivational states that may infl uence the behaviour of the animal over time (Hutson, 1984). Furthermore, this short-term choice may refl ect the ani-mal’s proximate (immediate or present) requirements, rather than the animal’s ultimate requirements or those necessary for survival, growth and reproduction (Lawrence and Illius, 1997). For example, a hen continues to return to a ‘trap-nest’ to lay eggs even though it is assumed that the trap-nest is aversive because the hen cannot escape, there is no food or water and it is handled to be released. The hen’s behaviour on entering the nest simply refl ects its motivation at the time and not the future consequences (Duncan, 1978). Preference tests therefore need to be comprehensive enough to identify these sources of variation (Fraser and Matthews, 1997).

Preferences may vary with familiarity. The studies by Phillips et al. (1991, 1996) showed that previous experience strongly infl uenced the sows’ initial choice of farrowing crates in which design (solid or open sided crates) and fl ooring (con-crete, plastic coated metal mesh, or bare mesh fl oors) varied. Similarly, laying hens with experience of cages initially preferred this environment to a free-range run (Dawkins, 1977). In both cases, the preferences waned as the animals gained experience with the alternatives. Prior experience can be controlled in preference tests by using naïve animals, familiarizing the animals with the resources prior to testing or, as Fraser and Matthews (1997) suggest, prolonged testing.

Preferences for specifi c resources may also be affected by the context in which the animals are tested. Pedersen et al. (2002) and Sherwin (2003), using operant conditioning techniques to measure the motivation of pigs for food and straw, and of mice for a running wheel, respectively, found that the social context at the time of testing affected motivation to access these resources. There is also evidence that the stimuli from the test resources at the time of testing may affect motivation for the resources under study. For example, the motivation of mink to gain access to resources is affected by whether or not they can see the resources (Warburton and Mason, 2003). The context in which the animals are studied therefore should relate to the commercial conditions to which the experimental question is directed. Indeed, Dawkins (2003) has proposed the development of in situ methods of assessment to allow for the effects of factors such as the ani-mal’s development history and the precise commercial conditions experienced by the animal.

Concerns have been raised about the operant responses used to quantify the level of demand in behavioural demand studies. Fraser and Matthews (1997) suggest that certain operant responses required to be performed to obtain the reward may not be appropriate for certain types of reward. Using the example of hens required to peck a key in order to activate a motorized barrier and enlarge

42 Chapter 2

their cages, they questioned whether an alternative response, such as walking, would have provided a better measure of the hens’ motivation for additional space. Fraser and Matthews (1997) recommend that experiments should be repeated using different methods when there are questions about the appropri-ateness of the required operant response.

Fraser and Matthews (1997) recognize that limitations in using preference tests arise when animals are exposed to potential dangers or benefi ts that are beyond their sensory or affective capacity, or if the choice requires a level or type of cognitive ability that the animal does not possess. They suggest that the best safeguard is to base preference tests on the types of choices that the species argu-ably evolved the capacity to make and that the individual animals are accus-tomed to making.

The conceptual basis of the link between animal preferences and welfare has been criticized because animals may not necessarily prefer or be motivated to obtain resources that are truly in their best interests (Duncan, 1978). For an ani-mal of a wild genotype living in the wild (i.e. successfully evolved in the environ-ment), natural selection and behavioural development are expected to produce a set of environmental preferences that promote the health and survival of the individual and offspring. But without such natural selection, should this be expected in domestic animals? There are examples in livestock production where farm animals can learn to avoid foods that are associated with gastrointestinal distress (Provenza, 1995). There are also examples where farm animals, if given appropriate ‘choices’, are capable of selecting a mixed diet that can ameliorate health issues, such as lameness in meat (broiler) chickens (Danbury et al., 2000), that provides physical characteristics to optimize organ development such as giz-zard growth in chickens (Cummings, 1994; Wu and Ravindran, 2004), or that allow the optimization of energy and protein requirements for growth and health (Rutter et al., 2004). However, there are also examples to the contrary where choice may not promote the animal’s health and survival. Using operant condi-tioning techniques to study the demand of pigs for varying concentrations of sweet solutions, Kennedy and Baldwin (1972) found that a number of pigs had repeated bouts of illness after drinking large amounts of sugar solution. Confi ned sows show minimal demand for exercise (Ladewig and Matthews, 1996), yet a degree of exercise has been shown to have a longer term benefi cial effect on leg strength (Marchant and Broom, 1994).

In conclusion, as with biological functioning, clarifying the conceptual link between animal preferences and animal welfare is an issue for some. The indi-vidual’s concept of animal welfare clearly underscores the methodology used to judge or measure animal welfare. However, as commented by a number of authors (e.g. Fraser and Matthews, 1997), preference research should be integrated with other measures used in animal welfare research. Furthermore, Widowski and Hemsworth (2008) recommend that, while studies of motivation can provide compelling evidence that the performance of some behaviour (or preference) may be important to the animal, additional evidence, particularly on the occurrence of abnormal behaviour, and on stress physiology and health, are necessary to provide a more comprehensive assessment of the impact of restriction on animal welfare.

Farm Animal Welfare 43

2.5.1.3 The expression of normal or ‘natural’ behavioursThe principle underlying this concept of animal welfare is that animals should be raised in ‘natural’ environments and allowed to behave in ‘natural’ ways. How-ever, of the three concepts described in this chapter to assess welfare, the ‘nature of species’ approach has the least scientifi c credibility because it fails to defi ne both ‘natural’ and the welfare risks if either such ‘natural’ conditions are not provided or there are no opportunities to behave in ‘natural’ ways.

The view that animals should perform their full ‘repertoire’ of behaviour was common in early welfare research and is still common today, as often seen in the promotion of so-called ‘welfare friendly’ production systems. Thorpe (1965) argued that animals need to perform all the behavioural patterns that are dis-played by free-living members of their species and that they suffer if they cannot display all these behaviours. There are some obvious shortcomings with this approach as a criterion for animal welfare. There is broad agreement within sci-ence that it is often diffi cult to attribute actual suffering when the expression of certain behaviours is prevented or is absent when it would be expected to be present (Dawkins, 2003). Furthermore, as discussed by Dawkins (1980), ‘wild’ behaviour often represents an animal’s efforts to survive in a life and death strug-gle or contest, and therefore some ‘natural’ behavioural responses are adapta-tions to cope with extreme adverse situations. The following commentary by Martel (2002, p. 16) provides a less optimistic view than the populist romantic view of animals living in the wild: ‘Animals in the wild lead lives of compulsion and necessity within an unforgiving social hierarchy in an environment where the supply of fear is high and the supply of food low and where territory must constantly be defended and parasites forever endured.’ Thus the ‘natural behav-iours’ that are desirable or undesirable in terms of animal welfare require defi ni-tion, together with the rationale for their inclusion or exclusion. To date, there has been no attempt to reach agreed defi nitions or provide rationales.

The more general idea that we can improve animal welfare by respecting the ‘nature’ of animals is, as suggested by Dawkins (1980), intuitively appealing. However, the concept of an animal’s ‘nature’ would need to be more specifi c before it could give guidance in assessing animal welfare. For example, as aptly noted by Dawkins (1980), the sea is no longer an essential part of seagulls’ ‘nature’ even though they have evolved to live in such close association with the sea: in the past 30 years, gulls have changed their habits and now live voluntarily in very artifi cial environments created by humans; they nest on buildings, roost on playing fi elds and forage in garbage dumps.

This issue of what is the natural environment for domestic animals is put into sharp focus when we consider the history of the domestic hen as described by Appleby et al. (1992). The progenitor of the domestic fowl was the red jungle fowl (Gallus gallus), subspecies of which are still found in Asia. This is a tropical species confi ned to forested areas and to thick vegetation. It is believed to have been domesticated over 8000 years ago and the modern hen is regarded as a subspecies, viz. Gallus gallus domesticus. During early domestication, the hen was valued as a sacrifi cial or religious bird or for cockfi ghting. The Romans developed specialized breeds, including layers that were laying an egg every day. This industry collapsed with the decline of the Roman Empire about 2000 years

44 Chapter 2

ago and was not resumed until the 20th century. After this initial domestication, there was little selection for many centuries, except for cockfi ghting. Indeed, in the early 1900s there were only six breeds in England, but this increased to about 70 breeds by 1980. As a consequence of this period of rapid selection, we now have two modern hybrids, the egg-laying bird that reaches point of lay at 16–18 weeks of age at a body weight around 1.8–2.0 kg and lays close to an egg a day, and the meat bird that reaches a slaughter weight of about 2.0 kg as quickly as at 5 weeks of age. This quick history of the domestic hen raises questions on provid-ing ‘a natural environment’. What is the natural environment of a young bird selected for meat production or an adult hen selected for egg laying, both of which are the same species, but have followed about 8000 years of selection for fi ghting capabilities and a 100 years of intense selection for production attributes? Is an outdoor area with relatively little structural diversity, except perhaps for some grass, a natural environment for a tropical species? Furthermore, while the wild bird was theoretically an appropriate species to domesticate, based on both its adaptability and its social behaviour (e.g. it lived in small groups), the ongoing selection for fi ghting ability has resulted in a bird that is abundant but is diffi cult to manage in modern production systems. For example, the laying hen routinely requires beak-trimming to control feather pecking and cannibalism, particularly in large group production systems (Appleby et al., 2004; Lambton et al., 2010).

Domestication has resulted in behavioural changes between wild and domestic stocks. Adaptation to the captive environment is achieved through both genetic changes occurring over generations, and environmental stimulation and experiences during an animal’s lifetime (Price, 2002). While there is little evidence that domestication has resulted in the loss of behaviours from the species’ repertoire or that the basic structure of the motor patterns for such behaviours has been changed, behavioural differences between wild and domes-tic stocks in nearly all cases are quantitative in character and best explained by differences in response thresholds of behaviour to, for example, sexual or novel stimuli, humans and environmental conditions (Price, 2002). These compari-sons are diffi cult because of problems in both determining an appropriate wild population and interpreting differences between wild and domestic populations under one environment, in nature or in captivity. However, studies of farmed and wild Atlantic salmon (Salmo salar), for example, both reared in either cap-tive or wild environments, indicate that farmed salmon show fewer predator responses, while also having increased growth, increased disease resistance and decreased stress responses (Price, 2002).

Thus, the concept of an animal’s ‘nature’ would need to be more specifi c before it could give guidance in assessing animal welfare, as generalizations may lead us astray and achieve the opposite of what is intended. Similarly, the ‘natu-ral behaviours’ that are desirable or undesirable in terms of animal welfare require defi nition, together with the rationale for their inclusion or exclusion. More recently the emphasis has been on behavioural indicators of poor coping such as fearfulness, aggression and stereotypies (EFSA, 2005), responses that are also utilized in the biological functioning-based concept of animal welfare.

The so-called ‘fi ve freedoms’, that is freedom from hunger and thirst, from discomfort, from pain, injury and disease, and from fear and distress, and

Farm Animal Welfare 45

freedom to express normal behaviour (FAWC, 1993), include aspects of all three of the animal welfare concepts described above. While most would accept that these freedoms are necessary to avoid a lack of suffering, in terms of a consensus on animal welfare assessment, there has been little attempt to defi ne the levels of freedom that are desirable, together with the adverse consequences of not providing such freedoms.

2.5.2 Conclusion on welfare assessment

The different animal welfare concepts or views on animal welfare can lead scien-tists to use different criteria or methodologies in assessing an animal’s welfare. For short-term animal welfare issues involving acute stress, such as painful hus-bandry procedures, there is considerable agreement on the need to assess ani-mal welfare from a perspective of biological functioning (Mellor et al., 2000). However, for longer term issues, such as confi nement housing, disagreement over these welfare concepts or criteria, especially when consequent interpreta-tions confl ict, leads to debates concerning animal welfare and the varying interpretations (Fraser, 2008; Barnett and Hemsworth, 2009).

While there is scientifi c uncertainty in relation to animal welfare concepts or views, this does not necessarily diminish the rigour of the research utilizing crite-ria or methodologies arising from these different concepts. Furthermore, there are several commonalities in the rationale for two of the concepts of animal wel-fare: biological functioning and affective states (Barnett and Hemsworth, 2009). For example, it is considered that animals, at least in the wild, will be motivated to choose those resources (or behaviours) that maintain homeostasis to optimize their fi tness, that is, optimize their growth, reproduction, injury status, general health and survival. It is also considered by some that feelings or subjective affec-tive states have evolved to motivate behaviour in order to meet needs that have to be satisfi ed in order for the organism to survive, grow and reproduce. While there is limited evidence that deprivation of highly preferred resources results in biological dysfunction (e.g. Stevens et al., 2009), research utilizing well-accepted stress models is required to understand the relationships between these concepts and methodologies.

In the meantime, until science can broadly agree on the best methodology or methodologies to evaluate animal welfare, these two approaches should guide current welfare research methodology. The basis of the methodology used by scientists to assess animal welfare should routinely be provided so that individu-als using science in their decision making appreciate both the rationale for the methodology and its limitations (Fraser, 2003; Sandøe et al., 2004).

2.6 Conclusions

We have not attempted to address the issue of animal rights in this chapter. There has been considerable argument at the community level and among philosophers and scientists about the extent to which humans are entitled to use animals for a

46 Chapter 2

variety of purposes, including use as companion animals, as objects of hunting or fi shing and in agriculture. It is beyond the scope of this book to adequately address this issue. Nevertheless, current community values do accept the moral right for humans to use animals in agriculture as a source of food but place con-siderable emphasis on the welfare issues discussed here. While the general com-munity and, at times, the livestock industries have implicated housing systems and husbandry practices as potential causes of reduced welfare in farm animals, the infl uence of the stockperson on the welfare of animals is not necessarily obvious to many people both within and outside agriculture.

In this chapter we have attempted to provide the background information that is important in exploring the role of human–animal interactions in livestock production, particularly with respect to the implications for animal welfare. It will become clear in subsequent chapters, for example, that the welfare of livestock is at serious risk in situations where the stockperson’s commitment to surveillance of and attendance to welfare and production issues is less than optimal. Simi-larly, animal performance is also at risk in such situations. This book will there-fore highlight the integral role that the stockperson plays in determining the welfare of farm animals.

In the next chapter, we will consider human–animal interactions from the point of view of current research. The aim will be to identify some of the funda-mental principles underlying the way in which animals react to humans.

© CAB International 2011. Human–Livestock Interactions, Second Edition (P.H. Hemsworth and G.J. Coleman) 47

3 Human–Animal Interactions and Animal Productivity and Welfare

3.1 Introduction

In most walks of life humans frequently interact with animals and in many situations these interactions are such that relationships develop between humans and animals. The human–companion animal relationships that are so common in Western society households are an excellent example of the intense and close relationships that may develop between humans and animals. Some of the ben-efi ts that the relationship offers to both partners are recognized. It is obvious that pets are critically dependent on humans for their care and maintenance; how-ever, there is growing appreciation of the benefi ts that companion animals may have on the physical and psychological health of their owners. In an extensive review of research on the relationship between dogs and human health, Wells (2007) concluded that although not all studies in this area have been method-ologically robust, the studies suggest that dogs can assist in the prevention of ill-ness in their owners. For instance dog owners, as a group, tend to be a healthier cohort of individuals than non-owners in terms of both minor physical ailments and more serious medical conditions, such as coronary heart disease. Further-more, Wells (2007) concluded that dogs may facilitate their owners’ recovery from certain types of ailment, including coronary heart disease. In relation to psychological health, Wells (2007) concluded that dogs may assist the psycho-logical well-being of patients in hospitals and residential nursing homes on the basis of assessments of patients being ‘happy, alert and socially responsive’. In relation to children, there is evidence that pet ownership may improve social competence (Edney, 1992). In contrast, there is less appreciation, both within the general community and, to some extent, within the livestock industries, of the implications of human–animal relationships in livestock production.

In modern livestock production, particularly in intensive production systems, human–animal relationships similar to those in domestic settings develop because farm animals receive frequent and, at times, close human contact. Even

48 Chapter 3

though there may be considerable automation with these production systems, stockpeople are required to regularly monitor animals and their conditions and impose routine husbandry procedures. As a consequence, the amount of human contact that these animals receive is considerable. For example, a stockperson in modern meat chicken units may manage tens of thousands of birds at a time and, although the stockperson may not physically interact with his or her ani-mals, the stockperson will often be in close visual contact with most of the birds as many as six times daily during routine inspection of birds and their conditions. Lactating dairy cows are intensely handled twice daily during lactation and a stockperson may handle hundreds of cows twice daily at the time of milking (Fig. 3.1). Similarly, breeding female pigs may be handled frequently around mating and parturition. Handling of pigs and dairy cattle generally involves tactile inter-action by the stockperson, which may either be negative or positive. Even though in modern production systems the number of animals managed by each stock-person has substantially increased, farm animals in these systems are likely to receive, at times, frequent and close human contact. Furthermore, many of these interactions by stockpeople may be biased towards negative ones, given that opportunities for positive human contact are probably minimal, and many rou-tine husbandry tasks undertaken by stockpeople may contain negative elements, such as restraint, vaccinations and surgical interventions.

Many people in both the farming and general community recognize that an animal’s early experiences with humans may have marked effects on how it

Fig. 3.1. In modern livestock systems, even with considerable automation, such as in this rotary dairy, stockpeople work with and interact frequently with large numbers of farm animals.

Human–Animal Interactions 49

subsequently behaves towards humans. Popular reports on the effects of early experience with humans, such as those in the studies by Konrad Lorenz (Lorenz, 1970) on hand-reared birds, have probably contributed to this general apprecia-tion of the effects of early human contact. However, what is surprising for many is the profound effects not only of this early experience with humans, but also of the effects of regular human contact on the long-term behaviour and physiology of the animals. There is good evidence based on handling and fi eld studies examining fear–productivity relationships that human–animal interactions may markedly affect the productivity of farm animals. As fear and stress appear to be implicated in the effects of human–animal interactions on productivity, these interactions between humans and farm animals may also have implications for the welfare of these animals. This chapter examines these effects of human–animalinteractions on farm animal fear of humans, productivity and welfare. The chap-ter also forms an introduction to Chapters 4 and 5, in which the development of these human–animal interactions in livestock production are explored.

3.2 Human–Animal Interactions and the Human–Animal Relationship

From an ethological perspective, the relationship between a human and an animal can be viewed in terms of inter-individual relationships, with the quality and fre-quency of interactions between the two individuals, as well as the context in which they occur, determining the quality of their relationship. This conceptual frame-work, which is used to consider human–animal relationships in the book, is based on that described by Estep and Hetts (1992), which, in turn, was derived from the views of Hinde (1970) on social relationships. Social structures can be considered to be an accumulation of relationships, and relationships develop from inter actions between individuals. Utilizing the view of Hinde (1976) that inter-individual rela-tionships are based on the history of regular interactions between two individuals that are familiar with each other, Estep and Hetts (1992) proposed that human–animal relationships can be viewed in a similar manner and that studies of this relationship should be undertaken by investigating each partner’s perception of the relationship. Each individual partner’s perception of the relationship allows the interpretation and prediction of future interactions. Therefore, if animals are able to learn and anticipate future interactions, the concept of the human– animal rela-tionship exists not only for each partner of the relationship but also for an observer (Aureli and Schaffner, 2002), thereby allowing the relationship to be studied.

Thus, human–animal relationships can be considered to be constructed from a series of interactions between humans and animals. An important methodological feature associated with this concept is the necessity of characterizing those inter-actions that have signifi cance for both the human and animal partners, so that the infl uence of the quantity and nature of these interactions on the human–animal relationship can be understood. This approach has been clearly illus-trated by our research conducted on stockpeople and farm animals. As discussed later (Section 3.5 and Section 5.4), evidence from handling studies and observa-tions on human–animal interactions in the livestock industries indicate that it is

50 Chapter 3

this history of interactions between humans and the animal that leads to the development of a stimulus-specifi c response of farm animals to humans: through conditioning, farm animals may associate humans with rewarding and punishing events that occur at the time of human–animal interactions and so conditioned responses to humans develop. These behaviours by humans may be tactile, visual, olfactory and auditory, and the nature of these behaviours may be positive, neutral or negative for the animal. For example, fear-provoking behaviours, such as the sudden unexpected appearance of a human or a human looming over an animal may be negative for the animal, while painful behaviours such as a hit by a human are obviously negative to animals. Therefore, it is the nature and frequency of these human behaviours that markedly determine the quality of the human–animal rela-tionship for farm animals. Similarly, as discussed in Chapter 4 (Section 4.2), it is the stockperson’s direct and indirect experiences with animals that are infl uential deter-minants of the stockperson’s attitudes and behaviour towards farm animals.

The quality of this relationship from the perspective of the animal can therefore be assessed by measuring the behavioural response of the animal to the handler or to other humans. In other words, the behaviour of the animal in the presence of the handler or other humans (e.g. the approach and avoidance responses of the animal to the human) will provide information on the quality of the human–animal rela-tionship for the animal. Most of the previous research on these relationships in the livestock industries has been specifi cally focused on the fear responses of the ani-mals to humans because of their implications for animal productivity and welfare. However, it needs to be recognized that because the relationship from the animal’s perspective develops from the history of its interaction with humans, the animal’s perspective is likely to be determined not only by negative emotional states, such as fear, but also by positive emotional states generated by interaction with humans. Indeed, as recognized for example by Boivin et al. (2003) and Waiblinger et al.(2006), the range of emotions in an animal generated by the interaction with humans, which may vary from positive/pleasant to negative/unpleasant, is likely to determine the strength of an animals’ relationship with humans, which may therefore vary from negative through neutral to positive. The consequences of such emotional states on animal welfare during handling are easily appreciated, with fear of humans causing stress and therefore impairing the welfare of the animal. Conversely, positive emotional states may provide some protection from unfamiliar handling practices or situations or even painful husbandry proce-dures. While this chapter focuses more on the implications of poor human–animal relationships, the implications of a positive human–animal relationship will be considered later (Section 8.3).

3.3 Behavioural Responses of Domestic Animals to Humans

There are marked between-species and within-species differences in the behav-ioural responses of farm animals to humans. For example, the fl ight distance to humans, which generally is defi ned and measured as the distance at which an animal withdraws or escapes as a human approaches (Hediger, 1964), varies markedly both between and within farm animal species. Murphey et al. (1981)

Human–Animal Interactions 51

reported marked differences in the fl ight distance between Bos indicus and Bostaurus breeds of cattle to humans, and Hearnshaw et al. (1979) reported marked differences between crossbred Brahman cattle and British breeds in the speed that they exit a solid-sided box or squeeze chute (stall) following brief restraint. Indeed, ‘fl ight speed’ in response to this brief restraint, which some have consid-ered a measure of cattle temperament, and possibly innate fearfulness (Petherick et al., 2002), is moderately heritable in Bos indicus cattle (Burrow and Corbet, 2000). Furthermore, the fl ight distance of extensively grazed farm animals is generally reported to be greater than that of intensively managed farm animals, perhaps as a consequence of less contact with humans in extensive situations. For example, there are reports of fl ight distances of 6–11 m for extensively grazed or rangeland sheep, and 31 m for extensively grazed beef cattle in comparison with 4–5 m for more intensively handled sheep on pasture, 2–8 m for feedlot beef cattle and 0–7 m for dairy cattle (Grandin, 1980, 1993; Hutson, 1982; Purcell et al., 1988; Hargreaves and Hutson, 1990c).

These differences in the avoidance responses of animals to humans may, in part, refl ect inherent species differences in their fear of unfamiliar stimuli or gen-eral fearfulness. Selection is more likely to affect the general fearfulness of the naïve animal rather than infl uence responses to specifi c novel stimuli, a scenario expressed in the behaviour of both wild and domestic animal populations (Price, 1984). Inherent species differences in general fearfulness will affect the initial responses of naïve animals to novel stimuli such as humans; however, over time, experience with humans should modify these responses to the extent that they become stimulus specifi c. Murphy and Duncan (1977, 1978) studied two stocks of chickens, termed ‘fl ighty’ and ‘docile’ on the basis of their behavioural responses to humans, and found that early handling affected the behavioural responses of these two stocks of birds to humans, with the docile birds showing a more rapid reduction in their withdrawal responses with regular exposure to humans than the fl ighty birds. Birds of the fl ighty stock originated from a hybrid derived from the White Leghorn breed and those of the docile stock from a hybrid derived from a cross of Light Sussex and Rhode Island Red breeds. These stock differences in the responses to humans may be stimulus specifi c because observations indicated that the docile birds did not necessarily show fewer with-drawal responses to novel stimuli, such as a mechanical scraper and an infl ating balloon, than the fl ighty birds (Murphy, 1976).

Further evidence that the handling effects on the behavioural response of animals to humans may be specifi c to humans and not generalized to a range of fear-provoking stimuli is provided in a series of studies by Jones and colleagues (Jones et al., 1991; Jones and Waddington, 1992). These scientists examined the effects of regular handling on the behavioural responses of quail and domes-tic chickens to novel stimuli (such as a blue light) and humans, and found that handling predominantly affected the responses of birds to humans rather than to the novel stimuli. Handled birds showed less avoidance of humans but their responses to novel stimuli were unaffected. These data indicate that experience with humans results in stimulus-specifi c effects rather than effects on general fearfulness. In contrast, Lyons et al. (1988) reported that early human contact not only affected the behavioural responses of goats to humans but also affected

52 Chapter 3

the behavioural responses of goats to a range of novel stimuli: in comparison with dam-reared goats, those reared by humans showed increased approach to and less avoidance of a number of stimuli, including humans. However, humans were involved in testing the responses of these goats to novel stimuli and thus their response to humans at the time of testing may have infl uenced their responses to novel stimuli. In fact, many of the studies that have examined the behavioural responses of animals to novel stimuli have used testing procedures that involved the presence of humans around the time of testing. Furthermore, in the study by Lyons et al. (1988), and indeed in a number of other studies, early weaning also occurred in the human-rearing treatment, and the stress of early weaning may have affected the subsequent behaviour of the human-reared goats to novelty in a similar manner to that observed in the so-called ‘early han-dling studies’. In many of these early handling studies, a stressor was shown to have marked effects, through its effects on developmental processes, on the ani-mal’s subsequent behavioural and physiological responses to novelty (Schaefer, 1968). The early handling effects are discussed in more detail later in this chapter (Section 3.5.1).

Over time, young domesticated animals that have had limited experience with humans may habituate to the presence of humans and thus may perceive humans as part of the environment without any particular signifi cance. Habitu-ation will occur over time as the animal’s fear of humans is gradually reduced by repeated exposure to humans in a neutral context, that is, the human’s presence has neither rewarding nor punishing elements. Even wild strains that are highly fearful of humans show evidence of habituation to humans. Galef (1970) tested the effects of several rearing experiences on the behavioural response of wild Norway rats to handling by humans. Second- and third-generation rats, which were housed in a laboratory, were reared by either wild or domestic rats, reared with either wild or domestic littermates and received either regular handling or no handling from 10 to 23 days. At weaning at 23 days of age, the behavioural response of these animals to handling was observed and it was found that only those animals that had had physical contact with humans (handling) showed minimal withdrawal responses to capture. Wild-caught deer and deer bred in captivity both habituate to the presence of humans, with their fl ight distance to humans reducing to 30 m or less (Matthews, 1993).

Some domestic animals, such as farm and laboratory animals and, indeed, some pets, such as aviary birds housed in groups, which may receive limited human contact, may perceive humans as predators. Selection for increased docility in the presence of humans has accompanied domestication; however, based on their withdrawal responses to humans, domestic animals may still fi nd human contact aversive and thus perceive humans as predators. Suarez and Gallup (1982) suggested that the predominant response of naïve domestic poultry to humans may be a response to predators. Bos indicus cattle extensively grazed with infrequent human contact display extreme avoidance responses to restraint and human presence including, at times, tonic immobility or a catatonic-like state during restraint (Grandin, 1980), which are indicative of anti-predator responses.

Domestic animals in situations in which they frequently interact with humans, may, through conditioning, associate humans with rewarding and punishing

Human–Animal Interactions 53

events that occur at the time of these interactions and thus conditioned responses to humans may develop. Studies examining the effects of a range of handling treatments on the behaviour of pigs (see Section 3.5.1) indicate that conditioned approach–avoidance responses develop as a consequence of associations between the stockperson and aversive and rewarding elements of the handling bouts. Pigs that were slapped or shocked with a battery-operated prodder when-ever they approached or failed to avoid the experimenter in daily handling bouts of 15–30 s, learned to associate the presence of the handler with the punishment of the handling bouts. In contrast, pigs that received pats or strokes during brief daily handling bouts subsequently showed increased approach to humans. There is also evidence that pigs may associate the rewarding experience of feed-ing with the handler and that this conditioning results in pigs being less fearful of humans (Hemsworth et al., 1996d). Although there is some controversy over the mechanism by which avoidance behaviour becomes conditioned by punishment (Walker, 1987), it is well established that animals learn to avoid conditioned stimuli that are paired with aversive events. Thus, through conditioning, the behavioural responses of farm animals to humans may be regulated by the nature of the experiences occurring around the time of interactions with humans.

Animals may also perceive humans as social partners and this is likely in situations where young animals form a long-lasting bond or attachment to humans. As reviewed by Rushen et al. (2001), the relationship between humans and domestic animals, particularly companion animals and, to a lesser extent, farm animals, is often considered a social relationship. If the defi nition of a social relationship between two individuals includes preference for interaction and proximity for each other (i.e. affi liative behaviour) which is similar to that for conspecifi cs, it is questionable that human–animal relationships in commercial livestock systems are genuine social relationships. However, one could argue that such relationships exist between humans and their companion animals, as demonstrated by the extensive studies by John Paul Scott (see Scott, 1992) on the effects of early human contact on the responses of dogs to humans. These studies showed that early human contact can result in the formation of social attachment by dogs to humans, whereby dogs act to maintain proximity with humans and show signs of separation stress when separated from humans. Social attachment, whether to dogs or humans, occurs most rapidly in dogs from 3 to 12 weeks, with a peak between 6 and 8 weeks, and as little as a few minutes per day of visual contact or 20 min of visual contact twice a week appears to affect attachment (Scott, 1992).

3.4 Fear of Humans by Domestic Animals

Fear of humans by animals is often assessed on the basis of the behavioural and physiological responses of animals to humans. The behavioural responses may include escape–avoidance behaviour or aggressive behaviour to humans. Approach behaviour to humans may also occur when fear wanes and the animal commences to explore and investigate the human stimulus, and thus approach behaviour can be usefully considered as an inverse measure of level of fear of

54 Chapter 3

humans. The physiological responses associated with fear include responses of the autonomic nervous system (e.g. secretions of adrenaline and noradrenaline) and the neuroendocrine system (e.g. secretions of the corticosteroid hormones cortisol and corticosterone). These physiological responses are the same indica-tors used to study stress (Section 2.5.1.1).

Fear, as with other emotions, cannot be measured or observed directly, but can only be inferred from the behaviour and stress physiology of the animal. This limitation in its measurement has contributed to the considerable controversy that exists with the defi nition of fear, its use by scientists and its measurement (see Hinde, 1970; Murphy, 1978). Toates (2004, p. 189) described the function of emotions as ‘certain states of the central nervous system that play a role in the co-ordination of cognition, internal physiology and external behaviour. For example, a state of fear simultaneously underlies (a) excitation of escape, avoid-ance or freezing and (b) inhibition of appetitive behaviour’.

3.4.1 Defi nition

At its simplest, fear can be viewed as an intervening variable, linked, on the one hand, to a range of stimuli that may pose some risk or danger to the welfare of the animal and, on the other hand, to a series of responses, both behavioural and physiological, by the animal that enable it to respond appropriately to this source of danger. However, many authors consider fear as a motivational state which is aroused by certain specifi c stimuli and normally gives rise to defensive behaviour or escape (McFarland, 1981; Hogan, 2008). Motivate means to move and is concerned with what ‘moves’ an animal to act in a certain way (Toates, 1980) and thus the study of motivation aims to understand the modulation of stimulus–response relationships. While motivation cannot be observed directly and is dependent upon internal and external stimuli, Toates (1980) proposes that this is no argument against motivation having a provisional usefulness. Thus, while we consider fear in this book as an emotion like pleasure/happiness, fear has been considered by some writers in the same sense as thirst, hunger and sex are considered motivational states, affected by both internal and external stimuli.

In relation to external stimuli, Gray (1987) recognizes that fear may be trig-gered by environmental stimuli that are novel, have high intensity such as loud and large stimuli, have special evolutionary dangers such as heights, isolation and darkness, arise from social interaction such as contagious learning or have been paired with aversive experiences. While fear appears to be activated almost exclusively by external stimuli (Hogan, 2008), intrinsic factors are also impor-tant. For example, social, gender, breed and strain effects on fear responses have been shown in both laboratory and farm animal species. In open-fi eld tests, which are often used in behavioural studies to assess fear, Jones (1983) found that pair-tested chickens generally froze less, vocalized and ambulated sooner and preened and pecked more than those tested individually. Breed and strain differences in fear responses have been demonstrated in many farm animals (Mills et al., 1997). For example, Faure (1981) has shown that selection for high ambulation in an open-fi eld test is associated with reduced latency to and

Human–Animal Interactions 55

increased ambulation in domestic chicks. Female rodents generally exhibit more ambulation and less defaecation than males in open-fi eld tests (Gray, 1987), while Vandenheede and Bouissou (1993) found that rams show more approach to a stationary human than ewes.

Fear is usually listed among the emotions and Gray (1987) defi nes fear as a form of emotional reaction to a stimulus that the animal works to terminate, escape from, or avoid. Furthermore, Jones and Waddington (1992) consider fear as an undesirable emotional state of suffering. While the existence of emotional feelings in animals is a contentious issue in science, biologists generally consider that animals are restricted to a few basic emotions, such as anger, fear, joy and hap-piness (Bolles, 1981; Section 2.5.1.2). Therefore, fear can be considered as an undesirable emotion that gives rise to defensive behaviour or escape. In concert with these behavioural effects, fear normally activates the autonomic nervous system and the neuroendocrine system and, as discussed in Section 3.4.2, these systems, through their effects on regulatory mechanisms such as energy avail-ability and use, and cardiac and respiratory functions, assist the animal to meet the physical and/or emotional challenges of the fear-provoking situation.

Farm animals in the presence of humans commonly display behavioural patterns that can be labelled fear responses, such as withdrawal from or avoidance of humans, as well as immobility responses such as freezing or crouching in the close presence of humans (Hemsworth and Barnett, 1987; Jones, 1987; Mills and Faure, 1990). It is reasonable to label these responses as fear responses because it is generally accepted that fear responses are normal, adaptive responses that function to protect the animal from harmful stimuli (Toates, 1980). What is surprising when studying farm animals is the magnitude of these fear responses to humans given that these animals have been domesticated over many generations and that there is generally substan-tial contact between humans and farm animals in modern livestock production systems.

The view that fear is an emotion, affected by both internal and external stim-uli that elicit behavioural and physiological responses which, in turn, may assist the animal to respond appropriately to a source of danger, is a useful construct that enables the study of the proximate or immediate causes and ontogeny of these fear responses to humans. An understanding of the causal factors may provide the opportunity to reduce fear responses to humans in livestock and, therefore, the main factors that have been identifi ed that regulate these responses in livestock are considered in detail in Section 3.5.

3.4.2 Behavioural and physiological responses to humans

The behavioural and physiological responses that fearful animals may display in the presence of humans will be considered in more detail, because these responses have important implications for the productivity and welfare of live-stock. These biological responses are the stress responses considered earlier (Section 2.5.1.1) and they will be reviewed here in the context of exposure of the animal to a fear-provoking stimulus.

56 Chapter 3

It will be recalled from the previous chapter (Section 2.5.1.1) that the animal has four types of interrelated biological responses available when confronted with a potential challenge (stressor): behavioural responses and responses of the autonomic nervous system, the neuroendocrine system and the immune system. Thus, when an animal that is highly fearful of humans is in the close presence of a human, the central nervous system integrates these responses and it is these responses that provide the principal resources that the animal utilizes in its attempts to cope with this stressor. The close presence of a human is likely to initiate a series of adaptive or coping responses, often called the ‘fi ght or fl ight responses’ or the ‘emergency reaction’ (Cannon, 1914), which may include escape or avoidance responses, as well as responses of the autonomic nervous system, with secretions of catecholamines (e.g. adrenaline and noradrenaline) and elevations in heart rate, blood pressure and body temperature, which pre-pare the animal for these behavioural responses. A particularly important bio-logical effect is the adrenaline-dependent production of glucose from liver glycogen (glycogenolysis) for an immediate energy supply. These autonomic nervous system responses therefore function to mobilize the body’s reserves for an appropriate and immediate reaction to this challenge.

These ‘fi ght or fl ight’ responses last for only a short period and, if the stressor (human) is not removed, a second series of events occurs. This is the short-term or acute stress response, which may last from minutes to hours and predomi-nantly consists of increased secretions of the corticosteroid hormones cortisol and corticosterone. The magnitude of acute physiological responses should refl ect the intensity of the stressor (Harbuz and Lightman, 1992) and, indeed, studies on a number of farm animals have shown that highly fearful animals, assessed on the basis of their behavioural responses to humans, have a greater acute cortisol response to brief exposure to humans than animals that are less fearful of humans (Fig. 3.2). This acute stress response has the major function of providing glucose from food or muscle protein (gluconeogenesis) for the required increased metabolic performance necessary to respond to the fear-provoking situation. If the stressor (human) is removed, this physiological state will disappear with possibly no real ill-effects on the animal apart from a deple-tion of energy reserves.

The short-term behavioural and physiological responses of the two strains of birds studied by Murphy and Duncan (1977, 1978), termed ‘fl ighty’ and ‘docile’ on the basis of their behavioural responses to humans, were monitored by Jones et al. (1981). Orientation, withdrawal and heart rate were measured as a human slowly approached. The behavioural and physiological changes had similar time courses and the differences observed refl ected the original classifi cation of the two strains: when the human was close, the fl ighty strain showed greater with-drawal and increased heart rate. It is of interest that when the human was fi rst observed at a distance, the docile birds showed evidence of higher fear, per-haps because the human was not recognized as a human at this distance. As mentioned earlier, these stock differences may be stimulus specifi c, as the docile birds do not necessarily show fewer withdrawal responses to novel stimuli, such as a mechanical scraper and an infl ating balloon, than the fl ighty birds (Murphy, 1976).

Human–Animal Interactions 57

If the stressor continues (i.e. these short-term responses are ineffective in enabling the fearful animal to avoid or alleviate the challenge of the close pres-ence of the human), the response continues to the third series of events, which is the long-term or chronic stress response. This response is also corticosteroid dependent and comes at a physiological cost to the animal: prolonged activation of the hypothalamic–pituitary–adrenal axis results in decreased metabolic effi -ciency and thus growth performance, impaired immunity and reduced reproduc-tive performance. Research, predominantly on pigs, has shown that high levels

50

30

10

–10

–30

Cha

nge

in c

ortic

oste

roid

s (%

)C

hang

e in

cor

ticos

tero

ids

(%)

–50

240

190

140

90

40

–10

–60

0 5 10 15 20

Positively handled

Negatively handled

Positively handled

Negatively handled

0 10 20Time (min)

Time (min)

30 40

Fig. 3.2. Percentage change in plasma cortisol concentrations in positively handled and negatively handled cows (top fi gure) and pigs (bottom fi gure) after a 2-min exposure to a human (data reanalysed from Breuer et al., 2000, and Hemsworth et al., 1981a).

58 Chapter 3

of fear of humans resulted in a chronic stress response, which in turn markedly reduced the growth and reproductive performance of the animal (see Section 3.6.1). Handling experiments with young chickens have also shown that birds that received brief human contact, of an apparently positive nature, had improved growth rates, feed conversion and antibody response to an antigen, and were more resistant to Mycoplasma gallisepticum and Escherichia coli than birds that either received minimal human contact or had been regularly scared (see Section 3.6.2).

3.4.3 Measurement of fear for humans

A number of studies have assessed fear levels by measuring the amount of avoid-ance of the stimulus or, conversely, the amount of approach to the stimulus in standard testing situations. The rationale behind these assessments is that while there may be a number of behavioural patterns available to the animal that may be equally effective in the fear-provoking situation, the amount of avoidance or, conversely, approach provides an integrated measure of the fear levels without having to make judgements about the relative signifi cance of specifi c behav-ioural patterns. For example, a decision that freezing is indicative of higher or lower levels of fear than orientation away from the stimulus or vigorous escape from the stimulus is not required. Acute physiological stress responses such as rises in plasma corticosteroid concentrations and heart rate have also be used to assess fear, often in conjunction with behavioural measures.

In studying the behavioural responses of farm animals to humans, we have adopted a functional view (Hemsworth et al., 1993) and have used the avoid-ance behaviour of the animal to an approaching experimenter or, conversely, the approach behaviour of the animal to a stationary experimenter to assess the animal’s fear of humans. For example, in some studies on pigs and cattle, we have measured the time to closely approach and time spent near a stationary experimenter in a standard arena. Hemsworth and Barnett (1987) have described the sequence of behavioural responses of pigs to humans that may occur in an approach test of this type, and this sequence of behaviour as affected by the animal’s fear of humans is depicted in Fig. 3.3. It should be appreciated that there are likely to be competing motivations of avoidance and exploration in these approach tests. While the degree of novelty of the test arena may be reduced by the similarity of the arena with the animals’ home pen or familiar environment, animals introduced into this new environment will be motivated to explore and familiarize themselves with the test environment once the initial fear responses have waned. Therefore, although the animals may be motivated to both avoid and explore the arena and the human stimulus, the animal’s fear of humans will have a major infl uence on its approach to the human stimulus. That is, when comparing animals in the same novel environment, there will be a systematic difference in approach behaviour between fearful and non-fearful animals.

There is evidence for this interpretation. A signifi cant negative association between the magnitude of the cortisol response of pigs exposed to an experimenter

Human–Animal Interactions 59

in their home pen and the approach behaviour of pigs to a stationary experimenter in the standard test support this behavioural assessment of fear (Hemsworth andBarnett, 1987). Similarly, Lyons et al. (1988) found that dam-reared goats that showed less approach to and greater avoidance of humans had a higher cortisol response to human presence than human-reared goats. Furthermore, the impo-sition of handling treatments designed to differentially affect pigs’ fear of humans produced the expected variations in the approach behaviour of pigs to a station-ary experimenter in the standard test (Hemsworth et al., 1981a, 1986c, 1987; Gonyou et al., 1986; Hemsworth and Barnett, 1991).

As poultry show little locomotion in a novel arena in the short term, the avoidance responses of birds to an approaching human have often been used to assess fear levels. Orientation away and withdrawal from the approaching human have been equated with high fear levels. Handling treatments intuitively expected to reduce the fear of humans by poultry have reduced orientation away and withdrawal from the experimenter in several standard tests measuring the avoidance of humans (Jones and Faure, 1981; Barnett et al., 1992; Jones and

Human

Stage 1Orientiationresponse

Stage 2Escape

Vocalization

Stage 3Immobile

Orient away

Stage 4ImmobileOrient to

Stage 5Approach and

investigatehuman

Stage 6Closely approachand investigate

human

Fig. 3.3. Sequence of behavioural responses of pigs to exposure to a stationary human. Asterisk refers to habituation of fear and, given suffi cient time, the animal will move to the next stage (adapted from Hemsworth and Barnett, 1987, with permission from the publishers).

60 Chapter 3

Waddington, 1993; Hemsworth et al., 1994c, 1996a; Edwards, 2009). These tests are similar to tests that have been utilized to measure fl ight distance in sheep and cattle (e.g. Hargreaves and Hutson, 1990c), defi ned as the distance at which an animal withdraws as a human approaches.

In some studies on cattle, pigs and poultry, we have used both approach behaviour to a stationary experimenter and avoidance behaviour to an approach-ing experimenter (Fig. 3.4) to study fear of humans. In a study by Hemsworth et al. (1981b), in which the fear responses of commercial breeding sows to humans were assessed on the basis of both the approach behaviour of sows to a stationary experimenter and avoidance behaviour to an approaching experi-menter, the two measures were correlated with reproductive performance. Approach behaviour of sows at the farm, on the basis of the time spent near a stationary experimenter, was positively correlated with farrowing rate (percentage

Fig. 3.4. The avoidance behaviour of poultry to humans has been used to assess their levels of fear of humans. This photograph shows an experimenter fi lming the withdrawal of chickens as he moves through the meat chicken unit in a standard manner.

Human–Animal Interactions 61

of sows mated that farrowed) at the farm, while avoidance behaviour, on the basis of time taken to return to feeding after the experimenter approached, was negatively correlated with farrowing rate. Breuer et al. (2003) used similar mea-sures to assess the fear responses of heifers to humans – approach behaviour to a stationary experimenter and fl ight distance to an approaching experimenter – and found that differing handling treatments resulted in corresponding effects on these measures of fear.

As mentioned earlier in this section, there are several criticisms of the use of behavioural responses as measures of fear levels. Behavioural responses and indeed physiological responses can be criticized on the grounds that the responses are not specifi c to the emotional state of fear and that motivational states, such as hunger, may confound the assessment. However, this is also true when attempting to assess motivational states. For example, mounting may occur in pigs in both sexual and aggressive encounters. A lack of sexual activity by a ram in a sexual setting may not be a consequence of reduced sexual motivation at the time, but may be a consequence of a competing motivational state, such as fear of the close presence of a dominant ram, inhibiting sexual motivation. As with fear, measurement of these states can also be diffi cult, relying on inference from observation of the responses. Studies under controlled conditions in which con-fl icting motivations are eliminated or controlled enable the study of the proxi-mate causes and ontogeny of specifi c motivational states or behavioural systems of interest.

Because a number of behavioural patterns may be available to the animal in a fear-provoking situation, and as a number of these patterns may be equally effective for the animal in avoiding the danger, a number of authors have also argued that there is little scientifi c basis for ranking these behavioural patterns in terms of fear levels (Murphy, 1978). For instance, is freezing indicative of higher or lower levels of fear than orientation away from the stimulus or vigorous escape from the stimulus? This particular problem is one of the main reasons that a functional approach, which does not rely on judging particular patterns of behav-iour has been used to measure fear: a functional approach to assess fear is based on the amount of avoidance of the stimulus or, conversely, the amount of approach to the stimulus in standard testing situations.

In a recent review of the assessment of the human–animal relationship, Waiblinger et al. (2006) concluded that while few tests of fear of humans were validated, tests measuring an animal’s behavioural response to either a station-ary or approaching human have to contend with possible competing motivations or behavioural systems that may differ between the test categories. In tests in which humans approach as distinct from those in which humans are stationary, fear responses may be easier to interpret than motivations such as curiosity or exploration. In contrast, when testing an animal’s approach to a stationary human, latency in approaching and interacting by the animal may vary accord-ing to the level of curiosity. However, as discussed earlier, although an animal may be motivated to both avoid and explore the arena and the human stimulus, the animal’s fear of humans will have a major infl uence on its approach to the human stimulus. This approach to assessing fear is supported by the fi ndings of behavioural and physiological correlates in these tests (see earlier in this section),

62 Chapter 3

together with fi ndings that imposition of handling treatments designed to differ-entially affect an animal’s fear of humans generally produced the expected vari-ations in the behavioural responses of animals to humans (see Section 3.5). Furthermore, the issue of competing motivations such as curiosity or exploration of the test location and effects such as novelty and social isolation may also con-found assessment of fear responses to humans in tests involving an approaching human. However, when comparing groups of animals of the same strain under conditions that are similarly novel for both groups but in which the groups have different histories of interactions with humans, their approach behaviour to humans should be attributable to fear because the competing motivations would be similar for both groups.

There are other concerns relating to assessing fear responses to humans, some of which have been discussed by de Passille and Rushen (2005) and Waiblinger et al. (2006). These include: the validity and repeatability of fear measures, particularly in commercial settings; the different fear measures avail-able; effects of context (e.g. testing setting, individual person and his/her stimulus properties, including posture and clothing); effects of testing location relative to the routine handling location; identity of test person (e.g. stimulus generalization versus stimulus discrimination); pretesting effects, particularly pretest handling; and the artifi cial nature of the test (e.g. novelty of the testing setting). Clearly these limitations need to be recognized and tests to assess fear of humans need to be validated.

3.5 Human Contact Affecting the Behavioural Responses of Farm Animals to Humans

Modern intensive management of livestock involves several levels of interaction between stockpeople and their animals. Many of the human interactions are associated with regular observation of the animals and their conditions and, consequently, this type of interaction often only involves visual contact between the stockperson and the animals. For example, stockpeople inspect meat chick-ens and their conditions in indoor deep-litter production systems by moving through the groups several times a day. Similarly, in caged housing systems for laying hens, stockpeople move along the corridors several times a day inspecting birds and their conditions. Larger animals in most production systems have to be moved and, in addition to visual and auditory contact, stockpeople will often use tactile behaviours to move their animals. Extensively grazed animals, such as cattle and sheep, are moved between pastures as part of optimal pasture man-agement, and extensively grazed dairy cows and indoor-housed dairy cows are moved several times a day during lactation to be milked. Growing pigs are gen-erally moved from pen to pen in order to provide accommodation suitable to their stage of growth, while breeding pigs may be regularly moved according to the stage of their breeding cycle. It is in these situations, in which animals have regular albeit brief contact with humans, that human behaviours have consider-able potential to infl uence the immediate behaviour of the animals, as well as the subsequent behavioural responses of the animals to humans.

Human–Animal Interactions 63

Human interactions also occur in situations in which animals must be restrained and subjected to management or health procedures. Some animals may never be restrained in a production system, while others are restrained on a regular basis. The association of fear and pain from these husbandry procedures with the humans performing them may increase the fear of humans that animals subsequently exhibit in both similar situations involving humans and different situations involving humans, including routine inspections.

Handling studies on farm animals are useful in identifying the type and nature of human contact that may regulate the behavioural responses of farm animals to humans, and some of the main studies in this literature are considered in this section.

3.5.1 Tactile contact by humans

Handling studies, predominantly on dairy cattle, pigs, poultry and sheep, indi-cate that the behavioural response of farm animals to humans is particularly affected by tactile interactions from humans.

Handling studies on pigs have consistently shown that negative handling imposed briefl y but regularly over weeks or months increases their fear of humans. Pigs that received a brief slap, hit or shock with a battery-operated prodder whenever they approached or failed to avoid an experimenter during brief handling bouts, subsequently were slower to approach and physically inter-act with, spent less time near and interacted less frequently with the stationary experimenter in an unfamiliar arena than pigs that received a pat or stroke when-ever they approached the experimenter (Hemsworth et al., 1981a, 1986c, 1987; Gonyou et al., 1986; Paterson and Pearce, 1989, 1992; Pearce et al., 1989; Hemsworth and Barnett, 1991). Furthermore, pigs handled in a negative man-ner had a higher acute cortisol response to brief exposure to the experimenter in their home pens than pigs handled positively (Fig. 3.2; Hemsworth et al., 1981a, 1986c, 1987). In general, pigs receiving minimal human contact in these experi-ments were intermediate in their fear responses to humans. Pedersen et al.(2003) found that sows that were stroked regularly around weaning tended to approach and interact quicker with a stationary experimenter at weaning than sows that were snout snared once and then in subsequent daily handling bouts were shouted at and hit if they approached. Tanida et al. (1995) also found that pigs that were regularly stroked were quicker to physically interact and spent more time interacting with humans than pigs that were not handled, and Day et al. (2003) found that growing pigs that received daily close human contact in the form of a stockperson entering their pens and stroking approaching pigs, physically interacted more with the stationary stockperson in their pens in a series of tests than pigs that generally only received routine inspection without the stockperson entering their pens. In many of the above experiments, patting or stroking in the positive handling treatments was accompanied by slow movement and squatting by the experimenter to encourage pigs to approach (Fig. 3.5).

As with pigs, there is good evidence from handling studies on dairy cows that negative tactile interactions will increase their fear of humans, while positive

64 Chapter 3

tactile interactions will decrease fear of humans. Dairy heifers that were slapped or hit whenever they approached or failed to avoid the experimenter subse-quently were slower to approach and interact with, spent less time near and interacted less frequently with a stationary experimenter in an unfamiliar arena than those that were positively handled with slow and deliberate movements, pats and strokes (Breuer et al., 2003). The former group of heifers also showed a greater fl ight distance to an approaching experimenter. Treatment effects were apparent when either the familiar or an unfamiliar experimenter was used in the behavioural tests. Breuer et al. (1997) found that fl ight distance in lactating heif-ers was increased when moderate or forceful slaps were briefl y imposed before and after milking when animals failed to avoid the close approach of the experi-menter. While the handling treatments in these two experiments were imposed briefl y but regularly for weeks, negative handling imposed over a shorter time has also been shown to increase fear responses of dairy calves and cows. de Passille et al. (1996) reported that dairy calves spent less time near and were slower to interact with an experimenter that had restrained them by nose tongs or shocked or threatened them with a battery-operated cattle prodder than to an experimenter that either patted and fed them or spent time in close visual contact with them. Rushen et al. (1999) found that dairy cows in their home pens showed greater avoidance of an experimenter that had hit or shocked them with a battery-operated cattle prodder than of an experimenter that had brushed, fed and gently spoken to them. Schmied et al. (2008) found that stroking the ventral region of the neck of dairy cattle encouraged more neck stretching, lowered heart rate and increased interactions with the experimenter. In cattle, intraspecifi c

Fig. 3.5. Slow movement and squatting will encourage pigs to approach and interact with humans (photograph courtesy of Wageningen UR Communication Services).

Human–Animal Interactions 65

social licking in this area is very common and these results suggest that stroking this preferred region may be perceived more positively than a neutral interaction by humans.

Long-term handling studies on dairy calves have shown that positive tactile interactions will decrease their fear of humans. Lensink et al. (2001b,c) found that calves that were regularly stroked on their necks and shoulders and allowed to suck the stockperson’s fi ngers spent more time interacting with a stationary unfamiliar experimenter in their home pens and showed less withdrawal to an approaching unfamiliar experimenter in front of their pens than calves that received minimal contact with the experimenter around feeding. Similarly, Lensink et al. (2000b) found that calves that received similar positive handling withdrew less from the approach of both an unfamiliar experimenter and the familiar experimenter in their home pens, and interacted more frequently and for longer with both experimenters standing stationary in an unfamiliar arena than calves that received minimal contact with the experimenter around feeding. Lensink et al. (2001a,b) also found that the dairy calves that predominantly received positive handling during rearing had lower heart rates during loading for transport than those that received either minimal human contact or predom-inantly negative handling, such as hitting and shouting, during rearing. Boissy and Bouissou (1988) examined the effects of the same total duration of positive handling but imposed at different periods in life on the behavioural response of dairy heifers to humans. Positive handling, involving brushing and leading on a halter, imposed from 0 to 9 months of age, reduced fl ight distance to an approach-ing experimenter at 9 months in comparison with positive handling imposed at either 0–3 months or 6–9 months, or minimal human contact from 0–9 months of age. While cortisol concentrations were lower in the handled heifers after capture and leading than in the non-handled heifers, heart rates were lower in the animals handled from 0 to 9 months after heart rate monitors were attached and the animals were tested in several tests than they were in the other groups.

There is some evidence from short-term early handling studies on dairy calves that feeding by humans and isolation from the dam may infl uence the extent of handling effects on calves. Jago et al. (1999) found that bucket feeding but not positive handling (consisting of talking, patting and stroking), reduced the latency of calves to approach a stationary unfamiliar experimenter in their home pens. Krohn et al. (2001) found that handling involving an experimenter assisting the calves to feed from a teat bucket and talking, patting and stroking calves, was effective, particularly in the fi rst 4 days of life, in subsequently increas-ing approach behaviour and reducing fl ight distance of calves to an unfamiliar experimenter. However, Krohn et al. (2003) found that this handling and feeding treatment was only effective if the calves were isolated from their dams. This lat-ter fi nding suggests that the presence of the dam at the time of handling may infl uence the subsequent behavioural responses of calves to humans.

In a series of handling studies on both meat- and egg-production strains of poultry, Jones and colleagues demonstrated the effects of regular but brief han-dling from an early age on the behavioural response of birds to humans. In an experiment with one meat- and two egg-production strains, Jones and Faure (1981) found that regular handling – involving catching and briefl y relocating to

66 Chapter 3

another pen – subsequently increased the approach of birds to a stationary experimenter in an unfamiliar pen and reduced the duration of tonic immobility following restraint in comparison with minimal human contact. Similarly, regular catching, holding and stroking chickens of an egg-production strain reduced the duration of tonic immobility and increased the approach to a stationary experi-menter in an unfamiliar pen (Jones and Waddington, 1992; Jones, 1993). It is of interest that in one of three experiments, Jones (1993) found that chickens that had regular visual contact with an experimenter in the form of gentle stroking of their wire-mesh cage wall showed less avoidance of the stationary experimenter than chickens that were regularly caught and stroked by the experimenter which, in turn, showed less avoidance of the stationary experimenter than chickens that had had minimal human contact. These results suggest that the treatment involv-ing catching, holding and stroking chickens may contain some aversive elements, and that habituation of the bird’s fear responses to humans over time with repeated exposure to humans is the main factor responsible for the reduction in fear observed in these handling treatments. These fi ndings will be considered further in Section 3.5.2. Jones (1993) also found that both suspending chickens by their legs and stroking chickens reduced their avoidance of a stationary exper-imenter, however the former treatment did not affect the duration of tonic immo-bility. Zulkifl i and Siti Nor Azah (2004) found that catching and stroking meat chickens, but not catching and suspending chickens by their legs, reduced the duration of tonic immobility.

Others have also examined the effects of handling on poultry. For example, Hughes and Black (1976) found that regular catching and briefl y relocating chickens to another pen during rearing reduced their avoidance of the experi-menter standing near the front of their cage as well as their avoidance of a novel rod placed by the experimenter in the front feed trough. In a handling study on two stocks of birds, termed ‘fl ighty’ and ‘docile’, Murphy and Duncan (1977) found that handling involving catching and briefl y relocating birds to another pen with feed reduced their avoidance of the approaching experimenter, but only in young birds of the docile stock and not in adult birds of either stock.

In a series of experiments, Boivin and colleagues demonstrated the long-term effects of positive handling from an early age on the behavioural response of lambs to humans. Both stroking and stroking in combination with bottle-feedingby an experimenter increased the time that lambs subsequently spent near the stationary familiar experimenter in an unfamiliar arena than those exposed to minimal human contact (Boivin et al., 2000, 2001, 2002). While the combina-tion of feeding and stroking increased the approach of lambs to the experimenter in subsequent tests, artifi cial rearing increased the magnitude of the handling effects (Boivin et al., 2001, 2002). This infl uence of artifi cial rearing suggests that the motivation to maintain contact with their dams may reduce the opportunity of dam-reared lambs to develop strong positive relationships with humans (Boivin et al., 2002). In a similar experiment, Tallet et al. (2005) examined the contributions of feeding and positive handling, involving catching, stroking and talking, on the subsequent approach behaviour of artifi cially reared lambs to a familiar experimenter in an unfamiliar arena. While the two treatments of posi-tive handling and positive handling plus bottle feeding increased the subsequent

Human–Animal Interactions 67

approach behaviour of lambs to the experimenter in comparison with a treat-ment involving the experimenter squatting in the lambs’ pens, the treatments of positive handling with or without bottle feeding resulted in similar approach behaviour.

Several studies on adult sheep have also demonstrated the effects of positive handling imposed briefl y but regularly over a number of weeks. Hargreaves and Hutson (1990a) found that talking to, stroking and feeding hay and lupins to sheep reduced their fl ight distance and heart rate in response to the approaching experimenter in comparison with minimal human contact. Mateo et al. (1991) found that sheep that were spoken to and stroked on the shoulders and necks by an experimenter whenever they approached were subsequently quicker to approach and interacted more frequently with the stationary experimenter in an unfamiliar arena than those that received no additional handling. In addition, ewes that were restrained in a head gate, but received similar auditory and tactile contact from the experimenter to that in the positive treatment, showed a similar approach to the experimenter to the ewes that received no additional handling. These latter results suggest that the negative nature of restraint may mask or negate the positive elements of talking to and stroking sheep.

Research by Pedersen and colleagues shows that handling of a positive nature both early in life and after weaning affects the subsequent behavioural responses of farmed silver foxes to humans. Juveniles that had been regularly stroked and talked to early in life showed increased approach to and reduced avoidance of the familiar experimenter standing stationary in front of their home cage in comparison with juveniles that had not been regularly handled (Pedersen and Jeppesen, 1990). While there were no treatment effects on cortisol concen-trations prior to an open fi eld test, the handled juveniles had higher cortisol concentrations following this test. Pedersen (1993) and Pedersen et al. (2002b) found that post-weaning handling also reduced the fear responses of both juvenile and adult silver foxes. Regular positive handling involving a slow approach to the cage, talking, offering feed and stroking resulted in more juveniles approaching an unfamiliar experimenter standing stationary in front of their cage (Pederson, 1993). Furthermore, regular catching and carrying in a cage after weaning resulted in similar approach behaviour to the positive handling. Pedersen et al. (1994) also found that when tested as juveniles and adults, animals that were spoken to, offered feed and stroked or regularly caught and carried showed less avoidance of an unfamiliar experimenter standing stationary in front of their cage in comparison with non-handled animals. Non-handled adult vixens had heavier adrenal glands than adult vixens in the two handling treatments, indicating that the handling treatments may have reduced stress.

Therefore, there is considerable evidence in farm animals that regular hand-ling involving positive tactile interactions, such as patting and stroking, reduces fear responses to humans assessed on the basis of the animal’s approach or avoidance behaviour to familiar or unfamiliar humans in either familiar or unfa-miliar locations. In some of these studies, particularly those on pigs and cattle, there is evidence that this positive handling also reduces the acute stress response to humans based on a reduced corticosteroid response to the presence of

68 Chapter 3

humans. It needs to be recognized that in addition to presumably positive tactile interactions, such as pats and strokes, many of these handling studies involved the experimenter moving slowly, squatting and quietly speaking to the animals. The effects of visual and auditory contact with humans will be discussed in Sec-tions 3.5.2 and 3.5.3. Furthermore, some of the handling treatments reported here were associated with the provision of food, often in an attempt to attract the animal to the experimenter, and thus the reduced fear responses to humans in some of these studies may have developed because the animals associated the presence of humans with the reward of food rather than the consequences of handling per se. For example with lambs, Boivin et al. (2000, 2001) found that feeding together with positive handling resulted in a greater increase in the approach behaviour of lambs to humans than either feeding or positive handling alone. Jago et al. (1999) found that feeding by humans, but not positive han-dling, increased the approach behaviour of calves to humans. Pigs fed in the close presence of humans were quicker to approach and spent more time near stationary unfamiliar and familiar experimenters than pigs that were fed in isola-tion of humans (Hemsworth et al., 1996d).

As discussed earlier in this chapter (Section 3.3), farm animals may form a long-lasting bond or attachment to humans, particularly if as young animals they have regular contact with humans. While it is arguable that human–animal rela-tionships in commercial livestock systems are genuine social relationships, there is evidence that farm animals may be more sensitive to human contact early in life. For example, Hemsworth et al. (1986b) found that regular but brief pat-ting or stroking of pigs for the fi rst 8 weeks of life increased their approach and

Fig. 3.6. Handling treatments involving initial passive interactions by the handler, such as squatting and remaining stationary, followed by patting and talking on approach by cattle, will reduce fear responses to humans by cattle (Hemsworth et al., 1996c).

Human–Animal Interactions 69

interaction with a stationary familiar experimenter in tests from 10 to 24 weeks of age in comparison with minimal handling of pigs early in life. Hemsworth and Barnett (1992) also found that regular but brief patting or stroking of pigs from either 0 to 3 or 9 to 12 weeks of age was more effective in reducing fear responses to humans based on their approach behaviour to a familiar experimenter at 18 weeks of age than when imposed at 3–6 and 6–9 weeks of age. Thus, pigs may be more sensitive to handling at 0–3 weeks of age and handling at 9–12 weeks may also be effective perhaps because it was the most recent handling. However, no treatment effects were apparent from 20 to 24 weeks of age, sug-gesting that any early handling effects were dissipated by subsequent human contact during routine husbandry and/or testing. The effects of early handling of cattle and sheep have also been studied. Brief human contact, consisting of stroking and assisting calves and lambs to feed in the fi rst few days of life has been shown to have long-term effects on the behavioural response of calves and lambs to humans (Boivin et al., 2001, 2002; Krohn et al., 2001, 2003). In con-trast, in an experiment using the same total duration of positive handling but imposed at different periods in life, Boissy and Bouissou (1988) found that brushing and leading calves on a halter from 0 to 9 months of age was more effective in subsequently reducing avoidance responses to humans than the same handling at either 0–3 months or 6–9 months of age. These studies on cattle, pigs and sheep indicate that, although farm animals may be more sensi-tive to handling at an early age, subsequent handling is also infl uential and has the potential to modify such early learning effects.

There is some limited evidence that farm animals may be sensitive to human contact at other times of their life. Hemsworth et al. (1987, 1989a) studied the effects of human contact at the fi rst calving of dairy cows. Allowing cows shortly after parturition to approach a squatting experimenter and sniff the outstretched hands that were smeared with fetal fl uid subsequently reduced their restlessness during milking and their fear responses to a stationary familiar experimenter. These results suggest that the parturient cow may not only be attracted to her newborn at this time, but that she may be more sensitive in general, including to the presence of humans.

The literature on early handling of rodents is relevant to the discussion of the possible effects of early handling of farm animals. This literature is very extensive and basically consists of two types of studies, those termed ‘handling studies’, which involved brief removal of pre-weaned animals from their home cages, and those termed ‘gentling studies’, which involved brief stroking of post-weaned animals. Although the results have often been contradictory, these treatments at times have resulted in increased growth and accelerated development, reduced activity and defaecation in an open-fi eld test, improved performance in learning tasks and reduced stress responses to subsequent stressors (Dewsbury, 1992). These results have been interpreted as a consequence of either direct stimulation or acute stress advancing the rate of development of some behavioural and physiological processes. Thus, early handling effects may not necessarily be a consequence of handling per se but may be a consequence of acute stress early in life advancing the rate of development (Schaefer, 1968). Furthermore, while early handling may have specifi c effects on the behavioural response of animals

70 Chapter 3

to humans, early weaning may also affect these responses, but via effects on behavioural development. Indeed, while there is some evidence that both may contribute, Boivin et al. (2001, 2002) have shown that positive handling without early weaning affects the subsequent behavioural responses of lambs to humans. Similarly, Hemsworth et al. (1986b) found that while pigs that were artifi cially reared and regularly handled early in life showed increased locomotion in a novel arena in comparison with those that were dam reared and regularly handled, both groups of pigs subsequently showed similar approach behaviour to a stationary experimenter.

Thus, for many farm animals, tactile interactions by humans that are likely to occur in commercial settings are important determinants of the animals’ fear of humans. However, there is evidence that other forms of human contact may affect the behavioural responses of farm animals to humans.

3.5.2 Visual contact with humans

The handling treatments in the studies described in the previous section also involved visual and auditory contact with humans and thus this non-tactile con-tact may have contributed to the handling effects. While it is easy to appreciate that regular negative tactile interactions with animals, particularly obviously neg-ative ones such as a slap, hit or shock with a battery-operated prodder, will increase fear responses to humans, the sensitivity of livestock to brief tactile con-tact, or even to non-tactile contact that intuitively appears to be of a minor or moderate negative nature such as shouting or unexpected movement, is surpris-ing. Understanding the sensitivity of animals to human contact is obviously criti-cal in handling animals in a manner that minimizes their fear responses to humans, so the effects of visual and auditory contact with humans on farm animals are discussed in this section (3.5.3).

A number of studies, particularly on poultry, have specifi cally examined the effects of human visual contact on fear. Chickens and laying hens appear to be particularly sensitive to visual contact with humans. Murphy and Duncan (1978) examined the effects of regular visual contact with stockpeople undertaking rou-tine husbandry on two stocks of birds, termed ‘fl ighty’ and ‘docile’, and found that chickens reared with visual contact with humans showed less avoidance of a stationary experimenter standing in front of their cages than those reared in isolation of humans. The handling effects were more pronounced and longer lasting in the so-called docile stock. Jones (1993) found that young chickens regularly observing others being caught and stroked showed similar avoidance of the stationary experimenter standing in front of their cages to those that were regularly caught and stroked. Both groups of handled chickens showed less avoidance of the experimenter than non-handled chickens. Furthermore, as dis-cussed in the previous section, Jones (1993) found that regular gentle stroking of the wire-mesh wall of the chickens’ cages reduced the subsequent avoidance of the stationary experimenter in comparison with regular catching and stroking. As both treatments allowed birds the opportunity to observe the experimenter either catching and stroking birds or stroking the cage wall, habituation of the bird’s

Human–Animal Interactions 71

fear responses to humans over time with repeated exposure to humans may have been responsible for the reduction in fear of humans. The fact that visual contact without tactile contact in the treatment involving stroking the cage wall was more effective in reducing fear than picking up and stroking the bird, also suggests that the latter handling may contain aversive elements for birds, such as active interaction and tactile interaction by humans.

Recently, Edwards (2009), in a factorial design study, examined three main handling effects on adult laying hens. The birds received the following treat-ments: (i) during rearing, either minimal human contact or additional human contact (involving 12 min daily visual contact with a stationary experimenter); (ii) in adulthood (from 20 weeks of age), either 30 s daily of positive handling or 5 s daily of negative handling; and (iii) in adulthood, the positive or negative handling imposed with the experimenter standing at either 0–45 cm, 45–75 cm or 75–105 cm from the cage. Positive handling involved slow approach and standing stationary while negative handling included unexpected appearance and fast movement by the experimenter. The avoidance response of the hens in their cages to an approaching human was assessed twice in adulthood (at 31 and 36 weeks of age) in a test in which two approaches to and two withdrawals from the caged birds were made by the experimenter over 20 s. During each approach and each withdrawal, the experimenter recorded whether the hen was in the front half or the back half of the cage. As shown in Table 3.1, additional human contact during rearing reduced avoidance behaviour to the experimenter at both ages, based on the proportion of birds that remained close to the experi-menter. Furthermore, the nature and proximity of handling in adulthood affected

Table 3.1. The effects of handling on the proportion of hens remaining in the front half of their cages in the close presence of the experimenter at 31 and 36 weeks of age (from Edwards, 2009).

Proportion of hens at the front of the cage

Treatmenta 31 weeks 36 weeks

Human contact during rearingAdditional human contact 0.42 0.48Minimal human contact 0.38 0.31Nature of handling in adulthoodPositive 0.44 0.40Negative 0.40 0.40Proximity of handling in adulthood0–45 cm from the cage 0.47 0.4245–75 cm from the cage 0.46 0.4875–105 cm from the cage 0.37 0.34

aTreatments: (i) during rearing, either minimal human contact or additional visual human contact; (ii) in adulthood, either positive handling or negative handling; and (iii) in adulthood, positive or negative handling imposed with the experimenter standing at either 0–45 cm, 45–75 cm or 75–105 cm from the cage.

72 Chapter 3

avoidance behaviour at 31 weeks and at both 31 and 36 weeks of age, respec-tively. These results indicate that additional human contact during rearing and the proximity of human contact in adulthood affected the avoidance behaviour of adult hens to humans: additional human contact during rearing and close human contact in adulthood resulted in less avoidance of humans, perhaps owing to habituation of fear responses. The nature of visual human contact in adulthood had less infl uence on fear of humans.

Barnett et al. (1994) also demonstrated the infl uential effects of visual con-tact with humans imposed during adulthood on fear responses of laying hens to humans. Regular visual contact, involving positive elements such as slow and deliberate movements, reduced the subsequent avoidance behaviour of adult birds in comparison with minimal human contact that at times contained ele-ments of sudden, unexpected human contact (see Table 3.2). As the unexpected appearance of a stimulus is fear provoking for most animals (McFarland, 1981), the birds that received minimal human contact in the study by Barnett et al.(1994) may have sensitized rather than habituated to the repeated but unex-pected exposure to humans, with fear levels therefore increasing rather than declining. The birds that had received regular visual contact with humans also had lower plasma corticosterone concentrations following handling than birds that had received only minimal human contact. The observations that laying hens housed in cages in the top tier of a multi-tiered battery and in outside rows with narrow corridors are more fearful of humans than those in lower tiers and in inner rows have at least been partly interpreted in terms of both less contact with stock-people during routine husbandry and often unexpected close contact when it occurs (Barnett and Hemsworth, 1989; Hemsworth and Barnett, 1989).

Hemsworth et al. (1994c) examined the effects of regular close visual con-tact with humans on the fear responses of young chickens to humans. At 6 weeks of age, birds that had received regular human contact showed less avoidance of an approaching experimenter and had lower plasma corticosterone concentra-tions after handling than birds that had received minimal human contact.

There is some limited evidence that pigs may also be affected by visual contact with humans. Gonyou et al. (1986) found that pigs that were regularly exposed to an experimenter rapidly and closely approaching them showed similar avoidance

Table 3.2. Effects of handling on the behavioural responses, stress physiology and productivity of laying hens (from Barnett et al., 1994).

Handling treatments

VariableMinimal

human contactAdditional

human contact

Forward score to humana 1.22 2.12Corticosterone concentrations in

response to handling (nmol/l)11.7 7.9

Eggs produced per day per hen (%) 83.1 89.4

aA high score refl ects less avoidance of the experimenter.

Human–Animal Interactions 73

of the stationary experimenter to that of pigs that regularly received negative tactile interactions, such as a shock from a battery-operated prodder, whenever they failed to avoid the approaching experimenter. Regular exposure of young pigs to humans standing erect or approaching them results in less approach to a stationary unfamil-iar experimenter than regular exposure to humans squatting or avoiding them (Hemsworth et al., 1986a).

Thus, there is evidence that poultry and limited evidence that pigs are sensitive to visual contact with humans. Indeed, positive visual contact may be more effective in reducing levels of fear for humans in poultry than human tactile contact. Rela-tively little is known of the visual contact with humans that may elevate fear levels in farm animals; however, rapid speed of movement by humans and sudden and unexpected exposure to humans may be fear provoking, particularly for poultry.

3.5.3 Auditory and olfactory contact with humans

Few studies have examined olfactory and auditory contact with humans on farm animals. Waynert et al. (1999) studied the effects of noise on cattle and found that both heart rate and movement were greater when animals were exposed to an audio recording of humans shouting than to a recording of the clanging noise of metal-on-metal. Pajor et al. (2000), using aversion learning techniques in which animals have the opportunity to learn to associate a location with a specifi c treat-ment, found that dairy cows showed a similar aversion to being shocked with a battery-operated prodder as to shouting by humans, based on both the time and effort required to move the animals to the location in which they were treated. In contrast, cows showed greater aversion to being hit and shouted at than to brushing or feeding, based on the effort required to move the animals to the location in which they were treated.

While there were no differences in the approach behaviour of young pigs to humans using a loud harsh voice or soft quiet voice, young pigs showed less approach to a human wearing gloves than to an ungloved human (Hemsworth et al., 1986a). It is possible that human odours, particularly on the hands, are used by pigs in recognition of humans and that masking these odours may create uncertainty or novelty for the animals.

Care is required in interpreting studies in which individual cues have been studied, because animals experienced with humans may learn through condi-tioning to associate insignifi cant cues from humans with those that have signifi -cance for the animals. For example, auditory cues from humans may be associated with negative tactile interactions by humans, as non-tactile contact with humans may accompany negative tactile interactions by humans.

In conclusion, tactile and visual contact with humans are important deter-minants of the fear responses of many farm animals to humans. Research has shown that human interactions, such as slaps, hits and shocks with a battery-operated prodder, are likely to increase fear responses in cattle, pigs and sheep, while contact such as talking, slow speed of movement and stroking are likely to reduce fear in farm animals. Furthermore, limited interaction appears to be infl u-ential in determining fear responses to humans, indicating the sensitivity of

74 Chapter 3

farm animals to brief human contact, which at times may appear to be innocu-ous and inoffensive. Although farm animals may be more sensitive to human contact early in life, particularly if weaned early, subsequent handling is also infl uential and has the potential to modify such early learning effects.

3.6 Effects of Fear of Humans on the Productivity of Farm Animals

Some of the handling studies reviewed in Section 3.5 also examined handling effects on animal productivity and most of these studies indicate that productivity is reduced in situations in which handling increases the animal’s fear of humans. The results of these handling studies, together with some observations on fear–productivity relationships in the livestock industries, are reviewed in this section.

3.6.1 Pigs

Negative tactile interactions imposed briefl y but regularly on pigs not only result in high levels of fear of humans (see Section 3.5.1), but may also markedly reduce growth and reproductive performance in pigs. A summary of some of the results of handling studies at our laboratory are presented in Table 3.3.

Seabrook and Bartle (1992) also reported growth depressions in pigs follow-ing negative handling, while Paterson and Pearce (1989, 1992) and Pearce et al.(1989) found no effects of regular negative handling on growth performance. There are also reports of fear for humans affecting reproduction: negative handling reduced pregnancy rate but not sexual receptivity in gilts (Hemsworth et al., 1986c), fear of humans in oestrous sows reduced their time spent near boars when in the presence of humans (Pedersen et al., 2003) and fear of humans in sows has been reported by Hemsworth et al. (1999), but not by Andersen et al.(2006), to be positively associated with percentage of stillborn piglets, and by Lensink et al. (2009) to be positively associated with crushing of piglets within 24 h of parturition.

Furthermore, observations in the Dutch and Australian pig industries have revealed signifi cant relationships, based on farm averages, between fear of humans and the reproductive performance of pigs (Hemsworth et al., 1981b, 1989b). The direction of the relationships indicates that reproductive perfor-mance was low at farms where breeding females were highly fearful of humans and the magnitude of these relationships indicate that variation in fear of humans, accounted for about 20% of the variation in reproductive performance across the study farms. The magnitude of the associations between fear and productiv-ity were remarkably similar in these two on-farm studies (e.g. the correlation coeffi cients, which estimate the degree of association, between time to approach within 0.5 m of the experimenter and farrowing rate were −0.55 and −0.54, P < 0.05; Hemsworth et al., 1981b, 1989b). In a second study, as distinct from the fi rst, farms varied substantially in terms of size, housing systems, genetics, nutrition and locality, and yet signifi cant fear–productivity relationships were

Human–Animal Interactions 75

found, which demonstrate the robustness of the fear–productivity relationship in the industry.

As mentioned earlier, several studies have shown no effects of regular nega-tive handling on the growth performance and stress physiology of young pigs (Paterson and Pearce, 1989, 1992; Pearce et al., 1989). There is no obvious explanation for this lack of effects in the studies by Paterson and colleagues; however, differences between studies in the nature, amount and imposition of handling treatments, and in opportunity for the animals to approach or with-draw, may be responsible for these apparently contradictory results. For exam-ple, a behavioural response of animals to an apparently aversive stimulus (e.g. withdrawal to negative handling by humans) in some situations may be an effec-tive strategy to enable the animals to cope with this stimulus without having to resort to any long-term physiological adjustment. There may also be genetic dif-ferences between pigs in their ability to cope with chronic stressors, although

Table 3.3. The results of handling studies on fear of humans, stress physiology and productivity of pigs.

Handling treatment

Experiment and variable Positive Minimalc Negative

Hemsworth et al. (1981a)Time to interact with human (s)a

Growth rate (11–22 weeks, g/day)Cortisol concentrations (ng/ml)b

119709

2.1

–––

157669

3.1Gonyou et al. (1986)Time to interact with human (s)a

Growth rate (8–18 weeks, g/day)73

89781

881147837

Hemsworth et al. (1986c)Time to interact with human (s)a

Pregnancy rate of gilts (%)Cortisol concentrations (ng/ml)b

4888

1.7

9657

1.8

12033

2.4Hemsworth et al. (1987)Time to interact with human (s)a

Growth rate (7–13 weeks, g/day)Cortisol concentrations (ng/ml)b

10455

1.6

92458

1.7

160404

2.5Hemsworth and Barnett (1991)Time to interact with human (s)a

Growth rate (from 15 kg for 10 weeks, g/day)Cortisol concentrations (ng/ml)b

55656

1.5

–––

165641

1.1Hemsworth et al. (1996b)Time to interact with human (s)a

Growth rate (from 63 kg for 4 weeks, kg/day)Adrenal weights (g)

520.973.82

791.054.03

1450.944.81

aFear assessed on the basis of the approach behaviour of pigs to a stationary experimenter.bBlood samples remotely collected at hourly intervals from 08.00 to 17.00 h.cTreatment involving minimal human contact.

76 Chapter 3

there is little evidence of this in the literature. Barnett et al. (1988) found that although there were differences between two genotypes of pigs in their basal plasma cortisol concentrations, both genotypes exhibited similar hypothalamic–pituitary–adrenal (HPA) axis responses (in terms of percentage increases in cor-tisol in the long term) to restraint on tethers.

The magnitude of the fear–productivity relationships observed in the pig industry demonstrates that fear of humans should be considered as a major factor associated with reduced productivity of commercial pigs. The mechanism respon-sible for the adverse effects of high fear on productivity appears to be a chronic stress response, because, in a number of experiments on pigs, handling treatments that resulted in high fear levels also produced either a sustained elevation in the basal concentrations of the stress hormone cortisol or enlargement of the adrenal glands, together with depressions in growth and reproductive performance (Barnett et al., 1983; Table 3.3). There is considerable evidence in the literature that stress hormones may adversely affect growth and reproductive performance by disrupting protein metabolism and key reproductive endocrine events. As dis-cussed in Chapter 2 (Section 2.5.1.1), the growth axis is inhibited at several levels during stress (Kaltas and Chrousos, 2007). Prolonged activation of the HPA axis leads to suppression of growth hormone (GH) secretion, while cortico-steroids can induce resistance in target tissues to the effects of GH, insulin-like growth hormone factor 1 (IGF-1) and other growth factors, and consequently the suppression of growth. The catabolic effects of ACTH and corticosteroids are also well known (Elsasser et al., 2000). Corticosteroids also support the synthesis and action of adrenaline in stimulating glycogenolysis and lipolysis (Matteri et al.,2000). Stress-induced changes in the secretion of pituitary hormones have been implicated in failed reproduction (Clarke et al., 1992; Moberg, 2000).

3.6.2 Poultry

Handling studies on poultry generally indicate that handling treatments likely to increase the birds’ fear of humans may depress the growth performance and immune function in chickens. For example, in experiments with young chickens, Gross and Siegel (1979, 1980) found that frequent but brief human contact of an apparent positive nature, such as gentle touching, talking and offering food on the hand, from an early age improved growth rates, feed conversion, anti-body response to an antigen and resistance to M. gallisepticum in comparison with minimal human contact. Although the behavioural response of the chickens to humans was not quantifi ed, the authors stated that the handled birds were easier to handle during weighing and blood sampling. Furthermore, under con-ditions of water deprivation, there was better feed conversion in the chickens that had received brief positive human contact (Gross and Siegel, 1980, 1982) and, while weight loss after fasting was not affected by handling, chickens that had received brief positive human contact were more resistant to Staphylococcusaureus (Gross and Siegel, 1982). Gross and Siegel (1981) found that chickens that received regular positive human contact from an early age had improved feed conversion and were more resistant to E. coli infection than chickens that

Human–Animal Interactions 77

either received minimal human contact or had been regularly scared by shouting and banging on their cages. More recently, Zulkifl i et al. (2002) and Zulkifl i and Siti Nor Azah (2004) examined the effects of handling on the physiology and productivity of meat chickens. Birds that were regularly caught and stroked early in life had improved growth rates and feed conversion compared with those that received minimal human contact (Zulkifl i and Siti Nor Azah, 2004). These han-dled birds, as well as other birds that either were stroked but also observed others being stroked, were caught and suspended by the legs or were caught and sus-pended by the legs but observed others handled in a similar manner, had lower heterophil to lymphocyte ratios than non-handled birds. This change in the dis-tribution of white blood cells may be indicative of reduced stress because, as discussed in earlier (Section 2.5.1.1), stressors can impair immune function, with evidence that increasing corticosteroid concentrations results in a redistribution of white blood cells (e.g. Gross and Siegel, 1983; Kehrli et al., 1999). Further-more, Zulkifl i et al. (2002) found that while there were no effects on body weight, feed conversion and mortality, regular visual contact with humans early in life increased the antibody response to Newcastle disease vaccine and reduced heterophil to lymphocyte ratio following capture and restraint in a crate.

Other studies have also shown that handling, mainly involving capture and brief restraint, is associated with increased growth performance of chickens (Thompson, 1976; Jones and Hughes, 1981). In contrast, Reichmann et al.(1978) found no effects of similar handling on the growth performance of either young meat or layer chickens, whereas Freeman and Manning (1979) suggested that regular handling decreased growth performance in layer chickens. Because handling may vary from positive to negative in nature, variation in the nature of handling between these studies, through effects on fear and stress, may have been responsible for the variation in the effects of handling on growth perfor-mance. For example, Buckland et al. (1974) demonstrated negative effects of handling involving blood sampling by cardiac puncture on the growth perfor-mance of meat chickens.

A handling study on adult poultry at our laboratory also indicates that high fear levels will limit the productivity of poultry. Barnett et al. (1994) found that regular visual contact, involving positive elements such as slow and deliberate movements, that reduced the subsequent avoidance behaviour of adult laying hens to the experimenter, resulted in higher egg production than a treatment which involved minimal human contact that at times contained elements of sud-den, unexpected human contact (Table 3.2). The authors speculated that the lower productivity of birds in the latter treatment may be a consequence of a chronic stress response because there was evidence of immunosuppression in these fearful birds: the cell-mediated immune response to a mitogen was lower in laying hens that received reduced and unexpected human contact (Barnett et al., 1994). Edwards (2009) also recently found that regular exposure to an experimenter moving slowly and predictably increased egg production in hens in comparison with regular exposure to an experimenter using startling and unpre-dictable behaviour.

Studies conducted on meat chickens and laying hens in the fi eld also sup-port the proposition that high levels of fear of humans may limit the productivity

78 Chapter 3

of commercial birds. Signifi cant negative relationships, based on farm averages, have been found between the level of fear of humans and the productivity of commercial meat chickens. In a study involving 22 commercial farms, the level of fear of humans was signifi cantly and negatively related to the effi ciency of feed conversion of meat chickens at the farm (Hemsworth et al., 1994c). For exam-ple, avoidance by meat chickens of an approaching experimenter accounted for 29% of the variation between farms in feed conversion (Fig. 3.7). Similarly, Hemsworth et al. (1996a) found that avoidance of an approaching experimenter was negatively correlated with the effi ciency of feed conversion of meat chickens at 22 commercial farms (correlation coeffi cient of −0.49, P < 0.05). Avoidance of an approaching experimenter was also negatively correlated with growth rate of meat chickens in this study (correlation coeffi cient of −0.55, P < 0.01). In contrast, Cransberg et al. (2000) found no evidence of a signifi cant relationship between feed conversion and level of fear of humans at 24 commercial meat chicken farms; however, there was a signifi cant positive relationship between the level of fear of humans and mortality early in life. Farms in these three studies were under contract to the same processor, and Cransberg et al. (2000) sug-gested that differences in the company’s payment structure at the time of the most recent study may have put more emphasis on mortality at the expense of food conversion.

Signifi cant negative relationships, based on farm averages, have also been found between the level of fear of humans and the productivity of commercial laying hens. The egg production of laying hens at 14 commercial farms was negatively related to the level of fear of humans by birds at the farms (Barnett

80

70

60

50

40

30

20

10

0

–101.75 1.85

Ave

rage

num

ber

of b

irds

rem

aini

ng n

ear

expe

rimen

ter/

obse

rvat

iona

l sca

n

1.95

Feed to gain ratio (kg/kg)

2.05 2.15

Fig. 3.7. The relationship, based on farm averages, between fear of humans and feed conversion at 22 commercial meat chicken farms (Hemsworth et al., 1994c).

Human–Animal Interactions 79

et al., 1992). In an experiment examining the effects of cage position on fear and egg production of laying hens, level of fear of humans was also signifi cantly and negatively related to egg production and effi ciency of feed conversion (Hems-worth and Barnett, 1989). In observations on the behavioural response of laying hens to an experimenter, Bredbacka (1988) reported that egg mass production was lower in hens that showed increased avoidance of humans.

Thus, there is evidence that high fear levels may reduce the productivity of poultry. The mechanisms responsible are unclear. As seen in fearful pigs, a chronic stress response or a series of acute stress responses in the presence of humans may be responsible for the depressed productivity in fearful poultry. Support for this suggestion is provided by the known effects of corticosteroids on nitrogen balance, protein catabolism and energy retention or excretion in chick-ens (Siegel and van Kampen, 1984). Further, exogenous elevations of circulating corticosterone concentrations have been shown to adversely affect growth rate and feed conversion in chickens (Bellamy and Leonard, 1965; Adams, 1968; Bartov et al., 1980; Siegel and van Kampen, 1984; Saadoun et al., 1987). Therefore, regular exposure to stressful stimuli, with the consequent elevations of plasma corticosterone concentrations, might also be expected to impair poultry productivity.

3.6.3 Dairy cattle

Most of the cattle handling studies that have studied effects on productivity have been on dairy cattle. As with observations in the pig and poultry industries, there is evidence of negative fear–productivity relationships in the dairy industry. In a study examining inter-farm correlations between fear of humans, assessed on the basis of the approach behaviour of cows to a stationary experimenter in an unfa-miliar arena, and cow productivity at 31 dairy farms, Breuer et al. (2000) found that fear of humans was negatively correlated with the average milk yield of cows at the farm. While Hemsworth et al. (2000) found no signifi cant inter-farm cor-relations between fear of humans, assessed on the basis of both the approach behaviour of cows to a stationary experimenter and the avoidance behaviour of cows to an approaching experimenter, and milk yield at 66 dairy farms, the authors found that the time cows spent near the stationary experimenter was negatively correlated with milk cortisol concentrations at the farm. Furthermore, the frequency of negative interactions used by stockpeople, such as slaps, pushes and hits, was negatively correlated with average milk yield and conception rate to fi rst insemination and positively correlated with milk cortisol concentrations at the farm. Waiblinger et al. (2002) found that the frequency of negative behav-iours by stockpeople was negatively correlated with average milk yield. Seabrook (1972a) reported that cows in high-yielding herds in Britain tended to be the most willing to approach the milker, to return from pasture and to enter the milking par-lour. These fi ndings in commercial settings, therefore, also suggest that milk yield may be at risk when cows are fearful of humans.

There is some evidence from handling studies that negative handling may depress milk yield in cows. Rushen et al. (1999) found that dairy cows had

80 Chapter 3

increased residual milk when milked in the presence of an experimenter who had hit or shocked them with a battery-operated cattle prodder than in the presence of an experimenter who had brushed, fed and gently spoken to them. The former cows also had higher heart rates at milking, implicating the secretion of catechol-amines under the infl uence of the autonomic nervous system as affecting milk let-down in these fearful cows.

Moderate or forceful slaps imposed briefl y before or after milking when ani-mals failed to avoid humans increased fl ight distance and tended to reduce milk yield in heifers (6% reduction in yield over 8 weeks; Breuer et al., 1997). As reported earlier (Section 3.5.1), dairy heifers that were slapped or hit whenever they approached or failed to avoid the experimenter were more fearful of the experimenter based on their approach behaviour than those that were positively handled with slow and deliberate movements and pats and strokes (Breuer et al.,2003). Furthermore, while these fearful heifers showed higher plasma cortisol concentrations in the presence of the experimenter, they also had higher cortisol concentrations in the afternoon in isolation of humans, suggesting that they may have been chronically stressed by the negative handling treatment. Stressors that result in an acute stress response may depress milk yield owing to inhibition of milk let-down (Bruckmaier et al., 1993, 1997; Bruckmaier and Blum, 1998). The long-term stress response of cows and how these responses affect milk yield are poorly understood. One key function of the stress response is to divert food and substrates, such as acetates, glucose and amino acids, away from normal day-to-day functions such as growth and reproduction (Sapolsky, 1992), and so during a chronic stress response, the substrates may be diverted elsewhere, thereby interfering with milk synthesis (Breuer et al., 2003).

It is also of interest that Arave et al. (1985) found that dairy calves reared in visual and tactile isolation from other calves produced more milk in adulthood than herd mates reared either in groups or individually but with visual and tactile contact with calves. In contrast, no such effects were found in a more recent study (Arave et al., 1992). The authors of the earlier study proposed that human-reared calves may have ‘imprinted’ upon the stockperson and thus may have adapted more easily to the milking procedure, which involves intense human contact. Creel and Albright (1988) rejected this hypothesis on the basis of the similar approach behaviour of isolated and control calves to a stationary experi-menter. However, they also found that the isolated calves had a shorter fl ight distance to an experimenter than control calves. Dam-reared goats, which showed increased avoidance of humans, were found to have impaired milk let-down in comparison with human-reared goats (Lyons, 1989).

Lensink et al. (2000a) studied human–animal interactions at 50 veal calf units. Average daily weight gain of the calves was predicted by the stockperson behaviour towards calves, with higher weight gain associated with more positive behaviour, such as touching, patting, talking gently and allowing calves to suck fi ngers. Feed conversion was also predicted by the stockperson’s behaviour, with improved feed conversion associated with more positive behaviour. It is also of interest that several variables, such as positive behaviour to calves and farm size, were predictive of calf mortality: mortality was lower at farms that were large and in which the stockperson displayed more positive behaviour to calves.

Human–Animal Interactions 81

In contrast, Lensink et al. (2000c) found that calves which received positive handling during rearing had similar growth rates to those that received minimal human contact.

In conclusion, there is evidence from observations in the dairy industry and handling studies on dairy calves and cows that fear of humans may limit the pro-ductivity of commercial dairy calves and cows. These results support those of sim-ilar but more extensive studies on pigs and poultry. Further research is required to examine the effects of high fear levels on the productivity of dairy cattle.

3.6.4 Other farm animals

There is very limited evidence that high fear levels may reduce the productivity of other farm animals. This evidence applies generally to situations in which animals have intense or frequent contact with humans and, therefore, in which there are opportunities for fear responses to impair animal productivity and welfare. In more extensive management settings, chronic effects of negative human contact on the productivity of livestock may be less likely because of less contact with humans and the greater opportunity for the animals to control their contact with humans.

As referred to in the previous section, dam-reared goats, which showed increased avoidance of humans and lower cortisol responses in the presence of humans, exhibited greater milk ejection impairment than human-reared goats (Lyons, 1989). The dam-reared goats received considerably less human contact than the human-reared animals.

There is evidence that beef cattle which are the most diffi cult to handle may suffer depressions in their productivity and meat quality at slaughter. Fordyce et al. (1988) found that beef cattle that were the most active and vocal when restrained in a weighing stall had the most carcass bruising and tended to have tougher meat following slaughter. Although part of the behavioural responses of cattle when restrained in a weighing stall would be responses to restraint and novelty, a component of these responses would be specifi cally to humans. In studying a similar behavioural response to restraint, Burrow and Dillion (1997) found that the exit speed of beef cattle, called ‘fl ight speed’, was negatively correlated with weight gain. Although these extensively grazed cattle would have received limited human contact, this contact would have generally been associ-ated with aversive experiences such as restraint, castration and branding, and thus these observations may refl ect general fearfulness and/or fear of humans. However, Petherick et al. (2009b) found that while behavioural measures of fear for humans were not correlated with average daily gain in beef cattle housed in feedlots, fl ight speed was moderately and negatively correlated with average daily gain. As discussed earlier (Section 3.3), fl ight speed in response to this brief restraint is considered a measure of temperament (Petherick et al., 2002) and so this relationship with growth suggests that feedlot cattle which are more stressed by environmental change or uncertainty may suffer depressions in productivity. In contrast, while positive handling of feedlot cattle involving feeding and talk-ing reduced fear of humans, based on approach behaviour to a stationary

82 Chapter 3

experimenter outside the pen, positive handling did not affect fl ight speed (Peth-erick et al., 2009a). Thus, in more extensive management settings, chronic effects of negative human contact on the productivity of livestock may be less likely because of less contact with humans and the greater opportunity for the animals to control their contact with humans. Nevertheless, as discussed later (Section 5.6.3), animals that are fearful of humans are likely to experience both acute stress in the close presence of humans and handling diffi culties. So while stimu-lus-specifi c responses to humans will develop as a consequence of the history of interactions with humans, behavioural and physiological responses to humans will also vary across species and across strains within species, depending on the heritable components of temperament within species.

3.7 Conclusions: Fear, Productivity and Welfare

In intensive livestock production there is frequent and often close contact between stockpeople and animals, particularly young animals and breeding animals, and as a consequence of these interactions, many of which are far from superfi cial, long-term relationships develop between humans and animals. There is evi-dence from studies in the industry and in the laboratory that these relationships may exert substantial effects on the behaviour, physiology and productivity of commercial livestock. Most of the evidence in the literature is from studies on dairy cattle, pigs and poultry; however there is limited evidence on other species such as sheep and goats. It is therefore proposed that the human–animal rela-tionship may have practical implications for farm animals in production systems in which there are close or frequent human–animal interactions. In more exten-sive management settings, chronic effects of negative human contact on the behaviour, physiology and productivity of livestock may be less likely because of less contact with humans. Nevertheless, acute effects are likely in situations in which fearful animals are in close contact with humans.

It is also important to recognize the implications of fear for humans in the welfare of farm animals. Fear is generally considered an undesirable emotional state of suffering in both humans and animals (Jones and Waddington, 1992) and, indeed, one of the key recommendations proposed to the UK Parliament by the Brambell Committee in 1965 (Brambell et al., 1965) was that intensive-housed livestock should be free from fear. Research reviewed in this chapter has shown that farm animals that are both highly fearful of humans and in regular contact with humans may experience not only an acute stress response in the presence of humans but also a chronic stress response that is evident even in the absence of humans. For example, handling treatments that resulted in high fear levels in pigs resulted in not only an acute stress response in the presence of humans but also a chronic stress response measured on the basis of a sustained elevation in the basal concentrations of cortisol in isolation from humans (Table 3.3). Thus, in addition to the concern about animals experiencing an undesirable emotion state such as fear, it is also ethically unacceptable to have animals that are chronically stressed. Furthermore, fearful animals are more likely to sustain injuries trying to avoid humans during routine inspections and handling. Chronic

Human–Animal Interactions 83

stress may also result in immunosuppression which, in turn, may have serious consequences on the health of the animals. Clearly, fear in farm animals can have an impact on farm animal welfare and so this topic of fear in farm animals is a legitimate welfare consideration. The magnitude of the effects of fear for humans observed on the productivity of livestock in both commercial and labo-ratory settings indicates the potential for improving the productivity and the wel-fare of farm animals by identifying and manipulating those human factors that are infl uential determinants of animal fear in commercial settings. The following chapters explore these opportunities.

© CAB International 2011. Human–Livestock Interactions, Second Edition 84 (P.H. Hemsworth and G.J. Coleman)

4 Attitudes of Stockpeople

4.1 Introduction

In Chapter 3, the nature and frequency of human–animal interactions in live-stock production were discussed in detail and it was shown that these interac-tions, which can result in the formation of long-term relationships between humans and animals, have substantial effects on the behaviour, physiology and productivity of farm animals. The existence of these infl uential human–animal relationships indicates the potential for improving the productivity and welfare of farm animals by identifying and manipulating the key human factors that regulate these relationships. This rests on the human–animal relationships being causally related to the behaviour of humans and this behaviour, in turn, being causally related to some characteristics of humans that are susceptible to change.

The handling studies reported in the previous chapter have shown that the behaviour of humans infl uences the stress physiology and productivity of farm animals and thus the behaviour of stockpeople is an important factor in the effects of human–animal interactions on farm animal productivity and welfare. The aim in this chapter is to discuss the human characteristics that underlie stockperson behaviour. This review will lead to a discussion of a theoretical framework to be used to consider the development and maintenance of atti-tudes and how attitudes can be used to predict human behaviour. This discus-sion, in turn, will provide a sound basis for reviewing stockperson behaviours in Chapter 5: how they develop and their role in regulating farm animal behav-iour. Furthermore, the concepts and theoretical framework considered in both Chapters 4 and 5 will be utilized in Chapter 6, in which a model of stockperson–animal interactions, based on research in the livestock industries, will be described.

Attitudes of Stockpeople 85

4.2 Stockperson Attitudes and Animal Welfare and Productivity

The psychology literature reveals that the important dispositional factor in predict-ing human behaviour is attitude. Attitudes are learned and therefore able to be changed. In order to improve animal performance and welfare by manipulating stockperson behaviour in livestock production, a thorough understanding of the development of attitudes and their relationship with behaviour is obviously required.

4.2.1 What are attitudes?

The term attitude is widely used in everyday conversation and the media, and we all think we understand what it means. Typically, we think of attitudes as opin-ions, that is, what we think about other people, things and events. In fact, opinions are verbal expressions of attitudes and our describing of attitudes as opinions does not really provide any insight into the nature of attitudes, how they are formed and how they may infl uence behaviour. In order to understand the nature of attitudes and their relevance to the actions of stockpeople in the work-place, it is important to look more closely at the various meanings of the term.

Eagly and Chaiken (1993) defi ned attitude as ‘a psychological tendency that is expressed by evaluating a particular entity with some degree of favour or dis-favour’. There are three key features to this defi nition: (i) the idea that attitudes are directed at an entity or thing; (ii) the idea that attitude is a tendency or dispo-sition; and (iii) the idea that attitude expresses some positive or negative evalua-tion. It is important to realize that attitudes cannot be observed directly; we infer peoples’ attitudes from what they say and do.

Characterizing an attitude object is not simple. We may have attitudes towards people, animals, inanimate objects or even ideas. For example, we may have an attitude towards a political ideology, a religion, an individual, a race, a species of animal or a particular animal. As we shall see later in this chapter, to get some insight into expected behaviour it is very important to be quite specifi c about the particular attitude object of interest.

The notion of attitude as a tendency refl ects the view held by psychologists that our attitudes tend to direct our behaviour or, at least, our intended behav-iour (e.g. Ajzen and Fishbein, 1980). For example, if we believe that a particular brand of motor vehicle has outstanding performance and appearance and we like it more than any other brand, we are likely to have a positive attitude towards that car and, when in the market to buy a car, we would tend to buy that par-ticular brand. However, it is easy to see that there may be a whole range of issues, apart from attitude, that we would need to consider before buying the car. These include whether we can afford the car, whether we can sell our current car, what family and friends think of it and what cars friends and neighbours drive. The role of some of these kinds of factors, apart from attitude, in determining stockperson behaviour will be discussed in Chapter 6.

The evaluative nature of attitudes is what distinguishes them from other kinds of verbal expression. A statement that either explicitly or implicitly characterizes

86 Chapter 4

something as good or bad, liked or disliked or even something to be enjoyed or not enjoyed, expresses an evaluation of that object and, therefore, refl ects the underlying attitude. Attitudes are, therefore, favourable or unfavourable: they refl ect a tendency for or against, like or dislike, etc.

Historically, psychologists have defi ned three components to attitude: cogni-tion, affect and conation (Allport, 1935). Cognition refers to the thoughts that people have about some object. In other words, cognitions are beliefs or subjec-tive facts. They are things that people believe to be true about a person or object. Albarracín et al. (2005) make the point that affect, beliefs and behavioural ten-dencies are expressions of an underlying attitude rather than components of an attitude. In other words, an attitude is a person’s current judgement about a par-ticular object, which is, in turn, a product of an underlying judgement modifi ed by relevant available information. A stockperson may believe that pigs are very diffi cult to handle and require a lot of effort. This may refl ect an underlying negative attitude towards working with pigs. Affect refers to the emotional response that a person has towards some other person or object. The extent to which we like or dislike an object is an example of affective response. A stockper-son may express a dislike of pigs or may fi nd them dirty, greedy or smelly. Such expressions would be affective statements refl ecting an underlying negative atti-tude. Finally, conation refers to a tendency to behave in a particular way. A stockperson’s intention to avoid contact with pigs or to fi nish work in the piggery as quickly as possible are examples of conation and may also refl ect an underly-ing negative attitude towards pigs.

There has been much discussion about the three-component concept of atti-tude. In general, Eagly and Chaiken (1993) seem to favour the view that these three components independently contribute to a person’s attitude towards an object. Albarracín and colleagues’ (Albarracín et al., 2005) view that the three components are expressions of attitude rather than components of attitude is probably the most reasonable way to conceptualize attitudes in that the three components are outcomes rather than the underlying disposition as such. These three components are all correlated with each other and all contribute to an understanding of the underlying evaluative attitude dimension. This means that measuring any one of the components will provide some indication of a person’s attitude.

4.2.2 Measuring attitudes

Attitudes cannot be measured directly but a person’s responses to a series of attitude statements in a questionnaire can be used to infer an underlying attitude. These statements are usually designed to measure one or more of the three aspects of the expression of an attitude: the person’s beliefs about the object (cognition), emotional response to the object (affect) and behavioural tendency towards the object (conation). Thus, when we assess a person’s attitudes for the purpose of predicting subsequent behaviour towards the attitude object, we are actually using a series of verbal responses to predict this behaviour by the person.

Attitudes of Stockpeople 87

4.2.3 Functions of attitudes

Katz (1960) proposed that attitudes serve four functions. First, attitudes may meet the individual’s need to organize experience. Thus, a person may express a positive attitude towards a job if he or she has observed others enjoying similar work and if it is felt that it has status and security. A positive attitude to work is a consistent way of refl ecting these various aspects of the job. Secondly, attitudes may help the person to maximize positive experiences and to minimize negative experiences. They are a refl ection of the learning processes a person undergoes in the job. Thus, negative aspects of a job, such as a history of adverse workplace experiences, poor pay and little opportunity for advancement, may lead to a poor attitude to work. As a consequence, a person will minimize emotional and, perhaps, physical investment in the job. Thirdly, attitudes may serve to protect a person from negative events. The expression of authoritarianism in ethnocentric attitudes (that is, attitudes that express a negative view of other races) is an example of this. An authoritarian person protects him- or herself from the pos-sible threat from an out-group (a group other than one’s own social, ethnic or work group) by assuming a superior stance. Finally, attitudes are an expression of the personality of the individual. Thus, a person’s attitudes are consistent with self-image. In livestock production, a person who takes pride in being a good stockperson may hold positive attitudes towards husbandry, acquisition of knowledge and the work ethic. The identifi cation of these functions of attitudes refl ects the belief among psychologists that attitudes have a motivating infl uence on our behaviour.

In general, although there have been various elaborations of these func-tions of attitudes, Katz’s basic functions of attitudes remain current (Eagly and Chaiken, 1993).

4.2.4 Development of attitudes

Attitudes are generally regarded as learned dispositions of a person. They develop initially as part of the socialization of the individual. Young children are dependent primarily upon their parents for information about how they should behave towards others. As a result, children’s attitudes come initially from their parents and other close family. Once the child reaches school age, there is an increasing infl uence of teachers, school friends and television. Stagner (1961) reports a study by Blake and Dennis (1943) who tracked the development of white children’s attitudes towards American blacks. Initially, very young children showed no particular attitude towards blacks. As the children developed, they formed more and more negative stereotypes, which were like those of their par-ents. One important aspect of the development of these attitudes is that they need not be based on direct experience. Children learn to accept what parents say because parents control the rewards and punishment.

In school-age children and in adults, social pressure to conform in beliefs and behaviour also come from friends, school- or work-mates and from the media. In a classic study, Asch (1956) demonstrated that people would change

88 Chapter 4

their beliefs and their behaviour when pressured by others. How likely they were to change and the amount of this change depended on the size of the difference between the person’s beliefs and the others’ beliefs, and the extent to which the person was in a minority. This tendency to change is called conformity behaviour.

4.2.5 Attitude systems

A person’s attitude towards a specifi c subject or thing does not exist independ-ently of his or her other attitudes. If, for example, a person believes that excessive drinking is bad, that person will also have related attitudes towards the cost of alcohol, drinking at lunchtime and so on. In other words, there will be a system of attitudes which are more-or-less consistent with each other. Not only are these attitudes consistent, but the person will also tend to behave in a way consistent with those attitudes. For example, the person may prefer the lunchtime company of people who don’t drink so that he or she will not be under social pressure to drink. While the origin of this tendency for consistency is a matter for some argu-ment (Ajzen, 1988), there is widespread acceptance of it as a well-established characteristic of human behaviour.

One of the most widely accepted theories of attitude consistency is that of Leon Festinger. Festinger (1957) proposed his theory of cognitive dissonance to account for the way in which attitudes infl uence behaviour, and this theory was later extended by Brehm and Cohen (1962) and Aronson (1969). Dissonance theory proposes that cognitive elements (that is, beliefs or any pieces of knowl-edge) are dissonant if one element does not follow from the other and are con-sonant if one element does follow from the other. For example, the statement ‘I smoke cigarettes’ is dissonant with the belief ‘smoking is a health hazard’ but is consonant with the belief ‘smoking relaxes me’. The theory is very broad, a belief can be regarded as a cognitive element.

Cognitive dissonance theory has been the focus of a great deal of research. In particular, it has led to a focus on the ways in which people process new infor-mation in the context of their existing attitudes. It provides a framework within which the interrelationships between attitudes and behaviours can be under-stood. This is particularly important when we seek to understand attitude change and to predict behaviour from attitudes.

4.2.6 Attitudes and behaviour

Most people, and psychologists are no exception, attempt to explain a person’s behaviour in terms of dispositions. In other words, we believe that people tend to do those things that are consistent with their underlying characteristics. For example, an authoritarian person is likely to treat superiors with deference and unquestioning obedience, while treating inferiors with arrogance and detach-ment. As we have seen, attitudes represent a major class of dispositions. The question arises, therefore, how well do attitudes predict an individual’s behaviour?

Attitudes of Stockpeople 89

Despite an intensive research programme directed towards clarifying the rela-tionship between attitudes and behaviour, Hovland and his co-workers at Yale University were unable to demonstrate that attitudes were good predictors of behaviour (Hovland et al., 1953). This research involved th e assessment of pe o-ple’s attitudes t owards o bjects and the establi shment of the relationsh ip of these attitudes to specifi c be haviours. As we shall se e shortly, generic attitudes are not good predicto rs of behaviour.

The adver tising industry is based on the assump tion t hat ch anging peopl e’sbeliefs about a product wi ll infl uence them to buy t he product. This i s only true in a sense. If a new bee r is int roduced into t he m arket, it is very diffi cult to deter-mine wheth er any given individ ual will buy it. I t will depend on whether the person is already a beer drinker, how loyal he or she is to a current brand, whether or not the person is thinking of giving up drinking and so on. In fact, the advertising industry relies on a small percentage of consumers changing their attitudes and behaviour: it is an actuarial exercise rather than one of changing a particular individual’s attitudes.

A very interesting study that demonstrated the uncertain relationship between generic attitudes and behaviour was reported by La Piere (1934). In this study, all but one of the operators of 250 restaurants and hotels in the USA served a visiting Chinese couple when the couple visited unannounced. When La Piere sent questionnaires in the mail to the same proprietors, 92% indicated that they would not be willing to accept Chinese guests. The question that imme-diately arises is ‘Why the discrepancy?’. Here, the questionnaire answers were clearly at odds with the actual behaviour of the hotel proprietors. In fact it is pos-sible to think of many reasons. For instance, the proprietors may have wished to avoid a scene with other guests, or may have decided that this particular Chinese couple was respectable and not the kind of people they had in mind when answering the questionnaire.

This illustrates that it is very important to identify the attitude object pre-cisely if attitude is to be used to predict behaviour. In La Piere’s study, for example, the questionnaire would really have had to describe the particular Chinese couple in some detail, and also to have characterized the circum-stances of their visit, in order to get a reasonably accurate idea of how the proprietors would behave.

La Piere’s study also illustrates that a specifi c behavioural situation may not provide the ideal context in which to predict behaviour from attitudes. Campbell (1963) asserted that a single behavioural event may not be suffi cient to provide an accurate measure of behaviour and that patterns of behaviour should be measured. Fishbein and Ajzen (1974) reported that, when predicting a single behaviour from attitudes towards religion (e.g. regular church attendance), correlations of 0.12–0.15 were obtained, depending on the method used. When multiple behaviours were predicted from the same attitudes (e.g. fi nancial contri-bution to a church, owning a bible, practising Christian beliefs, etc.), the correla-tions increased to 0.61–0.71, depending on the method used. This has practical implications for studying stockperson behaviour because it suggests that a varietyof stockperson behaviours needs to be observed, not merely a single handling bout with an animal, for example.

90 Chapter 4

A major development in the conceptualization of the relationship between attitudes and behaviour came with Fishbein and Ajzen’s Theory of Reasoned Action (Ajzen and Fishbein, 1977, 1980). This theory was developed to deal with behaviours that were under the person’s control, in other words, volitional behaviours. The theory proposed that the three components of attitude dis-cussed earlier, affect, belief (cognition) and conation, can better be considered as three response tendencies, which represent a sequence in the development of behavioural outcomes. More specifi cally, the beliefs that people hold, when com-bined with their evaluations of those beliefs, lead to the formation of attitudes. Intentions and actions then follow from these attitudes. A comprehensive picture of a later revision of the model is presented in Fig. 4.1 (Ajzen and Fishbein, 1980). From Fig. 4.1, it can be seen that the immediate cause of a person’s behaviour is intention. So long as there are no impediments to intention being translated into behaviour, the theory is useful for predicting behaviour. In other words, if there are no physical constraints, such as inability to perform a behav-iour or lack of access to the behavioural situation, then a person is likely to do what he or she intends.

The immediate cause of intended behaviour is a person’s attitude towards the behaviour in combination with the person’s subjective norms with respect to the behaviour. A person’s subjective norms refer to the extent to which a person believes that relevant other people would approve of the behaviour and the extent to which the person feels willing to comply with other people’s expecta-tions. One important feature of this part of the theory is that the object of the attitude is not some general person (or animal) but a behaviour. The Theory of Reasoned Action relies on attitudes towards specifi c behaviours rather than

Demographic variablesPersonality traits

Attitudes towards targets

Beliefs that behaviour leads to outcomesEvaluation of outcomes

Attitude towards the behaviour

Intention

Behaviour

Fig. 4.1. A simplifi ed version of the Azjen and Fishbein (1980) model of the attitude–behaviour relationship.

Attitudes of Stockpeople 91

objects for prediction of acts. This is a major departure from the earlier approaches (for example the Yale studies by Hovland et al., 1953).

Attitudes are, in turn, determined by a combination of beliefs about the out-comes that are likely to occur following a particular behaviour and an evaluation of those outcomes. For example, if I believed that smoking leads to lung disease and I thought that lung disease was a particularly bad outcome, then I would have a negative attitude towards cigarette smoking. Similarly, if a child thought that his or her parents would not approve of smoking, and the child felt obliged to obey his or her parents, then the child would feel a strong subjective norma-tive pressure against smoking.

The antecedents of beliefs, evaluations and motivations are many and var-ied. As can be seen in Fig. 4.1, demographic variables, past experiences, various general attitudes and personality traits indirectly affect behaviour through their infl uence on beliefs, evaluations and motivations. It is important to recognize that the Theory of Reasoned Action proposes that the important dispositional factor in predicting behaviour is attitude and that other dispositional factors, including personality, operate indirectly through attitudes. These other factors will be considered in the next section.

The prediction of behaviour from attitudes based on the Theory of Rea-soned Action can be illustrated with an example from an agricultural industry. Lynne and Rola (1988) carried out a study to investigate the relationship between farmers’ attitudes towards soil conservation and their soil conservation behav-iours. Farmers were asked to respond to questions about their beliefs regarding soil conservation. For example, attitude questions used included ‘Farmers have a responsibility towards all those now living to use soil resources such as not to cause erosion’ and ‘Crops can be grown without soil, so erosion is irrelevant’. Because Lynne and Rola were interested in the relevance of economic factors in determining farmers’ conservation practices, they also asked questions about farmers’ beliefs with respect to economic factors in soil conservation. An exam-ple of such a belief statement was ‘A farmer must seek to maximize profi ts no matter what the costs in eroded soil or environmental damage’. Farmers responded to each statement using a 5-point evaluative scale ranging from strongly agree (5) to strongly disagree (1). Both negative and positive statements towards conservation activity were used. Behaviour was measured in terms of whether or not the farmer engaged in at least one conservation practice. In this example, the authors used farmers’ beliefs to measure their attitudes towards conservation behaviour. By using many such belief statements, the authors were relying on consistency theory to justify the measurement of an underlying atti-tude towards soil conservation with belief statements. As was discussed earlier, if a person holds a variety of beliefs that are consistent with each other, then it is possible to infer that person’s underlying attitude. Lynne and Rola (1988) ensured that, consistent with Fishbein and Ajzen’s model (Ajzen and Fishbein, 1977), attitude questions were directed towards conservation behaviour.

Attitude was found to predict conservation behaviour. For example, vari-ation in attitude accounted for about a third of the variation in conservation behaviour. It was also found that those farmers with higher incomes tended to have poorer attitudes towards conservation behaviour. This study is a clear

92 Chapter 4

example of how Fishbein and Ajzen’s model of the attitude–behaviour relation-ship can be applied in a practical situation. Results from studies in which live-stock are intensively handled will be discussed later in this chapter.

There has been substantial research to show that the Theory of Reasoned Action is an excellent predictor of behaviour (Eagly and Chaiken, 1993). However, one of the limitations of this theory is that intentions do not always translate into behaviour. This can arise because the beliefs associated with the relevant attitudes are unrealistic and do not accord with the actual behavioural situation. One way in which this can occur arises when the person feels that it is not possible to engage in the relevant behaviour. This may occur because the person is not physically capa-ble of performing the behaviour or because someone or something prevents the behaviour. An example of this that is relevant to livestock production arises when stockpeople at abattoirs think that they cannot engage in best practice in handling animals pre-slaughter because they believe that it is inconsistent with the demands of management that they keep up with the speed demand of the processing facility (Coleman et al., 2003). To address this, Ajzen and Fishbein (1980) introduced per-ceived behavioural control as a further factor in predicting behaviour (Fig. 4.2).

However, there have been several variants proposed for the theory. One important additional factor is habit (Bentler and Speckart, 1979). There is ample evidence to show that what people do is partly determined by past experience. This includes studies on voting behaviour (Echabe et al., 1988), seat-belt use (Budd et al., 1984) and blood donation (Bagozzi, 1980). Despite this and other evidence which shows that other factors, such as experience and opportunity, may infl uence behaviour (Eagly and Chaiken, 1993), the Theory of Reasoned Action and Ajzen’s (1985) development of this, the Theory of Planned Behav-iour, have provided a solid basis for predicting behaviour from attitudes.

Backgroundfactors

Individual

Social

Information

PersonalityMood, emotionIntelligenceValues, stereotypesGeneral attitudesExperience

Behaviouralbeliefs

Attitudetowards thebehaviour

Subjectivenorm

Intention Behaviour

Actualbehavioural

control

Perceivedbehavioural

control

Normativebeliefs

Controlbeliefs

EducationAge, genderIncomeReligionRace, ethnicityCulture

KnowledgeMediaIntervention

Fig. 4.2. A model of the theories of reasoned action and planned behaviour (adapted from Albarracín et al., 2005).

Attitudes of Stockpeople 93

4.3 Other Dispositional Factors

In Chapter 1, personality and empathy were two characteristics that were intro-duced as being relevant to work performance in general. The relevance of these for stockperson behaviour in caring for their animals has been discussed in some detail by Coleman (2004), and a brief review of the relevant data will be given here.

4.3.1 Personality

Underlying psychological characteristics of a person are referred to as personality traits. The idea of a personality trait is that it is a relatively enduring characteristic which exerts a general effect on that person’s behaviour and which we cannot observe directly, but can infer from the person’s behaviour. For example, the concept of introversion/extraversion is widely used in psychology. An extraverted person is outgoing, confi dent and talkative, while an introverted person is shy, timid and withdrawn. These underlying characteristics result in predictable kinds of behaviours. In social situations, the introvert will avoid being in the limelight, will feel uncomfortable and will express negative feelings, while the extravert will be just the opposite. In livestock production, it may be the case that there are characteristics, for example, degree of empathy or some temperament factors, which predispose people to be good stockpeople. These characteristics are usually assessed by affective statements by the person. Such statements, for example: ‘I feel uncomfortable when I see a distressed animal,’ are not easily distinguished from attitude statements, but they actually do differ in that they are self-directed, not directed towards some external person or object. There is little evidence relating personality directly to stockperson behaviour towards their animals.

As briefl y discussed earlier (Section 1.4.2), Seabrook (1972a,b) reported that the stockperson’s personality was related to the behaviour of the cows and milk yield of the herd. He found that high milk yield was associated with herds in which the stockpeople were introverted and confi dent, and where the cows were most willing to enter the milking facility and were less restless in the presence of the stockperson. In general, self-confi dence was associated with moderate- to-high milk yield regardless of degree of introversion. However, there were no data to show that the causal link between stockperson personality and cow behaviour and productivity was related to stockperson behaviour. In a more recent study by Waiblinger et al. (2002), stockperson personal characteristics, based on the mea-sures used by Seabrook, also did not correlate signifi cantly with milk yield, but did correlate with the attitudes of stockpeople. The relevance of the correlation between personality scores and stockperson attitudes will be discussed in the next section of this chapter. Beveridge (1996) investigated the relationship between personality types as measured by the MBTI (Myers and Myers, 1995) and the behaviour of dairy stockpeople towards cows. There was a negative relationship between judgement–perception scores and negative behaviour towards cows. Thinking–feeling scores tended to be correlated positively with

94 Chapter 4

positive and mildly negative behaviours, and to be negatively correlated with clearly negative behaviours. The MBTI correlated more strongly with measures of stockperson attitude and showed no correlations with milk yield. The rele-vance of the correlation between personality traits and stockperson attitudes will also be discussed in the next section of this chapter. Signifi cant relationships have been found in the pig industry between personality types of stockpeople and productivity in farrowing units. Seabrook (1996) reported that pig performance, measured by litter size, was associated with stockpeople with confi dent person-alities, emotional stability, independent personality, rational behaviour and low aggression. It should be noted that the measures of personality used here were not based on established psychological tests but relied on the face validity of a series of questions. Ravel et al. (1996), using an established but currently less-frequently used test, the 16 PF, found that self-discipline was a trait that appeared to be important at all farms studied: high insecurity and low sensitivity were associated with piglet survival at independent owner-operated farms, while stockpeople that were highly reserved and bold, suspicious, tense and change-able were associated with higher piglet mortality at large integrated farms. In both of these studies, there were no data to show that the causal link between stockperson personality and sow productivity was related to stockperson behav-iour.

The fact that several researchers, in different contexts and using different measures of personality, have been able to fi nd direct or indirect relationships between stockperson personality and production outcomes does suggest that personality may well be a relevant factor in livestock production. However, the fact that the dependent variables often are production outcomes rather than the behaviour of individual stockpeople means that, because many factors can inter-vene between stockperson characteristics and the productivity and welfare of the animals under his or her care, it may be diffi cult to determine the causal sequence between stockperson personality and animal productivity.

Empathy was defi ned earlier in this book (Section 1.4.2). While empathy is a dispositional characteristic, there is some argument about whether it is innate or learned. In a recent review, Duan and Hill (1996) distinguished between a trait approach, which is widely adopted by psychotherapists and others, and a situa-tion-specifi c social learning approach, which is amenable to training. This is important, because if a characteristic is learned, then it is amenable to change through training. Personality is normally regarded as relatively stable and not susceptible to change.

In the agricultural literature, the term ‘empathy’ has been used to describe the bond that exists between humans and animals under their care (English et al., 1992). In fact, an empathic bond may exist between stockpeople and their animals; however, empathy does not refer to the bond itself, which may have its origins in a number of factors of which empathy is one. Empathy refers to the way in which stockpeople may feel a bond with their animals because of being able to put themselves in the animal’s position or to understand the way in which the animals are reacting.

Only limited empirical data on empathy are available from livestock produc-tion. Beveridge (1996) found that empathy towards animals was positively

Attitudes of Stockpeople 95

associated with positive attitudes towards interacting with cows and positive beliefs about cows, but not directly with stockperson behaviour toward cows. Coleman et al. (1998) found that empathy toward animals was associated with positive beliefs about pigs and about handling pigs. Coleman (2004) reported that empathy was associated with positive behaviour towards pigs and a high level of intention to remain working in the pig industry. These three studies suggest that empathy may be a factor underlying the development of positive attitudes towards farm animals. This will be discussed in more detail below.

4.4 Attitudes and Personality

It may seem that attitudes and personality traits refer to the same thing. In fact, personality traits differ from attitudes in several ways. First, as previously men-tioned, personality traits produce thoughts and behaviours that are not directed towards some external object, but that relate to the individual. Attitudes, in con-trast, are specifi cally defi ned in terms of external objects or events. Secondly, attitudes are evaluative, that is, they refer to things in terms of good/bad, liked/disliked, etc. Then again, personality traits are descriptive: they describe charac-teristics of the person, which, while they may have behavioural implications, do not have a prescriptive element. Thirdly, personality traits are somewhat endur-ing. They are considered to be characteristics of the individual and, although there is some argument about the extent to which they are innate, they remain fairly stable from early childhood to adulthood, while attitudes are thought to develop over time and are a result of the experiences of the person. Fourthly, because personality traits are enduring, they do not usually respond to efforts to change them except, perhaps, under the infl uence of psychoactive drugs or as a result of brain surgery. Attitudes, conversely, are susceptible to change. Indeed, the whole advertising industry, political campaigns and pitches by salespersons and the like are all based on the view that attitudes can be changed.

As already discussed, Ajzen and Fishbein (1980) identifi ed personality as one of the key antecedents of attitude along with other individual characteristics (Fig. 4.2). These factors may weakly correlate with stockperson behaviour, but would be expected to exert their infl uence by affecting attitudes. There is some empirical support for this (see Section 4.3.1).

4.5 Stockperson Attitudes and Behaviour in Livestock Production

Our early published research studying the relationship between stockperson atti-tudes and behaviour was conducted in the pig industry (Hemsworth et al., 1989b, 1994a; Coleman et al., 1998). Based on Fishbein and Ajzen’s Theory of Reasoned Action, Hemsworth et al. (1989b) used an attitude questionnaire to obtain infor-mation on the behavioural beliefs of stockpeople about interacting with pigs.

The attitude questionnaire was in two parts. The fi rst half contained a series of belief statements about characteristics of pigs and the second half contained

96 Chapter 4

belief statements about interacting with pigs. The rationale for the development of the questionnaire was that the inclusion of general attitudes towards pigs would allow for the possibility that general attitudes might be related to aspects of the stockperson in areas other than behaviour towards pigs. The relationship between general attitudes towards pigs and other variables will be discussed later in the book. Attitudes towards interacting with pigs were assessed because these were most likely to predict stockperson behaviour towards pigs under Fishbein and Ajzen’s model. The content of the questionnaires was as follows:

1. Statements about pigs. This part of the questionnaire comprised 26 questions. Questions required stockpeople to answer on a 5-point scale (1 = Disagree strongly to 5 = Agree strongly) and the questionnaire included items such as ‘Pigs are noisy animals’ and ‘Pigs are stubborn animals’. High scores were therefore associated with negative attitudes.2. Attitudes towards interacting with pigs. This part of the questionnaire consisted of 46 questions assessing the behavioural beliefs of stockpeople. Ques-tions were answered on a 7-point scale and included items such as ‘How much physical effort do you need to use when moving gilts in oestrus?’. The response categories ranged from ‘A lot’ (1) to ‘Very little’ (7). Several variants of each type of question were asked, each variant representing a different target animal. The target animals included ‘Gilts in oestrus’, ‘Non-oestrous gilts’, ‘First litter sows in oestrus’, ‘First litter sows with piglets’, ‘First litter sows after weaning’, ‘Older sows in oestrus’, ‘Older sows with piglets’ and ‘Older sows after weaning’. Two additional attitude questions were included: ‘How do you feel about frequent patting and stroking of pigs?’ and ‘What do you think other farmers feel about patting or stroking of pigs?’. These questions were answered on 7-point scales ranging from ‘Good’ (1) to ‘Bad’ (7) and ‘Wise’ (1) to ‘Foolish’ (7). All scores were recorded, so that a high score indicated a negative attitude.

The ‘attitudes towards interacting with pigs’ questionnaire is a questionnaire to assess interrelated behavioural beliefs in a way similar to that used by Lynne and Rola (1988), as described in Section 4.2.6. Consistent responses to these behav-ioural belief questions refl ect the stockperson’s underlying attitude towards the particular kinds of interactions with pigs.

The nature of the behaviour of these stockpeople towards pigs during rou-tine mating activities, such as moving pigs for mating, conducting oestrus detec-tion and assisting pigs to copulate, was recorded in this study. Negative or aversive tactile behaviours by stockpeople that were recorded included mild, moderate and forceful hits, slaps and kicks, while the positive behaviours included pats, strokes and the hand of the stockperson resting on the back of the animal. Data on stockperson behaviour were collected and collated on the basis of the number of positive and negative behaviours used by the stockperson per pig handled so that absolute numbers of both positive and negative behaviours used per pig handled and the percentage of negative behaviours (i.e. the ratio of negative behaviours to the total number of physical behaviours (sum of positive and negative)) used by each stockperson could be studied.

In this study (Hemsworth et al., 1989b), highly signifi cant correlations were found between stockperson attitudes and stockperson behaviour. For example, a

Attitudes of Stockpeople 97

positive attitude towards petting pigs by the stockperson, refl ected in positive beliefs about the frequency of patting, stroking and talking to the animals while working with them, showed a large negative correlation with the percentage of negative behaviours used by stockpeople (Table 4.1). Similarly, a positive atti-tude to the use of verbal and physical effort (i.e. beliefs that considerable verbal and physical effort are not generally required to handle pigs) was also associated with a large negative correlation with the percentage of negative behaviours. In other words, stockpeople with a positive attitude in respect to these items gener-ally displayed a lower percentage of negative behaviour when interacting with pigs. These correlations therefore indicate that stockpeople with a good attitude towards handling pigs exhibited fewer negative behaviours towards pigs. In con-trast, poor attitudes by stockpeople were associated with a high percentage of nega-tive behaviours. These results are supported, particularly the petting–behaviour correlation, in a reanalysis of data (Table 4.1) from the study by Hemsworth et al. (1994a), in which stockpeople were later trained to improve their attitudes and behaviour towards pigs.

Surprisingly, similar attitude–behaviour relationships appear to exist in the dairy industry. In a study of human–stockperson interactions at 29 commercial dairy farms in Australia (Hemsworth et al., 1995), observations on stockpeople indicate that the attitudes of stockpeople towards interacting with animals are also predictive of the behaviour of the stockpeople towards their animals. Beliefs about petting animals (patting and talking to animals) and the amount of physi-cal and verbal effort required to move animals were related to the use of negative behaviours by the stockperson (Table 4.2). Positive attitudes to the use of petting and the use of verbal and physical effort to handle animals were negatively cor-related with the use of negative tactile behaviours such as slaps, pushes and hits. Another study on 31 dairy farms in Australia indicated very similar attitude–behaviour relationships (Breuer et al., 2000). As shown in Table 4.2, positive attitudes towards petting and the effort required to handle cows showed moderate-to-large negative correlations with the percentage of negative behav-iours used by stockpeople. Therefore, as in the pig industry, stockpeople were likely to use fewer negative behaviours when handling their animals if they believed that: (i) petting should be frequently used; and (ii) verbal and physical effort should be infrequently used when interacting with animals. Seabrook

Table 4.1. Stockperson attitude–stockperson behaviour correlations in the pig industry.

Correlation between positive behavioural beliefs and negative stockperson behaviour

Study Petting and behavioura Effort and behavioura

Hemsworth et al. (1989b) −0.61** −0.47*Data reanalysed from

Hemsworth et al. (1994a)−0.55** −0.12

aSignifi cant correlations (* = P < 0.05 and ** = P < 0.01) indicate associations between the two variables.Attitudes assessed on the basis of behavioural beliefs. Variable used to measure negative stockperson behaviour was the percentage of negative tactile behaviours used by the stockperson.

98 Chapter 4

(1994, p. 254) also has provided some illuminating reports of unsolicited stock-person attitude statements that appear to be directed towards behaviour, for example, ‘Every day it is the same old routine: feed, move pigs, feed, move pigs. When they won’t go where you want them to, it’s so easy to lash out with the foot, they are so stubborn’. Such statements are consistent with stockpeople behaving in a way determined by their attitudes.

More recent research on dairy cattle has shown similar results, using ques-tionnaires similar to those previously used by us (Hemsworth et al., 1995, 2000, 2002; Breuer et al., 2000). Waiblinger et al. (2002) investigated the relationships among attitudes and personal characteristics of stockpeople, their behaviours to cows, the behaviour of cows during milking towards humans and average milk yield. Thirty small dairy farms in Austria where cows were loose housed indoors were studied. Stockperson behaviours were classifi ed as positive and negative as well as neutral (moderate use of the hand, moderately loud vocalizations and gently using a stick). A negative correlation between the absolute number and percentage of positive behaviours and the avoidance of humans by cows was found. Also moderate use of the hand, moderately loud vocalizations and gentle use of a stick correlated positively with cows stepping and kicking during milking and negatively with milk yield. It was found that behavioural attitudes correlated signifi cantly with stockperson behaviour and cow milk yield. General attitudes showed limited correlations with stockperson behaviour, but correlated strongly with behavioural attitudes. Stockpeople scoring high on agreeableness used more positive behaviours and fewer neutral ones when handling cows.

Lensink et al. (2000a) studied the behaviour of stockpeople towards veal calves during one morning feed at 50 farms and then asked them to fi ll in ques-tionnaires designed to measure their attitude towards calves, and to obtain infor-mation about their background. Production data (daily weight gain, feed conversion and mortality rates) were obtained from the local veal company. The frequency of gentle contacts was positively correlated with the self-reported behaviour of stockpeople towards their calves and their beliefs about the sensi-tivity of calves. Females showed more positive behaviour towards the calves, had more positive beliefs about the importance of contacts with calves and pro-vided a more positive description of their own behaviour. Positive behaviour was associated with better productivity.

Table 4.2. Stockperson attitude–stockperson behaviour correlations in the dairy industry.

Correlation between positive behavioural beliefs and negative stockperson behaviour

Study Petting and behavioura Effort and behavioura

Hemsworth et al. (1995) −0.47** −0.36*Data reanalysed from

Breuer et al. (2000) −0.55** −0.28

aSignifi cant correlations (* = P < 0.05 and ** = P < 0.01) indicate associations between the two variables. Attitudes assessed on the basis of behavioural beliefs. Variable used to measure negative stockperson behaviour was the percentage of negative tactile interactions used by the stockperson.

Attitudes of Stockpeople 99

In beef cattle, Boivin et al. (2007) found that stockpeople were concerned about animal behaviour-related handling problems. They also found that stock-people generally showed positive attitudes towards the animals, towards the ani-mals’ behaviour during handling and towards ease of handling. Stockpeople emphasized human contact as the most important factor in determining the ease of handling cattle. The authors argued that this may be important in motivating beef cattle stockpeople to engage in training to improve the human–animal rela-tionship. Unfortunately, they did not demonstrate that these stockperson atti-tudes were related to cattle behaviour. However, Windschnurer et al. (2009a) conducted a small study of stockperson attitudes and behaviour towards bulls at ten farms. The attitude questionnaire comprised two parts: (i) affective (emo-tional) attitudes, which consisted of questions assessing the degree of comfort that stockpeople felt in different situations when working with bulls; and (ii) behavioural beliefs, which consisted of questions assessing whether stock-people considered gentle contacts (talking to bulls, stroking bulls, etc.) to be important during their daily work. Signifi cant negative correlations, based on farm averages, were found between affective attitudes by stockpeople, positive stockperson interactions with cattle and the fl ight distance of cattle to humans. No signifi cant correlations were found between stockperson beliefs about posi-tive behaviours towards cattle and fl ight distance; however, several of these cor-relations, although small, were in the expected direction. In a study of bulls at the time they were loaded on to trucks for transport to slaughter, Mounier et al.(2006) found that positive beliefs by stockpeople about bulls were negatively related to the ease of loading and unloading them by a truck driver. This result is diffi cult to interpret, because the behaviour of the bulls was assessed in a novel situation (being loaded on to a truck) by a stranger (the truck driver). Most other research has been carried out on the relationship between the attitudes of an animal’s handler, the behaviour of the handler and the animal’s behaviour in response to that handler. The authors concluded that close contacts with bulls by farmers should be limited to facilitate later animal handling, but, as they point out, there were no data on the handling that the bulls had experienced under the stockperson, so the interpretation of the results is equivocal.

More recently, Edwards (2009) conducted fi eld studies on Australian and US commercial egg farms in which hens were housed in cages. Negative attitudes to laying hens in general, to the sensitivity of hens and to working with laying hens were associated with the stockperson making more noise and moving faster in the housing facility. While a general negative attitude to hens was also associated with the stockperson spending more time in the housing facility, a negative atti-tude to working with hens was associated with the stockperson spending less time closely inspecting the birds. These observations in two countries indicate relationships between stockperson attitudes and behaviour in commercial egg farms.

There is some limited evidence of attitude–behaviour relationships in stock-people handling livestock before slaughter at abattoirs. Research on stockpeople handling pigs before slaughter (Coleman et al., 2003) found that stockpeople who felt under pressure to keep up with the killing chain and who believed that it is important to move the pigs as quickly as possible tended to be less likely to

100 Chapter 4

use the electric goad when it was turned off, that is, as a relatively benign aid to move animals, than did those who believed it is not important to move the pigs quickly (Table 4.3). Furthermore, the belief that the way in which pigs are han-dled when waiting to be slaughtered does not affect their behaviour was associ-ated with high use of the electric goad when it was turned on, that is, as a deliberate aversive stimulus to the pigs. Research on stockpeople handling beef cattle and sheep before slaughter (G.J. Coleman and P.H. Hemsworth, unpub-lished data, 2009) found that the stockperson beliefs about the pressures imposed by perceived lack of control over their actions, perceived time constraints, the effect of poor facilities and the importance of arousing livestock were all associ-ated with frequent use of forceful handling behaviours, such as shouting, hitting and use of the electric goad. For example, stockpeople tended to use more nega-tive behaviours in handling sheep and cattle before slaughter (Table 4.4) when that believe that: (i) they don’t have control over how they handle their animals; (ii) they are under time pressure when moving their animals; (iii) fl ooring does not affect animal behaviour; (iv) external factors such as previous handling pre-dominantly affect animal behaviour; and (v) animals need to be aroused and kept moving. These observed relationships between stockperson attitudes and behaviour at cattle, pig and sheep abattoirs are consistent with previous research on pigs and cattle in other handling situations.

Table 4.3. Correlations between stockperson attitudes and stockperson behaviour to pigs in an abattoir (from Coleman et al., 2003).

Frequency of electric goad use when

Questionnaire item Turned offa Turned ona

Behavioural beliefsIt is important to move the pigs into the

carbon dioxide stunner as quickly as possible0.48* −0.04

I move the pigs no more or less quickly than my co-workers

0.35 0.06

How the pigs are handled by me when waiting to be slaughtered does not affect their behaviour

−0.03 −0.42*

Using the electric goad is the most effective tool to get the pigs to do what is needed

0.09 −0.36

While pigs are in the carbon dioxide stunner, I decide how tightly to pack the pigs in the twin race

−0.15 0.31

Normative beliefsMany employees think that it is important to move

the pigs into the carbon dioxide stunner as quickly as possible

0.47* −0.18

aSignifi cant correlations (* = P < 0.05 and ** = P < 0.01) indicate associations between the two variables; n = 23 stockpeople.The attitude scales are recorded so that a high score refl ects disagreement with the question, that is, a positive attitude.

Attitudes of Stockpeople 101

Despite the variability in species studied, differences in farming methods (particularly between Europe and Australia) and some variations in methodol-ogy, there is a general consistency in fi ndings to indicate that the attitude of stockpeople to handling animals is related to the behaviour of the stockpeople towards their animals. Key beliefs appear to be those relating to petting and use

Table 4.4. Stockperson attitude–stockperson behaviour correlations in sheep and cattle abattoirs (sample size varies from 41–87 stockpeople).

Stockperson behavioura

Attitude factorbShouting Hitting Use of electric goad

Lack of control 0.19* 0.05 0.29**Time constraint −0.05 0.28** 0.10Flooring affects behaviour −0.39** −0.04 −0.35**External factors affect behaviour −0.06 0.19* 0.18*Arousing animals 0.13 0.02 0.21*

aSignifi cant correlations (* = P < 0.05 and ** = P < 0.01) indicate associations between the two variables.bAttitude factors: stockpeople that believed that: (i) they don’t have control over how they handle their animals; (ii) they are under time pressure when moving their animals; (iii) fl ooring does not affect animal behaviour; (iv) external factors such as previous handling predominantly affect animal behaviour; and (v) animals need to be aroused and kept moving. High scores indicate agreement with the statements.

Fig. 4.3. Positive attitudes to handling pigs are associated with positive contacts with pigs.

102 Chapter 4

of verbal and physical effort, with a positive attitude to these behaviours associ-ated with less negative tactile behaviour directed towards farm animals. There-fore, these observations indicate that the attitude of the stockperson towards farm animals may be an infl uential determinant of how the stockperson behaves towards farm animals (Fig. 4.3).

4.6 Conclusion

As there is compelling evidence to suggest that the behaviour of stockpeople may infl uence the behaviour, productivity and welfare of farm animals, it is important to identify the origins of the behaviour of stockpeople. The Theory of Reasoned Action (Ajzen and Fishbein, 1980) holds that ‘as a general rule, we intend to behave in favourable ways with respect to things and people we like and to display unfavourable behaviours towards things and people we dislike. And, barring unforeseen events, we translate our plans into actions’. There is good evidence in the pig and dairy industries that stockperson attitudes do pre-dict stockperson behaviour towards farm animals. Positive attitudes to the use of petting and the use of verbal and physical effort to handle animals are negatively correlated with the use of negative tactile behaviours, such as slaps, pushes and hits. Of course, it is likely that factors other than attitudes will contribute to the prediction of behaviour, and these factors will be discussed further in Chapter 6. A key issue, however, is the way in which these attitudes and behaviour develop, and their consequences for the productivity and welfare of farm animals. This is the subject of the next chapter.

© CAB International 2011. Human–Livestock Interactions, Second Edition (P.H. Hemsworth and G.J. Coleman) 103

5 Stockperson Behaviour and Animal Behaviour

5.1 Introduction

So far in this book we have considered many handling and fi eld studies, mainly with dairy cattle, pigs and poultry, that have shown that human–animal inter-actions in livestock production may have marked consequences for farm ani-mals, particularly in terms of their behaviour, physiology and productivity. For example as discussed in Chapter 3, growth and reproductive performance of farm animals may be reduced in situations where fear of humans is high and a chronic stress response is implicated in the adverse effects of fear for humans on farm animal productivity. The implications of human–animal interactions on the welfare and productivity of livestock highlight the need to study the relationships between stockperson behaviour and animal behaviour.

Because attitudes are the main dispositional factor affecting volitional human behaviour, there are likely to be opportunities to manipulate human–animal interactions in order to infl uence farm animal welfare and productivity by improving the attitudes and behaviour of stockpeople towards farm animals. Therefore, in Chapter 4, the nature of attitudes and their relationship to behav-iour were discussed both from a theoretical perspective and in terms of some of the available data from the livestock industries. A key issue is how these stock-person attitudes and behaviours affect the behaviour of farm animals. This is the subject of this chapter.

While animals have some instinctive behaviours, for example avoidance of strange or novel stimuli or certain fi xed mating patterns in response to a hor-monal state during oestrus, many behaviours are learned. Farm animals that have had a history of aversive handling learn to avoid stockpeople. The basic principles of learning apply equally well to animal learning as to human learning. Because of the particular importance of learning in understanding how the attitudes and behaviour of stockpeople affect farm animals and how the behav-ioural responses of farm animals to humans develop, the concepts of learning

104 Chapter 5

and reinforcement need to be understood, as do the ways in which conditioning, in its various forms, operates.

5.2 Kinds of Learning

Learning is generally defi ned as a relatively enduring change in behaviour that occurs as a result of experience. Behaviour refers to observable actions on the part of individuals and we can measure underlying cognitive events by their expression in behaviour. Attitudes fall into this latter category. Not all relatively enduring changes are examples of learning because ageing or disease may also produce such changes. Experience in the particular behaviour or in a closely related behaviour is central to learning.

While there are other forms of learning recognized, such as habituation, sensitization and socialization or imprinting, the most fundamental type of learning involves the process where an event, the stimulus, comes to produce a particular behaviour, the response, under conditions where this would not have occurred previously. This process is called conditioning and has two basic forms: classical and instrumental. The establishment of this connection between stimu-lus and response in the process of conditioning requires the presence of a reinforcer.

5.2.1 Reinforcement and motivation

Reinforcement is diffi cult to defi ne and there have been many arguments in the literature about how to arrive at a non-circular defi nition. Reinforcement is generally defi ned as anything that follows a response and that increases the likelihood of the response occurring when the same stimulus is presented again. There are two basic kinds of reinforcers, positive and negative. A positive reinforcer is something that is pleasurable or rewarding. Thus food to a hungry individual or water to a thirsty individual would be positively reinforcing. Negativereinforcers are those that arise when some aversive stimulus is removed. It is the release from the aversive stimulus that has the reinforcing effect. So removal of an electric shock or the cessation of punishment are examples of negative reinforcers. Often it is fairly easy to determine, a priori, whether or not some-thing will act as a reinforcer. For example, those things, such as palatable food, which appear to be universally attractive to a person or another animal, would be expected to be reinforcers; however, even then, if the person has just eaten a large meal, this may not be the case. It is often not possible to determine whether such apparently obvious reinforcers really are reinforcing without observing the development of stimulus–response sequences. What this means is that the strength of a reinforcer often cannot be determined without considering the motivational state of the person or animal. For example, attempts by a boar to mate with a gilt or sow are quite aversive for the female if she is not in oestrus, but courtship and copulation are positively reinforcing if she is in oestrus.

Stockperson Behaviour and Animal Behaviour 105

As discussed in Chapter 3 (Section 3.4.1), the study of motivation is con-cerned with identifying the causal factors that are responsible for an animal’s behaviour, such as feeding, drinking, copulating and fl eeing. It has been recog-nized by many authors that it is generally necessary to consider explanations for observed behaviour in terms of states within the organism (motivations or drives), experiences and learned associations. Indeed, Toates (1980, p. 2) states that ‘the fact that drive cannot be observed directly and is dependent upon such things as hormone level and learning, is perhaps no argument against it having a provisional usefulness’. For example, if an animal is hungry, its physiological state will generate a motivation or drive that is directed towards food. It is impor-tant to realize that motivations directed towards satisfying the basic physiological needs, such as food and water, form only one type of motivational state, have an innate basis and are relatively non-contentious. In contrast, a motivational state associated with, for example, cognitive consistency may be learned rather than innate and is relatively controversial. Thus, there is a complex interaction between innate characteristics and learned characteristics in determining moti-vations. The key issue is that there are a large number of causal factors, which, by affecting arousal in the individual, may increase or decrease the intensity of a reinforcing stimulus.

5.2.2 Conditioning processes

Classical conditioning is the most fundamental form of learning and was discov-ered in the experiments by the famous Russian scientist Ivan Pavlov last century (Pavlov, 1960). Pavlov had observed that his dog began to salivate before actu-ally being given food. He reasoned that some stimulus other than actual eating must trigger the salivation. He then rang a bell each time he gave the dog food. Subsequently, he found that ringing the bell alone was suffi cient to induce saliva-tion. Thus the salivation response had become conditioned to the bell (Fig. 5.1). To put it more formally, Pavlov paired an unconditioned stimulus (food) with a conditioned stimulus (bell) to eventually produce an unconditioned response (salivation) to the bell alone. This process of conditioning relies on two princi-ples. First, food is unconditioned, that is, the dog has already established the reinforcing properties of food. Secondly, the conditioned stimulus (bell) has been singled out from all other possibly relevant stimuli and has been presented sys-tematically in conjunction with the food.

Instrumental (operant) conditioning differs from classical conditioning in that there is no explicit pairing of conditioned and unconditioned stimuli. B.F. Skinner (1969) discovered instrumental conditioning and it is best illustrated by describing his experimental procedure. Skinner described an experiment in which a pigeon was placed in a box with a button on the wall. Pressing the but-ton will cause the release of a food pellet. The bird has not learned that pressing the button will release the food, so it will explore the box. Eventually, by chance, it will peck the button, releasing the food pellet (Fig. 5.2).

Because the pigeon has not learned that pressing the button caused the release of food, it will continue with other behaviours until it pecks the button

106 Chapter 5

again. Eventually, the contingency between the button and food will become learned and the pigeon will peck the button to obtain the food. The button will have become operantly conditioned to the food. This form of conditioning dif-fers from classical conditioning in that the individual must discover the stimulus–response contingency.

Unconditioned stimulus(US)Food

Conditioned stimulus(CS)Bell

Pairing

Unconditioned response(UR)

Salivation

Unconditionedassociation

Conditionedassociation

Fig. 5.1. Classical conditioning. Repeated pairing of food and the bell will eventually lead to the bell alone being able to elicit salivation.

Conditioned stimulus(CS)

Button

R2 (Peck button)Conditioned response

R1

R3

Unconditioned stimulus(US)

Food pelletReinforcement

Fig. 5.2. Instrumental (operant) conditioning. When the correct response is elicited, initially by chance, reinforcement is given. Eventually the response R2 is learned as the correct response. R1 and R3 are behaviours other than pecking the button.

Stockperson Behaviour and Animal Behaviour 107

5.2.3 Primary and secondary reinforcers

When an unconditioned stimulus has pre-existing reinforcing properties, it is called a primary reinforcer. The notion of pre-existing is diffi cult to defi ne, but food and water are clearly primary reinforcers by this defi nition. The term ‘pri-mary’ is used in a way similar to its use in motivation theory, i.e. ‘primary’ implies ‘unlearned’. When a stimulus has become conditioned, it may also act as a reinforcer. It is then referred to as a secondary reinforcer because it is not itself a ‘natural’ reinforcer. In the example of classical conditioning in dogs given above, the bell could act as a secondary reinforcer once it has become condi-tioned. Thus, pairing of the bell with, for example, a fl ash of light could condi-tion the salivation response to the fl ash of light. In this case, the bell has taken on the reinforcing properties of the primary reinforcer, food, in the original experiment. In fact, most secondary reinforcers do not retain their reinforcing properties indefi nitely unless they are regularly paired with the original uncon-ditioned stimulus.

Reinforcement does not need to be constant for learning to take place. In fact, partial reinforcement schedules tend to result in learning that is resistant to later attempts to change the learned response. For example, if, in the classical conditioning experiment described earlier, the food was paired with the bell only 50% of the time, and if the occurrence of the food with the bell was random, the bell would become more strongly conditioned than it would have been in the original experiment.

5.3 Relevance of Learning Theory to Fear of Humans

In Chapter 3 we proposed that fear can usefully be viewed as an emotion, affectedby both internal and external stimuli that elicit behavioural and physiological responses, that may assist the animal to respond appropriately to a source of danger.

Level of fear in animals can be inferred from behavioural observations. When an animal is confronted with a fear-provoking stimulus, there is likely to be a number of behavioural responses available to the animal. However, because fear responses function to protect the animal from harmful stimuli, we have pro-posed that the amount of avoidance of the experimenter or, conversely, the amount of approach to the experimenter are useful measures of the animal’s fear of humans. For example, in studies on pigs, the latency to and the amount of approach to a stationary experimenter in a standard test have been used to measure the level of fear of humans in pigs (see Section 3.4.3). Furthermore, fear-provoking stimuli will also elicit responses of the autonomic nervous system and the neuroendocrine system, and thus fearful animals in the presence of humans will, for example, have increased secretions of the catecholamine hor-mones adrenaline and noradrenaline and of the corticosteroid hormones cortisol and corticosterone (see Section 3.6).

Many animal behaviours are acquired through learning. More specifi cally, animals learn behaviours through operant and classical conditioning, which

108 Chapter 5

optimize their capacity to live in their environment. For example, there is sub-stantial evidence to show that farm animals learn to avoid stockpeople from whom they have received aversive stimulation (Section 3.5). In this case, avoid-ing the aversive handling is acting as a negative reinforcer. Similarly, farm animals show greater approach behaviour towards stockpeople from whom they have received positive interactions (Section 3.5). Here, positive handling is acting as a positive reinforcer.

5.4 Stockperson Behaviour and Animal Behaviour

Over the past 30 years, we and, more recently, European and Canadian research-ers have been studying stockperson behaviour–animal behaviour relationships in the livestock industries. This research has shown some large and consistent relationships between stockperson behaviour towards animals and the behav-ioural response of animals to humans across a number of livestock industries (see Section 3.6), which generally confi rms the predictions of numerous handling stud-ies that have been conducted under experimental conditions (see Section 3.5). We will now review in more detail some of the industry studies that reveal these relationships between stockperson behaviour and animal behaviour.

Consistent correlations between stockperson behaviour and the level of fear of humans by pigs have been found in the Australian pig industry. The correla-tions found in two of these studies are presented in Table 5.1. In these two studies,the variable used to measure negative stockperson behaviour was the percentage of negative tactile behaviours used by the stockperson in handling breeding sows, and fear of humans was assessed by measuring the time that sows spent near a stationary experimenter in a standard test. Negative tactile behaviours by stock-people were defi ned as mild to forceful slaps, hits, kicks and pushes, while the positive tactile behaviours included pats, strokes and the hand resting on the pig’s back. As shown in Table 5.1, the use of negative behaviours, measured in terms of percentage of the total tactile behaviours by the stockperson, was positively correlated with the level of fear of humans by pigs: high fear levels were observed where stockpeople displayed a high percentage of negative tactile behaviours.

Table 5.1. Stockperson behaviour–animal behaviour correlations in the pig industry.

StudyNegative stockperson behaviour and

level of fear for humansa

Hemsworth et al. (1989b) 0.45*Data reanalysed from Coleman et al. (2000) 0.40*

aSignifi cant correlations (* = P < 0.05) indicate associations between the two variables. Negative stockperson behaviour was assessed by the percentage of negative tactile interactions, while fear of humans at the farm was assessed by the time sows spent near a stationary experimenter.

Stockperson Behaviour and Animal Behaviour 109

Surprisingly, high levels of fear of humans were best predicted when the classifi ca-tion of negative behaviours included not only forceful kicks, hits, slaps and pushes, but also negative behaviours used with less force, such as mild and moderate slaps, prods and pushes. This clearly indicates the sensitivity of pigs to mild and moder-ate negative behaviours by humans, something that is not intuitively obvious to most of us.

The variation observed in the behaviour of stockpeople towards breeding pigs in the Australian pig industry is shown in Fig. 5.3. Two types of farms are represented in this fi gure: small independent farms in which one stockperson is

Small independent farms

Large integrated farms

100

80

60

40

Neg

ativ

e in

tera

ctio

ns (

%)

Neg

ativ

e in

tera

ctio

ns (

%)

20

0

0

10

20

30

40

50

60

70

80

90

1 3 5 7 9 11 13 15 17 19 21

Stockpeople

Stockpeople

23 25 27 29 31 33 35

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45

Fig. 5.3. Variation in the behaviour of 35 stockpeople at small independent piggeries and 46 stockpeople at large integrated piggeries (P.H. Hemsworth and G.J. Coleman, unpublished data, 1998). The behaviour variable is the percentage of negative tactile behaviours used by stockpeople in handling breeding pigs in the mating facility.

110 Chapter 5

predominantly responsible for mating management, and large integrated units in which teams of stockpeople are responsible for mating management. As can be seen from these data, the variation in stockperson behaviour, as assessed by the percentage of negative tactile behaviours used in moving breeding pigs and when conducting oestrus detection and assisted matings, is very similar at both types of farms. The important observation here is that there is substantial varia-tion in the use of negative behaviours used by stockpeople at both types of farms. In particular, there is a large number of stockpeople at both types of farms that display a high percentage of negative tactile behaviours. For example, 57% and 43%, respectively, of stockpeople at small independent and large integrated units consistently displayed more than 60% negative behaviours when handling breeding pigs. This high percentage of stockpeople that display predominantly negative behaviours towards pigs is of serious concern for the industry and its consequences will be considered in detail for all animal industries later in this chapter and in Chapter 6.

It should be appreciated that the pig industry in Australia is typical of inten-sive pig production systems in other countries. A comparison of the fear levels observed in sows at Dutch farms with that of sows observed at Australian farms (Fig. 5.4) indicates the relevance of the Australian data to the international pig industry.

Observations in the Australian dairy industry indicate that behavioural pat-terns of stockpeople, similar to those observed in the pig industry, also regulate the fear responses of commercial cows to humans. In two studies involving 31 and 66 farms (Breuer et al., 2000; Hemsworth et al., 2000) in which cows grazed on pasture all year in herds of 100 to 350, signifi cant correlations were found between the use of negative tactile behaviours used by stockpeople to move cows for milking and the level of fear of humans by cows (Table 5.2). The main negative behaviours were pushes, slaps, hits and tail twists while moving cows in and out of the milking facility and into position for milking, while the main positive behaviours were pats, strokes and the hand on the cow’s fl anks or legs during milking. In the fi rst study, the percentage of forceful negative behaviours was most correlated with fear, while in the second study, the frequency of both moderate and forceful negative behaviours was most correlated with fear. These associations indicate that fear of humans was high at farms in which stockpeople frequently displayed negative tactile behaviours.

More recent observations by Waiblinger et al. (2002) at 30 small Austrian dairy farms in which cows were loose housed indoors in herds of 25 to 50 cows found similar correlations between stockperson behaviour and fear of humans by cows (Table 5.2). In this study, negative stockperson behaviour included forceful hits as well as shouting. In a subsequent study, Waiblinger et al. (2003) found that the frequency of touching and talking to cows at 35 small Austrian dairy farms was signifi cantly and negatively correlated with the distance at which cows showed avoidance to an approaching experimenter. As discussed in Chapter 3 (Section 3.5.1), Lensink et al. (2000b, 2001b,c) found that commer-cial calves that were regularly stroked on their necks and shoulders and allowed to suck the stockperson’s fi ngers spent more time interacting with a stationary unfamiliar experimenter in their home pens and showed less withdrawal

Stockperson Behaviour and Animal Behaviour 111

from an approaching unfamiliar and familiar experimenter in front of their pens than calves that received minimal contact with the experimenter around feeding time.

The variation observed in the behaviour of stockpeople towards dairy cows in the Australian dairy industry is shown in Fig. 5.5. As was observed in the pig industry, there is substantial variation in the behaviour of stockpeople towards cows, and because of the relationships between stockperson behaviour, animal fear and animal productivity, this variation raises concern for cow productivity and welfare. As discussed earlier (Sections 3.6 and 3.7), fear and the associated stress have important implications for both animal productivity and welfare.

160

19 Australian farms

12 Dutch farms

140

120

100

80

Tim

e to

inte

ract

(s)

60

40

20

01 2 3 4 5 6 7 8 9 10 11 12

1 2 3 4 5 6Commercial farms

7 8 9 10 11 12

13 14 15 16 17 18 19

160

140

120

100

80

Tim

e to

inte

ract

(s)

60

40

20

0

Fig. 5.4. Variation in fear of humans, as assessed by the approach behaviour of sows to a stationary experimenter at 19 Australian and 12 Dutch piggeries (data reanalysed from Hemsworth et al., 1981b, 1989b).

112 Chapter 5

Together with handling studies, these studies with commercial dairy cows and pigs indicate that learned approach–avoidance responses develop as a con-sequence of associations between the stockperson and the aversive and rewarding elements of the handling bouts. The main aversive properties of stockpeople include hits, slaps and kicks by the stockperson, while the rewarding

Table 5.2. Stockperson behaviour–animal behaviour correlations in the dairy and poultry industries.

Industry StudyNegative stockperson behaviour and

level of fear for humansa,b,c

Dairy Breuer et al. (2000)d 0.30Hemsworth et al. (2000)e 0.33**Waiblinger et al. (2002)f 0.47**

Meat chicken Hemsworth et al. (1996a)g 0.32Cransberg et al. (2000)h 0.43*

Egg (hens) Edwards (2009)i 0.59**

aSignifi cant correlations (* = P < 0.05 and ** = P < 0.01) indicate associations between the two variables. bNegative stockperson behaviour was assessed in studies: d,fby the percentage of negative tactile behaviours such as hits and slaps; eby the frequency of negative tactile behaviours such as hits and slaps; g,hby speed of movement; and iby the noise made by the stockperson.cFear of humans at the farm was assessed in study: dby the time cows spent near a stationary experimenter; eby the percentage of cows that approached close to a stationary experimenter; fby the distance at which cows showed avoidance to an approaching experimenter; g,hby the number of chickens that remained close to an experimenter moving in standard manner among the chickens; and iby the proportion of hens that remained close to an approaching experimenter.

100

80

60

40

20

01 4 7 10 13 16 19 22 25 28 31 34

Stockpeople

Neg

ativ

e in

tera

ctio

ns (

%)

37 40 43 46 49 52 55 58 61

Fig. 5.5. Variation in the behaviour of 63 stockpeople at dairy farms (P.H. Hemsworth and G.J. Coleman, unpublished data, 2000). The behaviour variable is the percentage of negative tactile behaviours used by stockpeople when moving dairy cows for milking.

Stockperson Behaviour and Animal Behaviour 113

properties include pats, strokes and the hand of the stockperson touching the animal. Furthermore, shouting and talking appear to be negative and positive, respectively, for dairy cows. It is the use of these negative and positive behav-iours that appear to determine the commercial animal’s fear of humans.

Observations have also been conducted in two studies on stockperson and bird behaviour at 22 and 24 Australian commercial meat chicken farms (Hemsworth et al., 1996a; Cransberg et al., 2000). It was found that the behav-iour of the stockperson in the accommodation facility was moderately associated with levels of fear of humans by chickens. As shown in Table 5.2, the speed of movement of the stockperson while routinely inspecting chickens and their facili-ties was positively correlated with the level of fear for humans by chickens. Fear of humans was assessed by measuring the avoidance response of birds to an approaching experimenter. These observations indicate that speed of movement by the stockperson is one of the visual behaviours of stockpeople infl uencing fear of humans at commercial meat chicken farms. There is some evidence that wav-ing and creating noise may affect fear responses in commercial chickens. While Cransberg et al. (2000) found that the combined frequency of waving and tapping on objects by the stockperson when moving through the facility was pos-itively associated with fear of humans (correlation coeffi cient of 0.41, P < 0.05), Hemsworth et al. (1996a) found that the frequency of waving by the stockperson was negatively correlated with the fear (correlation coeffi cient of −0.53, P < 0.01) but the frequency of tapping on objects by the stockperson was positively associ-ated with fear of humans (correlation coeffi cient of 0.41, P < 0.05).

Recently, observations have been conducted on stockperson and bird behaviour at 29 Australian and US commercial egg farms in which hens were housed in cages (Edwards, 2009). As shown in Table 5.2, there was a signifi cant correlation found between the frequency with which stockpeople made noise during routine inspection and maintenance in the accommodation facility and level of fear for humans by laying hens. Noise made by stockpeople included loud vocalizations, cleaning with an air hose or leaf blower and banging equip-ment during cleaning and maintenance. Fear at the farms was assessed by the avoidance behaviour of the birds to an experimenter who closely approached and withdrew twice from a sample of cages in the facility. A composite score representing the proportion of birds that had their heads out of the cages when the experimenter was in close proximity was calculated for each farm. It is also of interest that the frequency of tactile contact with cages or hens by stockpeople was signifi cantly and negatively correlated with the fear of humans for birds based on their avoidance behaviour to the approaching experimenter (correla-tion coeffi cient of 0.48, P < 0.05). These associations indicate that fear of humans was high at farms in which stockpeople frequently made noise and infre-quently approached closely and had tactile contact with birds and cages.

Handling studies also indicate that poultry appear to be particularly sensitive to close and regular visual contact with humans and, indeed, positive visual contact may be more effective in reducing levels of fear of humans than tactile contact (Section 3.5.2). In addition, rapid speed of movement by the stock-person, sudden and unexpected exposure to the stockperson and noise made by the stockperson may be fear provoking for poultry.

114 Chapter 5

In a recent study examining handling–animal stress relationships at cattle and sheep abattoirs, it was found that the frequency of goad use and auditory behaviours by stockpeople prior to slaughter were correlated with plasma cortisol concentrations in cattle post-slaughter, while intensity of dog use and frequency of auditory and tactile behaviours by stockpeople were correlated with plasma cortisol concentrations in sheep post-slaughter (Hemsworth et al., 2008). Overall then, increased dog use, reduced auditory behaviours, such as whistles, reduced tactile behaviours, such as touches and pushes, and increased goad use were all associated with increased cortisol concentrations. These limited data indicate that brief handling post-farm gate may affect stress responses in livestock (Fig. 5.6).

These observations in the dairy, pig and poultry industries indicate that an important determinant of the fear response of farm animals to humans is the behaviour of the stockperson towards the animals. The large variation in the behaviour of stockpeople towards humans raises some serious concerns for the welfare and productivity of farm animals in situations where the behaviour of the stockperson is predominantly negative, and this aspect is briefl y consid-ered later in this chapter in Section 5.6.

5.5 Farm Animals’ Discrimination Between Humans

We have proposed in this book that the history of interactions between the stock-person and the animal determine the subsequent stimulus properties of the

Fig. 5.6. Whistling and slow, deliberate and predictable movements by stockpeople are associated with reduced stress responses in sheep.

Stockperson Behaviour and Animal Behaviour 115

human for the animal, that is, the animal’s perceptions of humans. The charac-teristics of these stimulus properties include the extent to which the animal has habituated to the presence of the stockperson and the extent to which the stock-person has been associated with rewarding or aversive events. Furthermore, we have previously suggested that, through the process of stimulus generalization, the behavioural response of a farm animal to an individual human will extend to all humans (Hemsworth et al., 1993). Stimulus generalization can be defi ned as a tendency for stimuli similar to the original stimulus in a learning situation to produce the response originally acquired (Reber, 1988).

Evidence from a number of handling studies supports this view that the ani-mal’s response to a single human might extend to include all humans through this process of stimulus generalization. For example, pigs that previously were briefl y handled fi ve times per week for 6 weeks either by a handler in a pre-dominantly negative manner, or by two handlers who differed markedly in the nature of their behaviour towards pigs, showed similar behavioural responses to familiar and unfamiliar handlers (Hemsworth et al., 1994b). Similarly, Jones (1994) found that young chickens briefl y handled twice daily from 1 day of age in a positive manner showed similar behavioural responses to familiar and unfamiliar handlers wearing either similar or different clothing when tested at 10 or 11 days of age.

Studies conducted on commercial pigs and poultry also indicate that farm animals in commercial conditions may not discriminate between familiar and unfamiliar humans. Barnett et al. (1993) found that the avoidance responses of commercial laying hens to humans varying in a number of attributes, such as sex, height, clothing and wearing spectacles, were not different. The approach behaviour of adult pigs to a familiar handler and an unfamiliar human were similar in a standard approach test conducted at a one-person piggery (Hems-worth et al., 1981b), and Bouissou and Vandenheede (1995) reported that the avoidance responses of sheep towards a human and a human-like model were similar but greater than the response to a novel cylindrical model bearing little resemblance to a human.

These results suggest that, in commercial situations, the behavioural response of commercial livestock to one handler may extend to other humans. However, it is possible that there are handling situations in which commercial livestock may not exhibit stimulus generalization. We have previously suggested (Hemsworth et al., 1993) that in situations in which there is intense handling predominantly of a positive nature by a stockperson, animals may learn to discriminate between this handler and other handlers to which the animals may be subsequently exposed. For example, several studies with rodents indicate that animals can discriminate between the handler with whom the animals have had substantial contact, presumably of a positive nature, and a stranger (see review by Dewsbury, 1992). Following an extensive period of intense human contact (10 min/day for 3 weeks), Tanida et al. (1995) found that young pigs showed greater approach to the familiar handler than to an unfamiliar handler, even though both handlers wore similar clothing.

Furthermore, in situations in which the physical characteristics of the handlers differ markedly, farm animals may learn to discriminate between the

116 Chapter 5

handlers. For example, in a series of experiments, de Passille et al. (1996) found that dairy calves exhibited clear avoidance of a handler who had previously handled them in a negative manner in comparison with either an unfamiliar handler or a handler that had previously handled them in a positive manner. In order to increase the chances of discrimination, handlers wore different coloured clothing. Initially there was a generalization of the aversive handling, with calves showing increasing avoidance of all handlers, but with repeated treatment, calves discriminated between handlers, and in particular between the ‘negative’ and ‘positive’ handlers. It is of interest that discrimination was greatest when tested in the area in which handling had previously occurred rather than in a novel loca-tion. In fact, in one experiment when animals were tested in a location in which the handling was not performed, 40% of the calves actually approached and interacted with the negative handler. These data on calves indicate that discrim-ination between people by farm animals will be easier if the animals have some distinct cues on which they can discriminate, such as colour of clothing or loca-tion of handling.

Although there is evidence that farm animals are capable of discriminating between stockpeople, they may not do so under normal commercial circum-stances. Certainly, it is necessary to take into account the number of stockpeople, the environment in which handling occurs, the distinctive features of the stock-people and the intensity and duration of interactions before being able to predict that farm animals would discriminate between stockpeople. Nevertheless, even when farm animals learn to discriminate between humans, fear responses to humans in general are likely to increase in response to the most aversive handler (Hemsworth et al., 1994b; de Passille et al., 1996). Such a fi nding has important implications in situations in which several stockpeople may interact with farm animals.

5.6 Implications of High Fear of Humans for Farm Animals

Some of the implications of high fear of humans for farm animals have been con-sidered in Chapter 3 and will be considered further in Chapter 6; however, it is useful at this stage to briefl y review the potential adverse effects of high fear levels.

5.6.1 Fear and productivity

As considered in Chapter 3 (Section 3.6) and in this chapter (Section 5.4), there is evidence from both handling experiments and observations in the animal industries that the combination of high fear levels and regular human contact may reduce the productivity of dairy cows, pigs and poultry. The mechanisms responsible are unclear in some species, but studies on pigs and, to a lesser extent, dairy cattle show that in fearful animals, a chronic stress response or even a series of acute stress responses in the presence of humans may be responsible for the depressed productivity in these fearful animals. Support for this is provided by the known effects of corticosteroids on nitrogen balance, energy

Stockperson Behaviour and Animal Behaviour 117

balance, key reproductive events and immunosuppression. In more extensive settings, chronic stress responses in livestock as a consequence of negative human contact are less likely than in intensive systems because of less contact with humans and the greater opportunity for the animals to control this contact. Nevertheless, acute stress responses are likely in situations in which fearful animals are in close contact with humans.

5.6.2 Fear and welfare

Fear of humans may also reduce the welfare of farm animals. Concern for the welfare of fearful animals arises for a number of reasons. Fear is generally considered an undesirable emotional state of suffering. While fear may lead to injuries in animals trying to avoid humans during routine inspections and hand-ling, there is also substantial evidence that when regularly interacting with humans, fearful animals may experience a chronic stress response and immuno-suppression, which in turn may have serious consequences on the growth and health of the animals. Additional concerns for the welfare of these fearful animals will be considered in later in this book (Section 6.3).

5.6.3 Fear and ease of handling

Fear of humans can affect the ease with which farm animals can be observed and handled by stockpeople. There is limited evidence in pigs and other farm animals that highly fearful animals are generally the most diffi cult to handle. As shown in Table 5.3, consistent correlations have been found between the behav-ioural response of pigs to humans and their ease of handling. These correlations indicate that pigs which showed high levels of fear for humans, based on their approach behaviour to an experimenter in a standard test, were the most diffi cult pigs to move along an unfamiliar route. These fearful pigs took longer to move,

Table 5.3. Correlations between the behavioural response of pigs to an experimenter in a standard approach test to assess fear of humans and the ease of movement of 24 pigs along an unfamiliar route by an unfamiliar handler (Hemsworth et al., 1994c).

Variable recorded in ease of movement testa

Variable recorded to assess fear of humans Time to move Baulks Scoreb

Time to approach experimenter 0.34 0.44* −0.63**Number of interactions with experimenter −0.42** −0.42* 0.51*

aSignifi cant correlations (* = P < 0.05 and ** = P < 0.01) indicate associations between the two variables.bScore was given based on ease of movement, with 0 refl ecting substantial diffi culty and 4 refl ecting little or no diffi culty in moving the pig.

118 Chapter 5

displayed more baulks and were subjectively scored as the most diffi cult to move by the handler. Intuitively, while it might be expected that fearful animals are easy to move, perhaps because they might tend to move quickly, animals that are fearful of humans may actually be diffi cult to handle for several reasons. Pigs will avoid places containing urine from stressed pigs (Vielle-Thomas and Signoret, 1992), presumably because stressed pigs release pheromones in their urine and saliva which communicate to other pigs that they have encountered a diffi cult situation. Thus, pigs will often pack together when alarmed and stressed and, consequently, sorting and handling are likely to be diffi cult. Furthermore, pigs are generally wary of moving towards an unfamiliar or unpredictable situation (Hemsworth, 2000), and if they are fearful of both the environment and the handler, they are more likely to show exaggerated behavioural responses to han-dling, such as baulking or fl eeing back past the handler. In other words, faster movement by stressed pigs may be counterproductive. Other studies on pigs also indicate that high levels of fear for humans, either as a consequence of reduced human contact or human contact of a negative nature, may decrease the ease of handling of pigs (Gonyou et al., 1986; Grandin et al., 1987). Several studies have shown that pigs regularly moved out of their pens prior to slaughter were quicker to move during the early stages of transport (Abbott et al., 1997; Gever-ink et al., 1998). While increased familiarity with locations outside the pigs’ pens may be responsible for these effects, increased human contact may also be impli-cated as Eldridge and Knowles (1994) reported that commercial grower pigs that were regularly handled and moved out of their pens to a range of locations were easier to move in an unfamiliar environment. In contrast, Hill et al. (1998) found no effect of fear for humans on the time taken to regularly move pigs to and from a weighing area, perhaps because familiarity with the location may facilitate ease of handling.

Other studies, particularly on cattle, also indicate that fear of humans can affect the ease with which farm animals can be observed and handled by stock-people. For example, Boissy and Bouissou (1988) found that calves that were positively handled by brushing and leading on a halter, had reduced fl ight dis-tances, were easier to capture and lead on a halter and had reduced heart rates and cortisol concentrations after capture and leading. Lensink et al. (2001a,b) found that the dairy calves that predominantly received positive handling during rearing required less effort to load for transport and had lower heart rates during loading than those that received either minimal human contact or predominantly negative handling, such as hitting and shouting, during rearing. Waiblinger et al.(2004) found that previous handling of dairy cows in the form of talking, feeding and stroking reduced heart rates, kicking and restless behaviour in both the pres-ence of humans and during rectal palpation. In contrast, Mateo et al. (1991) found that while regular positive handling of ewes in the form of talking and stroking resulted in increased approach to a stationary experimenter, this was not general-ized to other situations, such as avoidance responses to shearing and halter restraint. Hargreaves and Hutson (1990c) found that although talking, stroking and feeding sheep hay and lupins reduced their fl ight distance and heart rate responses to the approaching experimenter, the sheep were slower to move along a familiar route to a location in which they were either caught or caught and shorn. It is possible that

Stockperson Behaviour and Animal Behaviour 119

sheep that were less fearful of the experimenter may show more exploration of the familiar route. In contrast, as proposed earlier, more handling diffi culties may be encountered with animals that are fearful both of the handler and the location in which they are being moved.

Some observations in the dairy industry indicate positive correlations between the use of negative stockperson behaviours, such as slaps, hits and loud vocaliza-tions during handling, and restlessness, in the form of fl inch, step and kick responses, in the close presence of the stockperson during milking (Breuer et al., 2000; Hemsworth et al., 2000; Waiblinger et al., 2002). Rushen et al. (1999) found that cows stepped more frequently during milking in the presence of an experimenter who had hit or shocked them with a battery-operated cattle prodder than in the presence of an experimenter who had brushed, fed and gently spoken to them. However, cows in the presence of the experimenter that had handled them nega-tively kicked less during milking preparation. While the close presence of the stock-person may affect restlessness in fearful cows, the close presence of dominant cows at milking or painful conditions, such as lameness and mastitis, may also affect restlessness at milking.

Therefore, while fear of humans may affect ease of handling other factors, such as the animal’s familiarity and experience with the procedure and location and the handling behaviour of the stockperson at the time of handling, are likely to affect ease of handling. Nevertheless, fear of humans is likely to have implica-tions for the ease with which farm animals can be observed and handled during routine handling and husbandry.

5.7 Conclusion

As with experimental studies, observations on commercial pigs and dairy cows indicate that conditioned approach–avoidance responses develop as a conse-quence of associations between the stockperson and aversive and rewarding elements of the handling bouts. The main aversive properties of humans, which will increase the animal’s fear of those humans, include hits, slaps and kicks by the stockperson, while the rewarding properties, which will decrease the animal’s fear of humans, include pats, strokes and the hand of the stockperson resting on the back of the animal. Negative tactile behaviours appear to be the main deter-minant of the commercial animal’s fear of humans although there is evidence in dairy cattle that shouting may increase fear of humans. Poultry appear to be sensitive to visual and auditory contact with humans, and observations at meat chicken farms indicate that rapid speed of movement by the stockperson is fear provoking for chickens. In relation to laying hens, both rapid speed of movement and noise made by stockpeople appear to be fear provoking, as well as the sudden and unexpected appearance of the stockperson.

The present chapter and Chapters 3 and 4 have considered the infl uence of human and animal factors on the productivity and welfare of farm animals. The next chapter will integrate and expand these discussions in order to identify opportunities to manipulate human–animal interactions to infl uence farm animal productivity and welfare.

© CAB International 2011. Human–Livestock Interactions, Second Edition120 (P.H. Hemsworth and G.J. Coleman)

6 A Model of Stockperson–Animal Interactions and their Implications for Livestock

6.1 Introduction

The relationships between fear of humans and productivity of livestock observed in commercial settings indicate opportunities to improve animal productivity and welfare. For example, identifying and consequently addressing the factors regu-lating fear of humans may enable reductions in fear, with possible improvements in the productivity and welfare of farm animals. Handling studies on farm ani-mals under laboratory conditions have shown that human factors are potentially infl uential in affecting the fear responses of commercial farm animals to humans. In particular, tactile and visual contact appear to be important in affecting fear responses. While genetic effects may be infl uential for naïve animals, experience with humans may ameliorate, maintain or exacerbate the initial fear responses of farm animals to humans.

Studies on stockpeople in the livestock industries have identifi ed some of the infl uential human characteristics that affect the behavioural response of farm animals to humans under commercial conditions. If fear responses are to be manipulated, it is these human factors and their antecedents that have to be targeted for improvement. In Chapters 3, 4 and 5 the infl uence of stockperson attitudes on stockperson behaviour and of stockperson behaviour on farm ani-mal behaviour and productivity were considered, and this chapter will integrate these fi ndings and present the key factors involved in the relationships between these stockperson and farm animal variables with the objective of presenting a model of human–animal interactions in livestock production. The chapter will form the basis of an introduction to the next two chapters which deal with the opportunities to change stockperson attitudes and behaviour utilizing the frame-work outlined in this model.

A Model of Stockperson–Animal Interactions 121

6.2 Relationships Between Human and Animal Variables

There is good evidence, based on handling experiments and observations in the livestock industries, of a sequential relationship between stockperson attitudes, stockperson behaviour, animal behaviour and animal producti-vity and welfare. This section will summarize and extend the previous discus-sion on these sequential relationships between human factors and key animal variables.

6.2.1 Stockperson attitudes and behaviour

Observations on stockpeople in the pig and dairy industries indicate that the attitudes of stockpeople towards interacting with their animals are predictive of the behaviour of stockpeople towards their animals (Section 4.5). Question-naires were used to assess attitudes of the stockpeople on the basis of the stock-people’s beliefs about their behaviour and the behaviour of their animals. Positive attitudes to the use of petting and the use of verbal and physical effort to handle dairy cows and pigs were negatively correlated with the use of negative tactile behaviours, such as slaps, pushes and hits (Table 6.1). These correlations indicate that stockpeople were likely to use less negative tactile behaviour when

Table 6.1. Stockperson attitude–stockperson behaviour correlations in the livestock industries.

Correlation between positive behavioural beliefs and negative stockperson behavioura

Industry/Study Petting and behaviour Effort and behaviour

Pig industryHemsworth et al. (1989b) −0.61** −0.47**Data reanalysed from

Hemsworth et al. (1994a)−0.55** −0.12

Coleman et al. (1998) −0.20 −0.10

Dairy industryHemsworth et al. (1995) −0.47** −0.36*Data reanalysed from

Breuer et al. (2000)−0.55** −0.28

Hemsworth et al. (2000) −0.19* −0.15Waiblinger et al. (2002) −0.35 −0.50**

Chicken meat industryCransberg et al.(2000) – 0.13Hemsworth et al. (1996a) – −0.13

aSignifi cant correlations (* = P < 0.05 and ** = P < 0.01) indicate associations between the two variables. Attitudes assessed on the basis of behavioural beliefs and negative stockperson behaviour assessed by the use of negative behaviours, such as hitting, slapping and shouting when handling pigs and cows and fast speed of movement and loud noise when inspecting poultry.

122 Chapter 6

handling their animals if they believed that: (i) petting should be frequently used; and (ii) verbal and physical effort should be infrequently used when interacting with dairy cows and pigs.

These attitude–stockperson behaviour relationships have not been estab-lished in the meat chicken industry (Table 6.1). Attitudes towards positive tactile behaviours (petting) with chickens were not recorded in these studies because there is little opportunity for routine tactile behaviours between stockpeople and chickens in commercial settings. The correlation between attitude towards the effort required to move birds and stockperson behaviour was small; this is most likely because stockpeople do not routinely move birds as part of the production process. The only exception to this is that when the birds are very young, stock-people regularly move them to ensure that the birds are exposed to food and water. It appears that those stockperson behaviours that predicted bird behaviour are related to visual cues and include speed of movement. However, attitudes towards these kinds of stockperson behaviour were not assessed. This clearly illustrates the need to identify the stockperson behaviours used as the attitude objects that are relevant to the species being studied. The attitude questionnaire should specifi cally target these behaviours. In one of the chicken studies (Hems-worth et al., 1996a), positive beliefs about using behaviours likely to move birds when moving through the facility were moderately and positively correlated with the time the stockperson spent in the facility (correlation coeffi cient of 0.43, P < 0.05). As reported earlier (Section 4.5), Edwards (2009), in a recent study, found some signifi cant relationships between stockperson attitudes and behav-iour at Australian and US commercial egg farms. For example, negative attitudes to laying hens and working with hens were associated with the stockperson mak-ing more noise, moving faster and spending less time near the birds.

The generally consistent attitude–behaviour correlations in the dairy, pig and egg industries indicate that the stockperson’s attitude towards interacting with his/her farm animals may affect his/her behaviour towards these animals. These results demonstrate that one of the antecedents of stockperson behaviour appears to be the attitudes that the stockperson holds towards specifi c behav-iour, and this proposal is underpinned by Fishbein and Ajzen’s Theory of Reasoned Action (Ajzen and Fishbein, 1980). It is important to recognize that this theory proposes that the important dispositional factor in predicting behav-iour is attitude and that other dispositional factors, including personality, operate indirectly through attitudes.

6.2.2 Stockperson behaviour and animal behaviour

As reviewed earlier (Section 3.5.1), handling studies, predominantly with pigs and dairy cattle, indicate that the level of fear for humans by farm animals is markedly affected by tactile behaviours from stockpeople.

Handling studies on pigs have shown that negative tactile behaviours, imposed briefl y but regularly by humans, will produce high levels of fear for humans. For example, handling treatments imposed daily for as little as 15–30 s and involving brief slaps, hits or shocks with a battery-operated prodder

A Model of Stockperson–Animal Interactions 123

whenever pigs failed to avoid humans, consistently resulted in pigs showing marked avoidance of humans (Section 3.5.1). In contrast, brief positive handling, involving talking, pats or strokes whenever pigs approached, resulted in low fear levels. Furthermore, regular handling involving rapid and close approach by humans resulted in pigs showing a similar avoidance response to humans to that of pigs that received a shock from a battery-operated prodder (Section 3.5.2).

Observations in the pig industry (Section 5.4) indicate that the nature of human tactile behaviours is an infl uential factor affecting the behavioural response of commercial pigs to humans. The percentage of negative tactile behaviours to the total tactile behaviours by the stockperson was positively cor-related with the level of fear of humans in pigs, as assessed by the time pigs spent near a stationary experimenter (Table 6.2). Negative tactile behaviours include moderate-to-forceful slaps, hits, kicks and pushes, while the positive tactile behaviours include pats, strokes and the hand resting on the pig’s back. Thus, pigs were most fearful of humans at farms in which stockpeople used a high percentage of negative tactile behaviours in handling the animals (Fig. 6.1).

As in the pig industry, negative tactile behaviours by stockpeople may regu-late the fear responses of commercial cows to humans. The use of negative behaviours by stockpeople was positively correlated with the level of fear of humans, as assessed by the approach or avoidance behaviour of cows to a sta-tionary experimenter in a standard test (Table 6.2). The main negative behav-iours were pushes, slaps, hits, tail twists and shouts while moving cows in and out of the milking facility and into position in the facility for milking, while the main positive behaviours were pats, strokes, hand on the cow’s fl anks or legs during milking and talking. Thus, cows were most fearful of humans at farms

Fig. 6.1. A high percentage of negative behaviours used by stockpeople, such as kicking or slapping, will increase the commercial pig’s fear of humans.

124 Chapter 6

in which stockpeople frequently used these negative behaviours in handling their cows.

Handling experiments have also shown that cows will quickly learn to avoid humans using negative behaviours (Section 3.5.1). Brief exposure to handling treatments involving restraint with a nose tong, slapping and brief shocking with a battery-operated prodder resulted in heifers and cows rapidly showing avoid-ance responses to humans in comparison with handling treatments involving talking, brushing and feeding. Moderate or forceful slaps imposed briefl y when-ever heifers failed to avoid the approach of humans subsequently resulted in reductions in their approach to humans and an increase in their fl ight distance from approaching humans in comparison with positive handling involving talk-ing, patting and stroking. Similarly, positive handling of cattle involving the imposition of positive tactile behaviours, such as pats, strokes and brushing (Fig. 6.2), have been shown to reduce avoidance of humans in a range of testing situations designed to assess the fear responses to humans (Boissy and Bouis-sou, 1988; Boivin et al., 1992).

As with cattle, positive handling of sheep will reduce their fear of humans. Several studies on adult sheep have shown that talking, stroking and feeding sheep reduced their avoidance of approaching humans (Section 3.5.1).

Studies with poultry have shown that chickens and laying hens are particu-larly sensitive to visual contact with humans (Sections 3.5.1 and 3.5.2). Treat-ments involving a human placing his/her hand either on or in the chicken’s cage and allowing birds to observe other birds being handled have been shown to

Table 6.2. Stockperson behaviour–animal behaviour correlations in the livestock industries.

Industry Study

Negative stockperson behaviour and level of

fear for humansa,b,c

Pig Hemsworth et al. (1989b)d 0.45*Data reanalysed from Coleman et al. (2000)e

0.40*

Dairy Breuer et al. (2000)f 0.30Hemsworth et al. (2000)g 0.33**Waiblinger et al. (2002)h 0.47**

Meat chicken Hemsworth et al. (1996a)i 0.32Cransberg et al. (2000)j 0.43*

Egg (hens) Edwards (2009)k 0.59**

aSignifi cant correlations (* = P < 0.05 and ** = P < 0.01) indicate associations between the two variables. bNegative stockperson behaviour was assessed in studies: d,e,f,hby the percentage of negative tactile behaviours such as hits and slaps; gby the frequency of negative tactile behaviours such as hits and slaps; i,jby speed of movement; and kby the noise made by the stockperson.cFear of humans was assessed in studies: d,e,fby the time animals spent near a stationary experimenter; gby the percentage of cows that approached close to a stationary experimenter; hby the distance at which cows showed avoidance to an approaching experimenter; i,jby the number of meat chickens that remained close to an experimenter moving in standard manner among the chickens; and kby the proportion of hens that remained close to an experimenter.

A Model of Stockperson–Animal Interactions 125

reduce the subsequent avoidance of humans shown by young chickens. Interest-ingly, close visual contact without tactile contact was more effective in reducing fear than picking up and stroking the bird, suggesting that this tactile handling by humans may contain aversive elements for birds, such as active interaction and tactile interaction. Handling studies on laying hens also demonstrate the infl uen-tial effects of visual contact with humans on the fear responses of birds to humans. Regular close visual contact with humans, involving positive elements such as slow and deliberate movements by humans, reduced the subsequent avoidance behaviour of mature laying hens to humans in comparison with minimal human contact or visual human contact involving aversive elements such as fast speed of movement and unexpected human contact.

Observations on stockpeople and birds at meat chicken farms reveal that the visual cues from the stockperson may regulate the fear responses of commercial birds to humans (Section 5.4). The speed of movement of the stockperson was correlated with the level of fear for humans by birds (Table 6.2). Frequency of tapping on objects by the stockperson while moving through the facility was positively associated with fear levels but, surprisingly, frequency of waving by the stockperson as he or she moved through the facility inspecting birds and equip-ment was negatively, not positively, associated with avoidance by birds of humans. These correlations indicate that birds are most fearful of humans at farms in which stockpeople move quickly through the facility, frequently tapping on objects in the facility and infrequently waving. It is possible that waving by the stockperson, which intuitively appears to be fear provoking, may be either rewarding, mildly fear provoking resulting in rapid habituation of the fear responses to humans, or a consequence of less fearful birds remaining closer to

Fig. 6.2. Positive interactions by stockpeople, such as the hand on the cow’s back, will reduce the commercial cow’s fear of humans.

126 Chapter 6

the stockperson as he/she moves slowly through the facility, necessitating the frequent use of this behaviour to move birds from under the stockperson’s feet. Furthermore, this distinct behavioural pattern may alert birds to the imminent approach of the stockperson and reduce the chances of unexpected exposure, which in itself may be fear provoking. In contrast, fast speed of movement appears to be fear provoking, and the tactile contact that accompanies fast speed of movement by the stockperson and the corresponding avoidance responses (fl ight and vocalization) by birds that receive tactile contact from the stockperson may exacerbate these fear responses throughout the fl ock.

In recent observations on stockperson and bird behaviour at Australian and US commercial egg farms (Section 5.4), noise made by stockpeople during rou-tine inspection and maintenance in the accommodation facility was associated with birds showing increased avoidance of humans. In contrast, close approach to and tactile contact with cages or hens by stockpeople were associated with birds showing reduced avoidance of humans. These associations indicate that fear of humans was high at farms in which stockpeople frequently made noise and infrequently closely approached and had tactile contact with birds and cages.

These studies with farm animals indicate that conditioned approach–avoidance responses develop as a consequence of associations between the stockperson and the rewarding and aversive elements of the handling bouts. For pigs and cattle, the main aversive properties of humans, which will increase the animal’s fear of humans, include hits, slaps and kicks by the stockperson, while the rewarding properties, which will decrease the animal’s fear of humans, include pats, strokes and the hand of the stockperson resting on the animal. Poultry appear to be particularly sensitive to close and regular visual contact with humans and, indeed, positive visual contact may be more effective in reducing levels of fear for humans than human tactile contact. In contrast, rapid speed of movement by the stockperson, sudden and unexpected appearance of the stock-person and noise made by the stockperson may be fear provoking for poultry. As noted in Chapter 3, little is known of the effects of other forms of human contact.

6.2.3 Animal behaviour and animal productivity and welfare

Handling treatments inducing high levels of fear of humans have been shown to reduce the growth and reproductive performance of pigs (Section 3.6.1). Furthermore, observations in the pig industry have revealed negative correla-tions, based on farm averages, between fear of humans and reproductive per-formance of pigs (Table 6.3 and Section 3.6.1). These correlations indicate that the productivity of pigs was lowest at farms in which pigs were most fearful of humans. It is noteworthy that Hemsworth et al. (1989b) found that variation in fear of humans accounted for about 20% of the variation in reproductive per-formance across the study farms, indicating the importance of fear for humans on pig productivity.

The mechanism responsible for these adverse effects of high fear on produc-tivity appears to be a chronic stress response, because, in a number of experiments

A Model of Stockperson–Animal Interactions 127

on pigs, handling treatments that resulted in high fear levels also produced either a sustained elevation in the basal concentrations of the stress hormone cortisol or enlargement of the adrenal glands, together with depressions in growth and reproductive performance (see Section 3.6.1).

As also reviewed in Chapter 3 (Section 3.6.2), handling of a negative nature has generally been shown to reduce the growth performance of chickens. Posi-tive handling of young chickens, involving gentle touching, talking and offering feed improved growth rates and feed effi ciency and increased resistance to infec-tion more than either minimal human contact or negative handling involving shouting and banging on cages. Positive handling of adult laying hens, involving slow and deliberate movements, reduced avoidance of humans and resulted in higher egg production than handling that involved minimal and at times startling and unpredictable human contact. These results and those of other studies pro-vide support for the proposition that handling treatments that increase fear of humans will depress the productivity of poultry.

Field observations on poultry have found signifi cant negative relationships, based on farm averages, between the level of fear of humans and the productiv-ity of commercial meat chickens and laying hens (Table 6.3 and Section 3.6.2). Both egg production of laying hens and the effi ciency of feed conversion of meat chickens were generally negatively correlated with the level of fear for humans by birds: high fear levels were associated with low productivity. In contrast, Edwards (2009) found a positive correlation between the level of fear for humans and the productivity of commercial laying hens. There is no obvious explanation for this single contrary fi nding.

Table 6.3. Animal behaviour–animal productivity correlations in the livestock industries.

Industry StudyLevel of fear for humans and animal productivitya,b,c

Pig Hemsworth et al. (1981b)d −0.51*Hemsworth et al. (1989b)e −0.55*Hemsworth et al. (1994a)f −0.01*

Dairy Breuer et al. (2000)g −0.46*Hemsworth et al. (2000)h −0.27

Meat chicken Hemsworth et al. (1994b)i −0.57**Cransberg et al. (2000)j −0.10Hemsworth et al. (1996a)k −0.49*

Egg (hens) Barnett et al. (1992)l −0.58*Edwards (2009)m 0.43*

aSignifi cant correlations (* = P < 0.05 and ** = P < 0.01) indicate associations between the two vari-ables. bFear of humans was assessed in studies: d,e,f,gby the time pigs or cows spent near a stationary experimenter; hby the percentage of cows that approached close to a stationary experimenter; i,j,kby the number of meat chickens that remained close to an experimenter moving in standard manner among the chickens; and l,mby the proportion of hens that remained close to an approaching experimenter.cProductivity variables were number of piglets produced per sow per year, milk yield per cow, feed conversion in terms of ratio of feed to gain in meat chickens and eggs produced per day per hen.

128 Chapter 6

Observations on commercial dairy cows also indicate the existence of a sig-nifi cant negative fear–productivity relationship (Table 6.3 and Section 3.6.3). The approach behaviour of dairy cows to humans was positively correlated with the milk yield of the farm, indicating that milk yield was lowest at farms in which cows were highly fearful of humans.

Negative handling resulting in increased fear of humans may depress produc-tivity in dairy cattle (Section 3.6.3). Several studies have shown that negative han-dling increases residual milk in cows and tends to reduce milk yield in heifers. Furthermore, in comparison with minimal human contact, positive handling improves growth rate and feed conversion of veal calves.

As discussed earlier (Sections 3.7 and 5.6.2), fear of humans may also reduce the welfare of farm animals. The concern for the welfare of fearful ani-mals arises because fear is generally considered an undesirable emotional state of suffering (Brambell et al., 1965; Jones and Waddington, 1992); fear may also lead to injuries in animals trying to avoid humans during routine inspections and handling; furthermore, there is the evidence that when regularly interacting with humans, fearful animals may experience a chronic stress response and immuno-suppression, which in turn may have serious consequences on the growth and health of the animals.

Therefore, handling studies and observations in the livestock industries on fear–productivity relationships indicate that high levels of fear of humans may limit the productivity and welfare of farm animals. One possible mechanism responsible for these effects is a chronic stress response, as in a number of handling studies, a sustained elevation in corticosteroids was found in animals showing high fear levels. The human–animal relationship thus may have practical implications for farm ani-mals in production systems in which there are close or frequent human–animal interactions. In more extensive management settings, chronic effects of negative human contact on the behaviour, physiology and productivity of livestock may be less likely because of less contact with humans. Nevertheless, acute effects are likely in situations in which fearful animals are in close contact with humans.

6.3 Model of Human–Animal Interactions in Livestock Production

As a consequence of this research on human–farm animal interactions, the fol-lowing model of human–animal interactions in livestock production has been proposed (Hemsworth et al., 1993). Because a stockperson’s behaviour towards animals is largely under his/her control, this behaviour is strongly infl uenced by the attitudes that the stockperson holds about the animals. These attitudes and consequent behaviours predominantly affect the animal’s fear of humans, which, in turn, affects the animal’s performance and welfare. The mechanism whereby fear affects performance and welfare appears to be through a chronic stress response. These sequential relationships between human and animal variables are depicted in Fig. 6.3.

Based on handling studies, there is evidence to implicate a causal basis for the relationships between stockperson behaviour, animal fear and animal pro-ductivity, and this aspect will be considered further in the next chapter. This

A Model of Stockperson–Animal Interactions 129

sequential relationship is not necessarily a surprise to some, particularly those working in the livestock industries, but what is a surprise is the magnitude and consistency of the relationships across a number of industries. The fact that some studies in the dairy, pig and poultry industries indicate that fear of humans may account for a quarter to a third of the variation in productivity seen across farms in these industries demonstrates the importance of human–animal interactions on farm animal productivity.

In our earlier discussion of the relationship between attitudes and behaviour, we mainly considered the role of attitudes as dispositional factors in determining behaviour where such behaviour is under the volitional control of the person. However, as we shall see in Chapter 7, the behavioural situation feeds back on attitudes. So, for example, the mere fact that a stockperson behaves in a particu-lar way will tend to reinforce that stockperson’s attitudes towards that behaviour. In addition, the outcome of the behaviour will also feed back on the stockper-son’s attitudes. Thus, if a stockperson has a negative attitude towards handling pigs and behaves negatively toward pigs making them more diffi cult to handle (because of increased baulking, fl eeing, slipping, etc.), the pigs’ behaviour will reinforce the stockperson’s original attitudes.

The feedback loop in the model provides an opportunity to modify stock-person attitudes and behaviour. In Chapter 7, we will discuss in detail how this feedback loop can be used to train stockpeople by targeting these key attitudes and behaviour simultaneously.

The major task of stockpeople in modern animal production systems is to care for farm animals by managing their social, climatic, nutritional and health requirements to ensure their optimal productivity and welfare. Such a responsi-bility requires stockpeople to have appropriate technical knowledge and skills to provide the animal in a timely and effi cient manner with the correct conditions to survive, grow and, if in a reproductive stage, reproduce at a high level. There-fore, it is obvious that the stockperson’s knowledge and skills in animal produc-tion, together with the motivation to utilize these attributes (work motivation), are important determinants of the productivity and welfare of his/her stock. Man-agers and employers of stockpeople can infl uence the knowledge and skills of their stockpeople by encouraging and providing opportunities for their staff to participate in external and internal training programmes, while also providing

Stockperson

Attitudes Behaviour Fear

Animal

Productivityand welfare

Stress

Fig. 6.3. A model of human–animal relationships in the livestock industries.

130 Chapter 6

the working conditions and rewards to encourage their staff to utilize their knowl-edge and skills. Job satisfaction is widely regarded as an important factor affect-ing work motivation. De Alencar et al. (2007) found that mortality of broiler chickens was lower at farms where stockpeople reported a favourable relation-ship with the birds and a liking for the job. Stockpeople with these favourable relationships with their birds and with high job satisfaction visited the facility more often, thus being more likely to observe any problems before they developed too far. This pathway in which a stockperson’s technical knowl-edge and skills may infl uence animal productivity and welfare can be addressed by the provision of appropriate training courses for stockpeople. Developments in defi ning the necessary knowledge and skills required in each animal industry, and aligning remuneration and promotional opportunities with these aspects, will not only promote technical training for stockpeople but also provide a range of benefi ts, such as a recognized career and remuneration pathway, while promoting self-esteem in stockpeople and the professional-ism and image of the industries.

In trying to integrate the stockperson attitudes and behaviour with these other job-related characteristics, it is not too diffi cult to recognize that the atti-tude of the stockperson to the subject of his work, the animal, may infl uence a number of these other infl uential characteristics and thus the work performance of the stockperson. For example, the attitude of the stockperson towards the animal may affect such job-related characteristics as work motivation, motiva-tion to learn new skills and knowledge about the animal and job satisfaction, which in turn may affect work performance of the stockperson. In many indus-tries outside agriculture, motivating factors that appear to be important determi-nants of job satisfaction and thus, in turn, work motivation have traditionally been considered to include achievement, recognition, responsibility, the work itself and advancement. Other factors, often called hygiene factors, which may not actually increase job satisfaction but, when they are at suboptimal levels, act to depress job satisfaction, include company policy, pay, working conditions and benefi ts. Therefore, it is clear that if the stockperson’s attitude towards inter-acting with farm animals (the subject of the stockperson’s tasks) is poor, in con-junction with the consequent effects that this may create, (such as handling, production and welfare diffi culties), the stockperson’s job satisfaction is likely to deteriorate, with adverse consequences on work motivation. Furthermore, if the stockperson’s attitude towards the animal is poor, the stockperson’s commit-ment to the surveillance of and the attendance to production and welfare problems facing the animal is likely to deteriorate. Thus, the attitudinal and behavioural profi les of the stockperson may have marked effects on animal productivity and welfare, both via fear of humans by the animal and via work performance of the stockperson.

In fact, some research on 87 stockpeople at a large integrated piggery in Australia (Coleman et al., 1998) has indicated relationships between the stock-person’s attitudes and a number of job-related variables. In this study, four scores were obtained from questions relating to aspects of the job. The fi rst score, labelled ‘Job enjoyment’, comprised fi ve items including ‘How boring is your job?’ and ‘How long do you think you will continue in the pig industry?’ The

A Model of Stockperson–Animal Interactions 131

second score, labelled ‘Family’, came from two items: ‘When you go away for holidays, do you go with a member of your family?’ and ‘Who normally super-vises your children when you are not able to?’ The third score, labelled ‘Work breaks’, was based on two items: ‘How much do you look forward to tea and lunch breaks?’ and ‘How much do you look forward to the end of the working day?’ The fourth score, labelled ‘Working conditions’, was also based on two items: ‘I often have to work in cramped conditions’ and ‘The air is clean at work’. High scores meant high job satisfaction in all cases.

Two scores were also extracted from those questions relating to technical knowledge and willingness to learn. The fi rst score, labelled ‘Learning’, came from four questions including ‘How often do you discuss work methods during tea and lunch breaks?’ and ‘Would you attend training courses in your own time if they were available?’ The second score, labelled ‘Knowledge’, comprised three questions including ‘How much do you know about diseases in pigs?’ and ‘How much do you know about factors which affect reproduction in pigs?’ Low scores meant low technical knowledge and willingness to learn.

It was found that the willingness of stockpeople to attend training sessions in their own time (score labelled ‘Learning’) was correlated with attitudes towards characteristics of pigs and towards most aspects of working with pigs (Table 6.4).

Job enjoyment and opinions about working conditions showed similar rela-tionships with attitudes (Table 6.4). Thus, the stockperson’s attitudes may indeed be related to aspects of work apart from handling of animals. The possible inter-relationships between attitude towards animals and these other job-related char-acteristics are depicted in Fig. 6.4.

Other stockperson characteristics may also infl uence the performance of stockpeople in livestock production. In the study of 87 pig stockpeople by Coleman et al. (1998) described earlier, some modest relationships were found

Table 6.4. Correlations among attitude sub-scales and work-related variables in the pig industry (from Coleman et al., 1998).

Attitude sub-scalea,b

Work-related variable

Negative beliefaboutpigs

Negative behaviour

Working withpigs

Characteristics of pigs

Pigsas pets

Handlingnon-

oestrouspigs

Handlingoestrous

pigs

Job enjoyment −0.30* −0.24 −0.42** −0.46** −0.25 −0.20 −0.23Work breaks −0.42** −0.32* −0.10 0.00 −0.15 −0.09 −0.21Working

conditions−0.20 0.10 −0.32* −0.30* −0.03 −0.34* −0.25

Learning −0.31* −0.28* −0.39* −0.26* −0.32* −0.25* −0.16Knowledge −0.13 −0.02 −0.25 −0.12 −0.38* 0.02 −1.01

aSignifi cant correlations (* = P < 0.05 and ** = P < 0.01) indicate associations between the two variables.bHigh scores for the attitude sub-scales indicate a negative belief, while high scores for the work-related variables indicate a positive response.

132 Chapter 6

between empathy and stockperson attitudes, suggesting that empathy may infl u-ence the development of beliefs about pigs themselves and about handling pigs (Table 6.5). The ‘Feelings’ measure consisted of three questions including ‘How do you feel when pigs are injured?’ and ‘To what extent do you think weaners are cute?’ The ‘Reactivity’ sub-scale comprised three questions including ‘How much do pigs react to you?’ and ‘I get very upset when I see an animal in pain’. As shown in Table 6.5, there was a general pattern of negative correlations between empathy and negative attitudes towards pigs. This fi nding is consistent with Ajzen and Fishbein’s (Ajzen and Fishbein, 1980) assertion that empathy, as a characteristic of a person, is an antecedent to attitude rather than a direct determinant of behaviour.

Stockperson

Attitudes Behaviour

Jobsatisfaction

Work motivation

Technical skillsand knowledge

Motivationto learn

Stockpersonwork

performance

Fear

Animal

Stress

Productivityand welfare

Fig. 6.4. The relationships between stockperson attitudes and behaviour and other job-related characteristics.

Table 6.5. Correlations among attitude sub-scales and empathy in the pig industry (from Coleman et al., 1998).

Attitude sub-scalea,b

Empathy variable

Negative belief about

pigsNegative behaviour

Workingwithpigs

Characteristics of pigs

Pigsas

pets

Handlingnon-oestrous

pigs

Handlingoestrous

pigs

Feelings −0.26 −0.27 −0.18 −0.48** −0.32* −0.12 0.01Reactivity 0.08 −0.31* 0.03 −0.42** −0.20 −0.35* −0.30*

aSignifi cant correlations (* = P < 0.05 and ** = P < 0.01) indicate associations between the two variables.bHigh scores for the attitude sub-scales indicate a negative belief, while high scores for the empathy variables indicate a positive response.

A Model of Stockperson–Animal Interactions 133

As reviewed earlier (Sections 4.3 and 4.4), personality is one of the key antecedents of attitude to animals, and while personality has been shown to be weakly correlated with stockperson behaviour, it is expected to infl uence behav-iour by affecting attitudes (Fig. 6.5). Similarly, empathy may be a factor under-lying the development of positive attitudes toward farm animals.

6.4 Conclusions

Research on the interactions between humans and farm animals has shown interrelationships between the stockperson’s attitudes and behaviour and the behaviour, productivity and welfare of farm animals, and a model of human–animal interactions in livestock production has been proposed, which can be outlined as follows. Because a stockperson’s behaviour towards animals is largely under volitional control, this behaviour is strongly infl uenced by the attitudes that the stockperson holds about the animals. These attitudes and consequent behaviours predominantly affect the animal’s fear of humans, which, in turn, affects the animal’s performance and welfare. The mechanism whereby fear affects performance and welfare appears to be through a chronic stress response. The stockperson’s attitude and behaviour may have either direct or indirect effects on other important job-related characteristics such as job satisfaction, work motivation and motivation to learn and, thus, also affect animal productivity

Fig. 6.5. Irrespective of the production system, the attitudes and behaviour of stockpeople to their animals have profound effects on these animals.

134 Chapter 6

and welfare. A less obvious risk to the welfare of farm animals arises in situations in which the attitude and behaviour of the stockperson towards the animals are negative because the stockperson’s commitment to the surveillance of, and the attendance to, welfare issues is most likely highly questionable.

The sequential relationships between human and animal variables indicate the opportunity to target stockperson attitudes and behaviour in order to improve animal productivity and welfare. Stockperson selection and training programmes addressing these key attitudinal and behavioural profi les appear to offer the live-stock industries potential to improve animal productivity and welfare, and this is the subject of the next two chapters.

© CAB International 2011. Human–Livestock Interactions, Second Edition (P.H. Hemsworth and G.J. Coleman) 135

7 Changing Stockperson Attitudes and Behaviour

7.1 Introduction

In the previous chapter, the research on the interactions between humans and farm animals was reviewed and a model of human–animal interactions in live-stock production was developed. Based on this research it is proposed that sequential relationships exist between the stockperson’s attitudes and behaviour and the behaviour, productivity and welfare of farm animals. There are also recip-rocal relationships between attitudes and behaviour: attitudes do not simply deter-mine behaviour but the outcomes associated with this behaviour feed back on the beliefs of the stockperson, with animal behaviour, for example, feeding back on the attitudes and thus behaviour of the stockperson. In addition to this direct pathway in which the stockperson’s behaviour affects animal performance and welfare, it is suggested that the stockperson’s attitudes may infl uence other impor-tant job-related characteristics of the stockperson, such as job satisfaction and job motivation, and thereby infl uence animal performance and welfare via this path-way. Studies on both experimental and commercial animals, particularly dairy cattle, pigs and poultry, support this model. Thus these sequential relationships between human and animal variables indicate the opportunity to target stockper-son attitudes and behaviour in order to improve animal productivity and welfare.

The aim in this chapter is to review the principles underlying attitude and behavioural change and to discuss the research in livestock production in which attitude and behavioural change has been attempted.

7.2 Factors Infl uencing Stockperson Attitudes and Behaviour

In order to understand how the attitudes and behaviour of stockpeople can be changed, it is necessary to appreciate the factors that infl uence the establishment and maintenance of attitudes and behaviour.

136 Chapter 7

7.2.1 Learning and attitude acquisition

Stockpeople, through both classical and instrumental conditioning, and in the context of varying motivational states, acquire their attitudes and behaviour towards the animals under their care. The processes of instrumental and classical conditioning in animals that were discussed in Chapter 5 also apply to human learning, including the acquisition of attitudes (Olson and Mitchell, 1975), and so experiences, particularly those when fi rst interacting with animals or fi rst observing others interacting with animals, are infl uential in the acquisition of attitudes and behaviour towards farm animals. Kerpelman and Himmelfarb (1971) demonstrated that attitudes, including the evaluation of objects, can occur through conditioning. Using partial reinforcement, and counter-condition-ing to the development of an attitude towards a hypothetical ethnic group, various reinforcement schedules were used to establish a positive attitude in 160 undergraduates. Some participants were exposed to a counter-conditioning regime to establish a negative attitude, while others simply fi lled out a question-naire to assess their attitudes. Whether their attitude was positive or negative varied directly with the amount of reinforcement. Furthermore, resistance to counter-conditioning was found to vary inversely with consistency of reinforce-ment. Interestingly, the attitude object had acquired a positive or negative evalu-ative meaning. This research clearly demonstrated that reinforcement schedules affect attitude formation and attitude change consistent with the learning theory principles outlined in Chapter 5. This has implications for training programmes that are designed to change stockperson attitudes because it demonstrates that attitudes can be both learned and ‘unlearned’. However, humans can also learn indirectly by observing behaviour in others rather than by direct conditioning processes. Such processes are particularly relevant to understanding the way in which attitudes and behaviours develop in stockpeople.

Kanekar (1976) proposed that attitudes can be acquired by conditioning without any fi rst-hand experience of the attitude object. This rests on the assump-tion that people can vicariously experience the emotions of others. Because these emotional experiences can act as reinforcers, mere observation of an atti-tude object and another person’s emotional response to that object can be suf-fi cient for the observer to develop an attitude by classical or operant conditioning as though the observer was directly involved in the experience. An example of this is as follows. Suppose that a person observes another person expressing a strong negative attitude towards a particular farm animal, say a pig, and at the same time shows clear disgust towards pigs. By classical conditioning, the observer would experience a pairing of the object ‘pig’ with the vicariously expe-rienced emotion of disgust. If this was repeated suffi ciently often, the observer would develop a negative attitude towards pigs.

The direct evidence in support of Kanekar’s view is scarce. Nevertheless, there is widespread evidence that people can and do learn attitudes through observation. Maio et al. (1994) carried out a study in which people were given information about a new immigrant group. Two variables were systematically varied. First, the immigrants were described in a way that made them either very relevant to the participants in the study, or largely irrelevant. Secondly, participants

Changing Stockperson Attitudes and Behaviour 137

were told about others’ emotional reactions to the immigrants. In two of the experimental conditions, one group was given descriptions of positive reactions, and the other group was given descriptions of negative reactions. When consis-tently positive emotional descriptions were provided and personal relevance was high, participants formed more positive attitudes than when consistently nega-tive emotions were provided.

It is clear, from the foregoing, that people can acquire attitudes indirectly without having any direct experience with the attitude object. Of course, direct experience with an attitude object can also lead to the development of attitudes through learning.

7.2.2 Effects of behaviour on attitudes

One of the major contributions of the theory of cognitive dissonance (Festinger, 1957), briefl y described in Section 4.2.5, was its emphasis on the reciprocal relationship between attitudes and behaviour. This refers to the fact that not only do attitudes infl uence behaviour, but also behaviour infl uences attitudes. Once a person carries out a particular behaviour, there is a tendency to modify those attitudes that are relevant to performing that behaviour. The classic study in this area was reported by Festinger and Carlsmith in 1959. What they did was to ask undergraduate students to spend an hour performing a very boring task. The students were then paid either nothing, $1.00 or $20.00 (a lot of money in 1957, when the experiment was done). Before being paid, each student who was to be paid was asked (persuaded really) to tell another person that the experiment had been very interesting and that the person would enjoy it. All students were subsequently interviewed and asked whether they enjoyed the experiment.

So, basically, students were told to tell another student that performing a very boring task was, in fact, not boring at all. In other words, those students who received either $1.00 or $20.00 were asked to lie about how enjoyable the task was. Those who were paid nothing were not told anything about the task. Curi-ously, those who were paid $1.00 agreed with the other student and said that the experiment was enjoyable, and that they would participate in a similar experi-ment. Those paid nothing or $20.00 said that they did not enjoy the experiment and were less willing to participate in the future. In other words, those paid noth-ing told the truth, and those paid $20.00 were not persuaded by the fact that the student had told them that the task was enjoyable.

Although the interpretation of the results becomes a little complicated, the explanation can be derived from the theory of cognitive dissonance. Basically, those students who were paid nothing and who had not been exposed to persua-sion felt comfortable in telling the truth about how boring the experiment had been. Those who had been paid $1.00 felt uncomfortable about this, because they had acted in a way that was inconsistent with their real beliefs. They there-fore changed their beliefs to be consonant with their actual behaviour. Those students who had been paid $20.00 were more prepared to see the payment as pressure and therefore that their behaviour was consistent with the amount of that pressure.

138 Chapter 7

While there has been much argument about the best theoretical explanation for the results of this study, there is a substantial body of evidence to show that there is a reciprocal relationship between attitudes and behaviour (Olson and Stone, 2005). If a person performs a behaviour that is inconsistent with his or her current attitude, then that person will tend to change the attitude to make it consistent with the behaviour. This has important implications for attitude change because changing attitudes should be most effective when the attitudes and the related behaviours are targeted concurrently. This approach was used in the cognitive behavioural training packages to be discussed later in this chapter (Section 7.4).

Most attempts to induce attitude change through the mass media are designed to effect a behavioural change in only a small percentage of the popu-lation. The idea is that a small percentage of people changing brand loyalty, for example, is suffi cient to produce a signifi cant profi t for a cigarette company. A discussion of how to induce change in targeted individuals is presented in Section 7.3. Before proceeding to that discussion, it is useful to have some insight into why individuals resist attitude change.

7.2.3 Resistance to attitude change

If an attempt is to be made to change the behaviour of specifi c individuals, then it is necessary to take into account explicitly those factors that may infl uence the change process. While there are many factors that may infl uence this process, a broad understanding of the principles governing such factors can be derived from Festinger’s theory of cognitive dissonance.

Festinger’s theory of cognitive dissonance has been applied to the develop-ment of an explanation for people’s resistance to attitude change (Festinger, 1957). It is normal for us to be exposed to a variety of information, some of which is consonant with our existing beliefs, and some of which is clearly dis-sonant. If we are to understand how attitudes are changed and what are the sources of resistance to change, then it is necessary to explore ways in which we manage, psychologically, discrepant or confl icting information.

Take, for example, the behaviour of a person who works in the pig industry. When entering the industry for the fi rst time, the person seeks out all the infor-mation relevant to the job. This includes a variety of things, including techniques for feeding, assisting with matings and farrowing (parturition), health manage-ment, handling techniques and so on. This information is then combined on the basis of expectancy value, that is, on the basis of the sum of the values of each attitude or belief relevant to a particular behaviour multiplied by how positively or negatively each behaviour is valued. The values of each behaviour are deter-mined by feedback from work mates, behaviour of the animals, supervisors’ comments and so on. Once a behavioural strategy is established, the person often engages in behaviour that may not be entirely rational to defend.

There are many specifi c strategies that an individual can use to resist attitude and behavioural change. While Festinger’s (1957) theory offers a general insight into how these strategies work, it is useful to look at some of the specifi c ways in which individuals resist change. A knowledge of these specifi c strategies can help

Changing Stockperson Attitudes and Behaviour 139

when designing a training procedure for experienced stockpeople, because such training does not simply involve the imparting of information, but involves the changing of established and habitual attitudes and behaviours.

One method of defending an established behaviour is to attempt to reduce that dissonance which comes from new information. Using the example from the pig industry, if the stockperson is told that even moderate negative behaviours affect pig handling and production, the stockperson may say ‘some pigs are dif-fi cult to handle and you need to use a moderate amount of force’. Here the person is reducing the importance of dissonant elements.

Another strategy to reduce dissonance is selective exposure. Selective exposure occurs when a person seeks out information that supports an earlier decision, and avoids information that is dissonant. For example, a person may refuse to discuss the effects of handling on pig reproductive performance. In this way, only new information supportive of the previous decision is processed.

Resistance to attitude change can also arise from what is termed ‘inocula-tion’ (McGuire, 1964). This refers to the process in which a person has been warned of the opposing view, and therefore has a prepared defence against it. Thus, a trainer may say to a stockperson, ‘When you go back into the piggery, some of your work mates will ridicule the things you have learned. In fact, you have learned some important husbandry techniques and your work mates will just be talking out of ignorance’. In this way, the stockperson has become inoculated against information counter to his/her newly acquired beliefs, and is therefore more resistant to change.

How strong the current attitude is will affect the amount of resistance to atti-tude change. If a person has a strong personal investment in a set of beliefs, then these beliefs will be resistant to change. For example, a stockperson may strongly believe that it is essential that farm animals be taught who is boss, otherwise they will tend to dominate the stockperson and be diffi cult or even dangerous to handle. This person, therefore, may resist attempts to persuade him or her that fewer negative behaviours are necessary.

Some personality variables may contribute to resistance to attitude change. The authoritarian personality discussed earlier in Section 4.2 is an example of this. The work of Adorno et al. (1950) on authoritarian personality showed that authoritarian people were reluctant to accept information from a person who was not seen to be an expert or to have high degree of authority. This was later extended by Rokeach (1960) to include dogmatism to describe people who hold very strong opinions, both positive and negative, who have a strong trust in authority and who are intolerant of disagreement. Thus, a stockperson high on dogmatism may resist attitude change because the trainer ‘has never worked in a piggery’ and is therefore not an expert.

Finally, some characteristics of the message and its source can affect resis-tance to attitude change. A major research programme conducted at Yale Uni-versity in the 1950s and early 1960s attempted to identify these factors (Janis and Hovland, 1959). These factors are summarized in Fig. 7.1.

When attempting to change the attitudes and behaviour of individuals, it is necessary to take into account all of these sources of resistance to change. We will discuss this in some detail in the next section.

140 Chapter 7

7.3 Changing Attitudes and Behaviour

As discussed in Section 7.2.2, advertising campaigns through the mass media are designed to induce change in only a very small percentage of the population. This is because the multitude of factors associated with the source of the information, the medium via which it is transmitted, the environment in which it is received and characteristics and prior experience of the individual, all serve to attenuate the impact of the message. Particular individuals are not the targets of such campaigns, because these campaigns work on the principle that only a small percentage of individuals will be behaviourally infl uenced by the campaign. When it comes to inducing behavioural change in individuals, the higher resources required need to be offset by a high success rate: techniques that merely deliver a message on a ‘take-it-or-leave-it’ basis are not appropriate. Approaches that take into account individual differences and that are targeted at the individual are more likely to lead to behavioural change (Coleman, 2010); this is now discussed in more detail.

7.3.1 Individual change

Inducing behavioural change at the level of the individual, rather than at the community level, involves processes somewhat different from those in the normal

Fig. 7.1. Factors that affect attitude change.

Source of information

Message characteristics

Effect of the message

Person characteristics

ExpertiseTrustworthinessLikeabilityStatusRaceReligion

Order of argumentsOpinion change

Perception change

Affect change

Action change

PersuadabilityInitial positionIntelligenceSelf-esteemPersonality

One-sided versus two-sidedType of appealExplicit versus implicit conclusion

Changing Stockperson Attitudes and Behaviour 141

classroom situation. Typically, in a classroom situation, the objective is to impart knowledge. This is usually accomplished by providing information in a variety of interesting ways and by reinforcing the knowledge acquisition process using var-ious motivational, assessment and feedback techniques. In the classroom situa-tion, no particular behavioural end point is used, students are merely expected to develop their conceptual and analytic skills and knowledge.

Inducing behavioural change not only involves imparting knowledge and skills, but also involves changing established habits, altering well-established atti-tudes and beliefs and preparing the person to handle reactions from others towards the individual following change. In other words, the process of inducing behavioural change is really a comprehensive procedure in which all of the per-sonal and external factors that are relevant to the behavioural situation, and that were outlined in the previous section, are explicitly targeted. Of necessity, the costs involved in such a process are substantially higher per individual than those in classroom techniques, but the magnitude of change can be expected to be much larger than that induced by classroom teaching.

To understand how attitudes and behaviour might be changed in individuals with a high success rate, cognitive–behavioural intervention techniques offer a powerful tool.

7.3.2 Cognitive–behavioural intervention

Cognitive–behavioural interventions are based on the idea that people have schemata (plural of schema) about a particular set of objects. For example, a person working in the pig industry may have a particular set of beliefs about pigs, including ‘they squeal, they have a distinctive smell, they are noisy’. These schemata also include affective components, that is, feelings about the objects that make up the schemata. Much of the cognitive–behavioural research in the general psychology literature has focused on treatment of people with inappropriate schemata. For example, people who inappropriately judge them-selves as incompetent or inferior may not be able to cope with routine work or family problems, may underperform and may avoid activities for which they feel inadequate. Turk and Salovey (1985a,b) argue that schemata induce peo-ple to selectively attend to the world around them. This means that such indi-viduals may behave inappropriately and that the schemata will be diffi cult to change.

However, attitude systems are examples of schemata, and normal attitude–behaviour relationships fi t the cognitive–behavioural model. As will be discussed later, cognitive–behavioural techniques are very useful in changing attitudes and behaviour in the agricultural industries.

Kendall and Hollon (1979) identifi ed two basic principles underlying cogni-tive–behavioural techniques. The fi rst principle was that cognitions are subject to the same laws of learning as are overt behaviours. These laws were discussed in Chapter 5. The second principle is that attitudes, beliefs, expectancies, attribu-tions and other cognitive activities are central to producing behaviour. These principles are quite consistent with all of our earlier discussions about the

142 Chapter 7

acquisition of attitudes and behaviour and about the complexity of factors that directly or indirectly infl uence the attitude–behaviour relationship.

Although cognitive–behavioural techniques were developed to deal with individuals suffering from various behavioural pathologies, for example, extreme fear of spiders or snakes, the principles also apply to non-pathological situations. As can be seen in Table 7.1, the use of cognitive–behavioural strategies appears to be entirely appropriate for use in normal situations where cognitive and behavioural change is required.

7.4 Modifying Stockperson Attitudes and Behaviour

Several studies have successfully applied cognitive–behavioural intervention techniques to improve stockperson attitudes and behaviour (Hemsworth et al.,1994a; Coleman et al., 2000; Hemsworth et al., 2002) and some of these will be described in detail in this section to illustrate the techniques and their implica-tions in livestock production.

7.4.1 Modifying the attitudes and behaviour of stockpeople at small independent pig farms

In the fi rst study (Hemsworth et al., 1994a), the cognitive–behavioural modifi ca-tion techniques involved retraining in behavioural areas as well as changing the attitudes and beliefs of people. Because of the reciprocal relationship between the attitudes and behaviour of the stockperson, and the equally strong relation-ships between the stockperson’s attitude and behaviour and pig fear and repro-ductive performance (Hemsworth et al., 1989b), the cognitive–behavioural intervention procedure targeted both the attitudes and behaviour of stockpeo-ple. This was achieved by aiming fi rst to improve the stockperson’s beliefs about pigs, particularly beliefs about handling pigs. This was done by providing stock-people with key information on commercial pigs, such as the ease with which they can and should be handled, their sensitivity to the range of negative behavioursused by stockpeople (and their sensitivity to stressors in general), and the adverse effects of these negative behaviours on their fear of humans which, in turn, can

Table 7.1. Cognitive–behavioural interventions in two situations.

Situation Target Approach

Behavioural pathology

Fear of snakes Information provisionModify inappropriate cognitionsSystematic desensitization

Stockperson behaviour

Inappropriate human–animal interaction

Provide factual informationModify inappropriate beliefsAnimal-handling skills training

Changing Stockperson Attitudes and Behaviour 143

have negative consequences on their productivity and ease of handling. The indi-rect effects of negative handling on the stockperson, such as a deterioration in job satisfaction via diffi culty in handling and closely inspecting fearful pigs, poor pro-ductivity and thus poor profi tability and depressed welfare in fearful pigs, were also raised with the stockpeople. In addition, the training provided stockpeople with information on the positive behaviours that can be used to reduce fear of humans.

Secondly, in order to address the behavioural aspects of the intervention, stockpeople were shown video footage of the behaviour of stockpeople in com-mercial units and emphasis was placed on those patterns, such as a high percent-age of negative behaviours, including moderately negatively behaviours, that have been shown to increase pigs’ fear of humans. Video footage of the behavioural responses of pigs to a range of stockperson behavioural patterns was also presented to assist stockpeople in recognizing and assessing fear responses in their animals.

Encouraging stockpeople upon their return to the piggery to practise immedi-ately actual handling techniques in a variety of situations was an essential part of this process. Behavioural modifi cation is necessary to ensure that there is a con-sonant change in beliefs and behaviour, so that the reciprocal relationship between these elements is maintained. In order to reinforce the information targeting improvements in both beliefs and behaviour, stockpeople were provided with written material in the form of a booklet, posters and a regular newsletter. The desired outcome of the study was to reduce the percentage of negative behav-iours used by stockpeople on commercial pigs (i.e. reduce the degree of aversive-ness of their behaviour towards pigs), and the following cognitive–behavioural intervention procedure, which incorporated the above principles, was used.

Thirty-fi ve commercial pig farms in Victoria and New South Wales, Australia, were selected. The farms varied in size from 75 to 300 breeding females and all breeding females were housed indoors on concrete, either in stalls or in groups during pregnancy and in farrowing crates during lactation. At each of these farms, one stockperson was predominantly responsible for conducting oestrus detection and supervising and assisting matings, and this stockperson was the sub-ject of study at these farms. In order to study the effects of the cognitive–behaviour modifi cation treatment, there were premodifi cation and postmodifi cation periods at each farm consisting of a minimum of 8 and 15 months, respectively. At the commencement of the premodifi cation period and the termination of the postmodifi cation period, observations and measurements were conducted on the attitude and behaviour of the stockperson at each farm responsible for mat-ing activities; observations were also conducted at each farm on the approach behaviour to an experimenter of recently mated gilts and sows (to assess level of fear of humans) in the standard test discussed in Section 5.4. Reproductive records for each farm were collected monthly during the two periods. At the end of the premodifi cation period, farms were randomly allocated, within level of fear strata, to two treatments, training and control.

The training treatment involved a 1-h individual session with each stock-person in which the cognitive–behavioural intervention procedure was used. During this session, stockpeople were fi rst shown the evidence for a relationship between stockperson attitude, stockperson behaviour, pig behaviour and pig productivity. This part of the procedure was designed to cover the cognitive

144 Chapter 7

aspects of the intervention. The evidence was derived from our on-farm studies and from the handling experiments described in Chapters 3 to 5. Graphs of the results from these studies, and the economic implications of the differences in productivity that could be attributed to stockperson behaviour, were discussed. Furthermore, the importance of minimizing negative behaviour to promote ease of movement by the pig and the infl uence of physical features in the environ-ment on pig movement were outlined. Stockpeople were given an opportunity to question and discuss the evidence, and if doubts or uncertainty were expressed, the relevant parts of the evidence were reviewed.

Following this information session, stockpeople were shown video footage of positive and negative behaviours displayed by stockpeople and emphasis was placed on the effect that these behaviours would have on conditioning the pig’s behaviour. This part of the procedure was designed to cover behavioural aspects of the intervention. It was emphasized that continual, even mildly negative behaviour towards the pig would tend to condition the pig to be fearful of the stockperson. An ‘inoculation’ procedure was used in which stockpeople were told that some negative behaviour might be necessary in some situations (e.g. a wary animal is reluctant to move out of a familiar area to an unfamiliar one), but that most of the negative behaviours in the industry were often avoidable and, in any event, should be greatly outnumbered by positive behaviours towards the pig.

In order to reinforce the information presented in this session, stockpeople were provided with posters to place in working areas of the piggeries. Regular newsletters were also used to summarize the important points of the cognitive–behavioural intervention programme and to prompt an assessment by the stock-person as to whether or not changes in his or her behaviour and the behaviour of his or her pigs were being achieved.

The control treatment involved two observers, the same two who imposed the cognitive–behavioural intervention procedure, visiting the farm for a 1-h ses-sion in which only general developments in the pig industry were discussed.

Stockperson attitudes and behaviour were assessed in the same way as in our earlier studies (see Chapters 4 and 5) and the behavioural response of pigs to humans was assessed in a standard test in which the approach behaviour of the pig to a stationary experimenter in an arena was measured.

Relative to the control treatment, the training treatment resulted in a greater increase in positive attitudes towards petting and talking to pigs (Table 7.2). Cor-responding with this relative improvement in the attitude of stockpeople at the training farms was a signifi cant reduction in the percentage of tactile behaviours displayed by the stockperson that were negative in nature.

These relative improvements in the attitude and behaviour of stockpeople at the training farms corresponded with a signifi cant relative reduction in the level of fear for humans by pigs at these farms (Table 7.2). There were increases in the time that pigs spent within 0.5 m of the experimenter and in the number of inter-actions with the experimenter in the standard test for pigs at the training farms relative to pigs at the control farms. Furthermore, there was a strong tendency for an increase in the number of pigs born per sow per year at the training farms relative to the control farms (a 7% increase in reproductive performance at the training farms, Table 7.2).

Changing Stockperson Attitudes and Behaviour 145

The intervention procedure used in this study was less intensive than that normally used in therapeutic environments, where a number of sessions over an extended period of time are used, and where the behavioural component would normally include actual behaviour sessions in which the person used role playing or even actual situations to learn the new behaviour patterns. Ideally, inter-ventions in this area should be conducted in a place where opportunities for behavioural practice are available and follow-up sessions should be used. The resources available for the research presented here did not permit such a com-prehensive programme. Nevertheless, despite these shortcomings, behavioural and attitudinal improvements were observed following intervention and the effects of these were refl ected in changes in pig behaviour. This suggests that interventions of this kind are a powerful tool for improving the productivity and welfare of commercial pigs.

The effects of the intervention on the reproductive performance of pigs, while in the expected direction, were not substantial in magnitude. This raises the issue of whether the results from this study provide support for the sequential model, described in Chapter 6, in which stockperson attitudes affect stockperson behaviour which, in turn, affects pig behaviour and performance. Had the results only shown an effect on stockperson attitudes and behaviour, no fi rm conclu-sions could be drawn, because such results could have been attributed to stock-people merely conforming to the expectations of the experimenters. However, the observed reductions in the level of fear of humans by the pigs at the training farms cannot be readily explained other than by the effects of changed stock-person behaviour. This, in conjunction with the small but marked improvement in reproductive performance of pigs at the training farms (7% relative to the control farms), strongly suggest a causal link between stockperson attitudes and behaviour on the one hand and pig behaviour and performance on the other.

Table 7.2. Summary of the effects of a cognitive–behavioural intervention procedure, applied at modifi cation farms, and targeting the attitudes and behaviour of stockpeople in the pig industry (from Hemsworth et al., 1994b).

Variable Control farms Training farms

Human attitudesa

Beliefs about petting and talking 89.2 102.9Beliefs about effort needed to handle 89.8 92.2Human behaviourNegative behaviours (%) 55.8 38.6Fear levelsb

Time spent near experimenter(s) 15.6 21.9Number of interactions with experimenter 1.3 2.0Reproductive performancePiglets born per sow per year 22.2 23.8

aHigh score indicates positive beliefs.bFear of humans was assessed by the time pigs spent near and number of interactions with a stationary experimenter in a standard test.

146 Chapter 7

Fear–reproduction relationships observed in the industry and the effects of hand-ling studies on the growth and reproductive performance of pigs further support this interpretation.

7.4.2 Modifying the attitudes and behaviour of stockpeople at a large integrated pig farm

A second study (Coleman et al., 2000) using cognitive–behaviour modifi cation techniques was conducted in a commercial piggery where large numbers of stockpeople worked together rather than on individual farms. The procedure used in this study was similar to that used in the fi rst study (Section 7.4.1). How-ever, a number of refi nements were made to improve both the effi cacy of the procedure and the practicability of the procedure for training stockpeople in a formal group setting. First, as before the procedure was designed to modify the beliefs of stockpeople about the sensitivity of pigs, handling pigs and the conse-quences of aversive handling on the ease of handling and productivity of pigs; secondly, it was designed to educate stockpeople to properly observe and handle pigs to avoid these adverse handling effects. An important component of this training procedure was a detailed series of reviews and discussions of the subject utilizing data and observations from our earlier studies, and emphasizing the important relationships between stockperson attitude, stockperson and pig behaviour, and pig stress and productivity. In addition, a booklet summarizing the procedure and specialized video footage, showing examples of appropriate and inappropriate behaviours by stockpeople and the accompanying behav-ioural responses by pigs, were presented. This video and written material were also used for revision by the stockpeople.

This intervention procedure was conducted on half of the stockpeople 1 month after the study commenced, and the remaining stockpeople were sub-jected to a control procedure. At the commencement of the study (i.e. pre-modifi cation for all stockpeople) and then 1 month later, the attitude and behaviour of all stockpeople and the behaviour of pigs assisted to mate by these stockpeople were assessed. Stockperson attitudes and behaviour were assessed in the same way as had been done in our previous studies. In this study, however, an attempt was made to assess the effects of the cognitive–behavioural interventions on other job-related variables as well. The questionnaire developed by Coleman et al. (1998) and described in the previous chapter (Section 6.3) was used to meas-ure a number of job-related measures such as job satisfaction and job knowledge.

In most of our previous research, the behavioural response of pigs to humans was assessed in the standard test described in Chapter 5 in which individual pigs were introduced to a stationary experimenter. It was not possible to conduct this test due to the amount of time needed and the costs of setting up specifi c arenas at various locations. A simpler test, which has been shown to be predictive of the above test in a commercial setting (Hemsworth et al., 1981b), was used in this study and basically involved measuring the withdrawal response of feeding gilts and sows to an experimenter approaching in a standard manner. These animals were housed in stalls and a sample of 30 pigs supervised and assisted to mate by

Changing Stockperson Attitudes and Behaviour 147

each stockperson were tested 3 weeks post-mating. Because, early in the study, stockpeople had diffi culties recording the identities of pigs they had assisted to mate, insuffi cient data were collected in the premodifi cation period to obtain reliable data from each stockperson. Nevertheless, the data were collated and analysed, and are presented with the qualifi cation that the premodifi cation data are unreliable because of the small numbers of animals assisted to mate by each stockperson.

Consistent with results from the previous study (Hemsworth et al., 1994a) (Section 7.4.1), stockperson attitudes improved following the training procedure (Table 7.3). In the training group, attitudes towards working with pigs improved, while those for the control group actually deteriorated. Similarly, beliefs about petting and talking to pigs improved for the trained stockpeople, while those for the control group became worse. Beliefs about handling oestrus pigs also improved in the trained group compared with the control group. No signifi cant changes were observed in job-related measures although there was a tendency for stockpeople in the trained group to report increased knowledge compared with those in the control group. It is interesting to note that all of the attitudes that showed improvement were in the area of behaviour. Given that the aim of the training programme was to change stockperson behavioural beliefs, this was a desirable outcome. As in the previous study, it is possible that changes in attitude score could refl ect the fact that stockpeople in the training group were aware of what was required, and simply answered in a way that was consistent with the experimenters’ expectations.

However, stockperson behaviour in the training group, which was assessed by different experimenters from those who supervised the attitude assessment, also showed a decrease following the training programme in both number and percentage of negative behaviours towards pigs relative to the control group. Furthermore, there was a strong trend, consistent with this improvement in stockperson behaviour, for pigs supervised and assisted to mate by stockpeople in the training group to show less withdrawal to an approaching experimenter. No such effect was evident in the control group. These changes in observed pig

Table 7.3. Summary of the effects of a cognitive–behavioural intervention procedure, applied at a large commercial farm, targeting the attitudes and behaviour of pig stockpeople (from Coleman et al., 2000).

Modifi cation group Control group

Variable Pre- Post- Pre- Post-

Beliefs about working with pigsa 12.6 11.8 12.5 13.1Beliefs about negative handlinga 17.3 16.2 17.6 18.7Beliefs about ease of handling sows

in oestrusa10.4 9.6 11.1 12.6

Percentage of stockperson negative behaviour

70.94 55.23 54.25 45.20

aLow score indicates positive beliefs.

148 Chapter 7

behaviour suggest that a genuine change in stockperson attitudes and behaviour had occurred. Therefore, despite the fact that the gilts and sows in this large com-mercial farm had been exposed to many stockpeople, there was some infl uence on their behaviour exerted by the stockperson who was involved in mating the gilts and sows. In the earlier study (Hemsworth et al., 1994a), one stockperson was predominantly responsible for supervising and assisting mating activities at each farm. Therefore, consistent with results from this previous research, stock-person attitudes, stockperson behaviour and, to a lesser extent, pig behaviour improved following the training procedure.

It is important to appreciate that regular handling of a predominantly posi-tive nature is required to reduce fear responses in pigs. Furthermore, the period required to reduce fear levels in those animals experiencing high levels of fear will be considerable, perhaps months (Hemsworth et al., 1981a). While changes in fear levels may be observed in the short term, a greater period of time may be required before stress responses, either acute or chronic, in those highly fearful animals are reduced to the extent where reproductive performance is not limited. It was not possible to monitor changes in the long term in this study, as in the earlier study (Hemsworth et al., 1994a), because stockpeople were regularly moved to different areas of production at the commercial farm.

The results of this study confi rm that stockperson attitudes and behaviour can be improved in a large commercial farm and that short-term effects on ani-mal behaviour can be observed. Given the peer pressure operating on individual stockpeople by work mates in the same unit, and the pressures that the unit manager would exert to produce conforming behaviour by stockpeople, this is a very encouraging result.

An interesting result that emerged from this study and that was not antici-pated when the study was commenced related to employee turnover. In the 12 months following the study, the group that had received training showed a 52% higher retention rate than did the stockpeople that received no interven-tion. This suggests that one effect of the modifi cation procedure was to improve long-term job satisfaction, even though no clear changes in job satisfaction were observed just 1 month after the modifi cation procedure.

7.4.3 Modifying the attitudes and behaviour of stockpeople in other livestock settings

Subsequent research (Hemsworth et al., 2002) using similar cognitive–behavioural modifi cation techniques has shown similar results in the dairy industry. Hemsworth et al. (2002) reported two experiments, one involving 29 commercial farms and the other involving 94 commercial farms, that examined the effects of a cogni-tive–behavioural intervention on stockperson attitudes and behaviour, cow behaviour and cow productivity. In both experiments, two treatments were imposed randomly: a training treatment, consisting of a cognitive–behaviouralintervention procedure designed to improve the attitude and behaviour of stock-people towards cows, and a control treatment, in which no intervention was attempted. In the fi rst experiment, an analysis of covariance, using variables mea-

Changing Stockperson Attitudes and Behaviour 149

sured during the previous lactation, demonstrated that stockpeople that had been trained held more positive beliefs about handling cows and used a lower number and percentage of negative tactile behaviours in handling cows than did stockpeople who had not been trained (Table 7.4). Further, cows at the training farms showed a shorter fl ight distances from humans, indicating a lower level of fear for humans by these cows compared to those from the control farms. How-ever, 36% of the training farms failed to show a reduction in average fl ight distance over the two lactations. Although there was no signifi cant treatment effect on milk yield, the training farms in which fear levels declined following the intervention had a signifi cantly higher milk yield than did the other farms.

In the second experiment, a signifi cant increase was found in the milk yield of cows following the training treatment (Table 7.4). Similar effects were observed on both milk protein and milk fat. An important difference between the fi rst and the second of these experiments on dairy farms was that the fi rst experiment used a face-to-face prototype intervention in which the experimenters delivered the training verbally. The second experiment utilized a multimedia programme that incorporated feedback from the fi rst experiment as well as input from the experimenters. Thus, the training programme for the second experiment was more developed and this may account for the larger and more general improve-ment in milk yield that was observed for the training group. As was the case for the pig industry, these results indicate that cognitive–behavioural interventions which successfully target the key attitudes and behaviour of stockpeople that affect cows’ fear of humans are able to improve the productivity and welfare of these animals. One unexpected outcome was obtained anecdotally from stock-people during debriefi ng after the experiment had been completed. A number of stockpeople who had undergone the training reported that they enjoyed milking the cattle more, that it was much quieter in the milking facility and that there

Table 7.4. Summary of the effects of a cognitive–behavioural intervention procedure, applied at commercial dairy farms, targeting the attitudes and behaviour of stockpeople (from Hemsworth et al., 2002).

Variable Control farms Training farms

Experiment 1 Stockperson attitudesa

Beliefs about effort needed to handle 27.7 32.0 Stockperson behaviour Negative behaviours (%) 80.6 40.1 Forceful negative behaviours (%) 10.1 2.6 Cow behaviour Flight distance (m) 4.5 4.2 Cow physiology Milk cortisol (nM/l) 2.05 1.40Experiment 2 Milk yield (l/year/cow) 550.6 579.82

aHigh score indicates positive beliefs.

150 Chapter 7

were fewer days when ‘everything seemed to go wrong’. One farmer who had expressed a prior aversion to milking the cows and who had previously chosen other jobs on the farm actually came to enjoy milking!

There are several studies that have, to a greater or lesser extent, shown similar results in different contexts. For example, as part of the Sixth Framework Welfare Quality Programme, Windschnurer et al. (2009b) investigated the effec-tiveness of a European cognitive behavioural training package tested on small dairy farms in Austria. Stockpeople were either allocated to training or control groups. Farms were visited twice to record stockperson attitudes towards dairy cows by means of a questionnaire and stockperson behaviour by means of behavioural observations during milking. Training farms were visited before and after the training. Compared with control stockpeople, the trained stockpeople showed an improvement in behavioural attitudes towards cows and towards handling cows, and an improvement in positive handling behaviours was found in the training group.

These researchers also conducted fi eld trials using cognitive–behavioural training on stockpeople in the pig, dairy and poultry industries in Europe (Ruis et al., 2010). The fi eld tests were carried out in the Netherlands (laying hens and pigs) and Austria (dairy cattle). Stockpeople were randomly allocated to training or control groups and all farms were visited twice. Stockpeople in the training group were trained within 2 weeks of the fi rst visit. The period between the two visits was approximately 6–8 weeks. Stockperson attitudes towards animals were determined by means of a questionnaire completed during the visits. Average scores were obtained for beliefs about animal characteristics (general attitudes) and handling situations (behavioural attitudes). Stockperson behaviour was assessed by means of behavioural observations during handling, and expressed as percentage of positive behaviours per unit or animal. Finally, the animal’s avoidance behaviour to the approach of an unfamiliar person was measured to assess fear of humans. A combined analysis was performed for the three species with stockperson as the replicate. Data for 64 stockpeople were fi rst standardized within each species to remove the effects of the species-specifi c units of measure-ment of each variable. Data were analysed by a 3 (species) by 2 (treatment group) analysis of covariance with the post-training score as the dependent vari-able and the pretraining score as the covariate. There was a signifi cant improve-ment in positive general attitudes and in positive behavioural attitudes towards animals for the trained group of stockpeople compared with the control group. Moreover, the percentage of positive behaviours towards animals increased sig-nifi cantly in the trained group compared with the control. The training did not signifi cantly affect avoidance behaviour to stockperson approach. The authors argued that the period between the training and the second visit may have been too short to result in an effect on animal fear. The time interval between the stockperson training and the assessment of animal behaviour in these studies was 4–6 weeks compared with between 3–6 months in the earlier Australian studies. This short exposure to any changes in stockperson behaviour would have had limited time to produce behaviour change in the animals. Nevertheless, a unique contribution made by these fi eld trials is that not only were the trials conducted in Europe, whereas the previous research had been performed in Australia, but

Changing Stockperson Attitudes and Behaviour 151

the training programmes were in several languages: English and Dutch (pig and laying hen programmes) and French, English and German (cattle programme).

There have been a number of other training approaches that are designed to either reduce stress in farm animals or modify the way that stockpeople handle animals. For example, Low Stress Stockhandling (Low Stress Stockhandling, 2005) is a commercial training programme designed to improve the stockperson–animal relationship. The aims of this programme are described in Table 7.5. The Low Stress Stockhandling programme appears to incorporate the principles of livestock handling developed by Bud Williams in the USA under the label of ‘Low stress handling methods’ (Bud Williams Stockmanship, 2001; Cote, 2004). It is claimed that low stress handling lowers animal stress and increases ease of animal handling (Bud Williams Stockmanship, 2001). Although there is little refer-ence in the scientifi c literature to these training approaches, it appears that they utilize some sound behavioural principles that have implications for animal stress and ease of handling, such as a consideration of the animal’s fear, exploratory and social behaviour as well as the animals’ sensory and cognitive ability (Hemsworth, 2007b). Grandin (2007) contends that calm cattle are easier to handle and sort than agitated, fearful cattle and that the secret to low-stress cattle handling is to keep them calm. However, at present there are no controlled studies that have evaluated the effectiveness of these programmes or similar programmes in improving wel-fare or productivity outcomes for the animals. Also, apart from the studies by Hemsworth et al. (1994a, 2002), Coleman et al. (2000) and Windschnurer et al.(2009b), none of the stockperson training programmes designed to either reduce farm animal stress or modify stockperson handling are based on theory-based, research-validated approaches that target both human and animal factors.

7.5 Review – Modifying Stockperson Behaviour

The results from the intervention studies reported in this chapter, taken in conjunction with the previous research on the sequential relationship between

Table 7.5. Aims of the Low Stress Stockhandling Programme (www.lss.net.au/index.htm).

Low Stress Stockhandling Mission• To foster an environment of low stress interaction between people and animals• To impart knowledge that promotes a positive attitude to low stress stockhandling• To show people the economic benefi ts of a low stress environment

Meeting the needs to handle stock in a calm and confi dent manner in all situations• Increase productivity and make more money• Improve meat quality from your livestock• Be more effective with your time and money • Improve management and profi tability • Reduce cost of production • Have quiet stress-free stock and people • Learn to work through ALL situations confi dently

152 Chapter 7

stockperson attitudes, stockperson behaviour, animal behaviour and animal pro-ductivity in commercial settings, and the research on handling livestock in exper-imental settings, indicate an important role for training stockpeople. Indeed, taken in conjunction with this earlier research, there is a strong case for introducing stockperson training courses in the livestock industries that target the attitudes and behaviour of the stockperson. The published intervention studies also demonstrate that such training is both practical and effective on a wide range of stockpeople working in a variety of situations. Although there is substantially less informa-tion available about livestock industries other than the pig and dairy industries, it is likely that these other industries would also benefi t from such training programmes.

It is important to reiterate the point that cognitive–behavioural interventions are effective because they are individualized. The programmes are self-paced to accommodate stockpeople who may not be used to formal learning situations or who may not be fast learners. The programmes also use multimedia techniques, including voice-overs, to limit any adverse impact of poor reading skills and to make the presentations interesting and engaging. Furthermore, the programmes include individualized feedback to participants on the basis of a questionnaire that they complete at the beginning of the training programme. All of these techniques are designed to overcome the limitations of mass communication techniques, including didactic classroom situations.

It is also important to emphasize that any evaluation of an intervention must include an assessment of the impact of the intervention on animal behaviour and productivity. It is not suffi cient to show that stockperson attitude and behaviour have improved following an intervention, because self-reports of this kind may be susceptible to social desirability effects and, because trained participants know what is expected of them, they may be on their ‘best behaviour’. Record-ing data on handled animals provides an independent measure of the effective-ness of the training as well as of its relevance.

In Chapter 2, we discussed at some length the need for stockpeople to be treated as professionals and to receive due recognition for the central role that they play in determining farm animal productivity and welfare. Training stock-people to have a sound factual knowledge base about human and animal behav-iour and to clearly understand and be able to apply sound handling techniques offers an important step towards meeting this need.

In the next chapter, we will review the material covered in this book and look towards the future for people working in the livestock industries. In particular, we will propose ways in which the research we have reviewed can be applied to industry for both the selection and training of stockpeople.

© CAB International 2011. Human–Livestock Interactions, Second Edition (P.H. Hemsworth and G.J. Coleman) 153

8 Conclusion: Current and Future Opportunities to Improve Human–Animal Interactions in Livestock Production

8.1 Introduction

We began this book by identifying the critical role of the stockperson in the management of farm animals and argued that the relationship between humans and the animals under their care was fundamental to their welfare and produc-tivity. To begin with we explored the nature of the stockperson’s job in terms of the duties that the stockperson performs, the status attached to the job and some of the issues that affect job satisfaction and job retention. We then turned our attention to the welfare of farm animals. In doing so, we looked at the rights of animals – their welfare and the ethical issues associated with our use of animals in livestock production. This was a key place to begin the book. Any discussion of the role of domestic animals in our society needs to take into account the rights of animals and our obligation towards them with respect to those rights. Also important is the need to optimize animal productivity and welfare and to recognize the place of the professional stockperson in the livestock industries, which then led to an account of human–animal interactions. This fi rst part of the book was designed to set the stage for a more detailed account of the empirical knowledge of human–animal interactions and their consequences for the live-stock industries, and the theory underlying this empirical research.

Human–animal interactions has been the topic of this book because there is an ever-increasing bo dy of evidence that demonstrates that these intera ctionsmay result in profound behavioural and physiological changes in the animal, with consequences on the animal’s performance and welfare. Furthermore, these interactions may also infl uence the stockperson to the extent that job-related characteristics, such as job satisfaction, motivation and commitment, may be affected, with implications for the job performance and career prospects of the stockperson. Notwithstanding, the translation of this evidence into agri-cultural practice is not widespread. In Australia, training packages to improve attitudes and welfare in the pig industry are widely used. Recent developments

154 Chapter 8

in Europe and the USA may see wider use in those regions too. However, uptake of such training in the other livestock industries and also the poultry industries has yet to occur. We will discuss some of the barriers to uptake later in this chapter. However, to begin, we will summarize our current understanding of the mechanisms for training stockpeople.

Understanding stockperson behaviour is a key to manipulating human– animal interactions to improve animal performance and welfare as well as the stockperson’s attitude and motivation to the job. An understanding of stock-person behaviour is necessary if we wish to infl uence or change stockperson behaviour. As discussed in Chapter 4, the Theory of Reasoned Action and the Theory of Planned Behaviour as proposed by Ajzen and Fishbein (1980) and Ajzen (1985) implies that behavioural change is ultimately the result of changes in beliefs about interacting with farm animals and the outcomes resulting from these interactions. This, in turn, implies that in order to infl uence behaviour, we have to expose stockpeople to information that will produce changes in their beliefs. Alternatively, we may wish to select stockpeople to work in the animal industries on the basis of their beliefs and thus their behaviour towards farm animals. The ability to manipulate these human–animal interactions, either via training stock-people in terms of their attitudes and behaviour towards their animals and/or by selecting stockpeople on the basis of these characteristics, has the potential to markedly and quickly infl uence animal welfare and performance. While training is the appropriate course of action for the existing workforce, selection tools may have limited utility where the pool of available recruits is also limited, as is the case in many rural communities. In this case, selection may be better used as a screen-ing tool to assist in identifying the nature and extent of training that may be neces-sary to ensure that the recruit is suited to the role of a stockperson. Another option is to completely automate the stockperson’s functions and thus avoid the deleteri-ous effects of adverse human–animal interactions. With the high cost of labour in many Western countries, this option may appear attractive to the livestock indus-tries. However, the trade-offs in terms of not only costs but also effect s on rur alsociety would need to be ca refully co nsidered. A utomated systems for all aspects of animal h usbandr y may not be pra cticable. For example , rou tine checking o fhealth and wel fare, moving animals from on e pl ace to a nother as part of normal farm prac tice and introduci ng ani mals or removi ng anima ls from the fa rm, m ayrequire som e direct human i nterventio n. Once this happens , ther e is an opportu-nity fo r the stoc kperson to infl uence animal pr oductivity and welfare. Th is chapter will examine the opportunities to improve not only the stockperson’s attitude and behaviour towards his or her animals, but also to improve the stockperson’s atti-tude and motivation to the job in order to improve human–anim al interactions and animal performance and welfare.

8.2 The Relative Importance of the Stockperson in Farm Production

The focus of production research and, recently, welfare research has been on factors such as housing, nutrition, health and climate. The clear justifi cation for

Conclusion 155

this research has been that paying attention to these factors will substantially improve the economics of farm production and the quality of the farm product. While these factors clearly affect farm production and, in fact, animal welfare, this focus has often been at the expense of recognizing and investigating the role of the stockperson. Research examining fear–productivity relationships in the dairy, pig and poultry industries has consistentl y shown that fear of humans is one of the single most important factors associated with variation between com-mercial farms in animal productivity (Section 6.2.3).

In Chapters 1 and 7, we discussed in detail the need for stockpeople to be treated as professionals and to receive due recognition for the central role that they play in animal productivity and welfare. It was emphasized that training stock-people to have a sound factual knowledge base about human and animal behav-iour, and to clearly understand and be able to apply handling techniques, offers an important step towards meeting this need. Furthermore, the opportunity to undergo training is likely to infl uence not only the skill and knowledge base of the stockperson, but also improve his/her self-esteem, job satisfaction and work moti-vation, with possible advantages in work performance and retention rates. It is not diffi cult to appreciate that, in addition to direct effects on fear and, therefore, effects on the productivity and welfare of the animals, improvements in the stockperson’s attitude and behaviour towards the animals may affect these other job-related characteristics through improvements in ease of handling, ability to closely super-vise and assist when necessary, and gains in animal performance and welfare.

Limited research has provided some evidence for the association of personality traits of the stockperson with animal productivity. For example, there is evidence in both the dairy and pig industries that certain personality traits may be associated with animal productivity. However, the causal basis of these relationships and the opportunities to manipulate any causal relationships need to be examined further. The available recent evidence on whether personality factors operate indirectly by infl uencing the formation and maintenance of stockperson attitudes or whether these factors have a direct effect on other aspects of the stockperson’s job perfor-mance suggests that personality operates indirectly by helping shape a person’s attitudes. However, because attitudes are l earned, they are suscept ible to change, notwithstanding the fact that the pers on’s p ersonality may infl uence his or her response to such attempt s. A good trainer should be ab le to take into account the idiosyncrasies o f the people being tra ined. Nevert heless, selectio n procedur es t hat include me asures of pers onality may offer opportunities to impro ve farm productiv-ity if a causal rel ationship between st ockperson personality and farm productivity can be identifi e d; an understanding of per sonality factors may then be useful in matching people to some types of jobs in agricul ture. As discussed in Chapter 1 (Section 1.4.2), the available evidence from the broader literature certainly does suggest that certain personality attributes are associated with job performance in some kinds of jobs. As work motivation and job commitment are infl uential in affecting job performance, there may also be opportunities to select stockpeople on the basis of their predicted work motivation and job commitment (Section 1.5).

However, bef ore exploring these aspects in too much detail , it is relevant to consider, as an alte rnative, the pos sibility of reducing or elimin ating the role of the stockperson in a nimal production entir ely. This is the substance of the ne xt sec tion.

156 Chapter 8

8.3 The Nece ssity of Human–Animal Interact ions

The necessity of hu man cont act in livestock product ion systems has been reviewed by Hemsworth and Gonyou (1997), and it is useful to use this previous discussion as a basis for a discussion on the possibility of eliminating all human–animal interactions in livestock production systems in order to eliminate the chances that these interactions may limit animal performance and welfare.

Modern livestock production involves several levels of interaction between stockpeople and their animals. Many interactions are associated with regular obser-vation of the animals and their conditions, and thus this type of interaction often involves only visual contact between the stockperson and the animals, perhaps without the stockperson entering the animals’ pen. Visual and auditory interactions may also be used to move animals, and in some industries tactile interactions may be used. All interactions contribute to the overall relationship that animals have with humans and determine whether that relationship is positive, neutral or nega-tive. Because of the potential for negative interactions by the stockperson, Hems-worth and Gonyou (1997) have questioned the necessity for interactions between humans and animals. To examine this question, it is necessary for consider the nature of human–animal interactions in livestock production in more detail.

An essential role of stockpeople in achieving high animal performance and welfare is the careful observation of animals under their care. Although animal conditions, such as ambient temperature, noxious gas levels and presence of feed and water, can be remotely monitored, direct observation of animals often pro-vides the fi rst evidence of departure from normality in animals. In particular, behavioural change can be utilized by stockpeople to identify abnormality, such as illness or stress. For example, an animal not feeding, a social animal voluntarily separated from the group or an animal that is unresponsive to environmental change, such as the approach of the stockperson, can be used to identify the early stages of a problem and enable a prompt response. Often the diagnosis of the problem by veterinarians relies on reports from the stockperson on the behaviour of the animal. For example, signs of pacing and kicking at its belly are indicative of colic in horses. One could argue that the prompt identifi cation of an impending problem for the animal relies heavily on the observations of the stockperson on animal behaviour. Use of video cameras is in general less effective than direct observation. For example, video cameras are limited by their inability to use localized non-visual cues, such as auditory cues (vocalizations), and observe the fi ne detail that is necessary to discriminate, for example, between a lesion and a smear of dirt on the skin. While animal movement can be quantifi ed with sophis-ticated tracking systems, a human observer is still necessary in many cases to make a decision on whether or not a video of an unusual behavioural pattern in an animal is indicative of a problem. Clearly, automation should be utilized to assist the stockperson in monitoring animals and their conditions. In fact, auto-mation of the tedious, laborious and objectionable tasks that may decrease job satisfaction, such as cleaning, provision of feed and water, etc., should be encour-aged. However, automation is unlikely, in the foreseeable future, to completely replace the stockperson. In the interests of animal welfare alone, the general pub-lic probably considers careful observation of animals under the stockperson’s care

Conclusion 157

as an essential part of good stockmanship. While not necessarily a legal require-ment, codes of practice for farm animals in many countries often state that daily observation of animals in confi ned conditions is essential.

Animals in most production systems have to be moved. Extensively grazed animals such as beef cattle and sheep are moved between pastures as part of optimal grass management, and extensively grazed dairy cows are moved sev-eral times a day during lactation to be milked. Growing pigs are generally moved from pen to pen in order to provide accommodation suitable to their stage of growth, and breeding pigs may be regularly moved according to their stage of the breeding cycle. It is during these situations that human–animal interactions have considerable potential to infl uence animal performance and welfare. As considered earlier in the book (Chapters 3 to 6), the risk of eliciting high fear of humans can be reduced by stockpeople minimizing their negative behaviours while maximizing their positive ones. This can be achieved by using negative behaviours, such as hits and slaps on dairy cattle and pigs, only when necessary and by seeking opportunities to use positive behaviours, such as pats, strokes and the hand of the stockperson resting on the animal (Fig. 8.1), when the opportunity arises. In addition, humans inadvertently interact with animals when they inspect the animals and equipment such as feeders and drinkers in their pens. Although these observations do not necessarily involve contact with the animals, these visual behaviours can also be potentially fear provoking. The avoidance of fast speed of movement and unexpected movement or appearance by the stockperson will assist in reducing fear of humans by poultry. Furthermore, reductions in shout-ing and increases in talking will assist in reducing fear of humans in dairy cows. An important opportunity to reduce fear of humans in farm animals is through regular exposure to humans in a neutral context, such as a human standing stationary and near the animals; habituation will occur over time as the animal’s fear of humans is gradually reduced by repeated exposure to humans in a neutral context (Fig. 8.2).

Human–animal interactions also occur in situations in which animals must be restrained and subjected to management or health procedures. Some animals may never be restrained during their lives, while others are restrained on a regu-lar basis. Some sort of restraint is used for weighing, milking, vaccinating and blood sampling, and animals are restrained for procedures that are painful, such as castration, branding, ear tagging and dehorning. It may be possible to reduce or eliminate some of these procedures, but others, such as vaccinations and blood sampling for diagnosis, are necessary to improve the health and thus the welfare of the animals, and some degree of discomfort or pain is justifi ed in the interests of the animal. Procedures such as milking and shearing are directly related to the reason the animals are kept, and could only be eliminated if the industry is closed. Weighing, ear tagging, castration and dehorning are justifi ed by facilitating management, improving product quality or reducing the possibility of injury to animals or humans. Although some reduction in these procedures may be possible, it is likely that they will remain part of animal care and production for some time.

The association of fear and pain from these husbandry procedures with humans performing them will increase the fear for humans that animals exhibit in other situations, such as during routine inspections. The effect these procedures

158 Chapter 8

have on the human–animal relationship relates both to the aversiveness of the procedure and the association by the animal of humans with that aversion. Rushen and colleagues (Munksgaard et al., 1995; Rushen et al., 1995; de Passille et al., 1996) have shown that performing an aversive treatment at a specifi c loca-tion, or by either an unfamiliar or familiar operator wearing different distinctive clothing, may avoid dairy cows associating the procedure with the regular stock-person. Rewarding experiences, such as provision of a preferred feed or even positive handling, at around the time of a stressful procedure may ameliorate the aversiveness of the procedure and reduce the chances that animals associate the punishment of the procedure with humans. For example, studies with pigs have shown that pigs will associate the rewarding elements of feeding with humans if handlers are present at feeding (Hemsworth et al., 1996d). Hutson (1985) found that although the effectiveness of food rewards diminished as the severity of the handling treatment increased, rewarding sheep with barley food improved

Fig. 8.1. The frequent use of positive behaviours by stockpeople, such as pats and the hand resting on the animal’s back, will decrease the pig’s fear of humans.

Conclusion 159

subsequent ease of handling in the location in which the aversive treatment was previously imposed. Surprisingly, daily injections were not highly aversive to pigs (Hemsworth et al., 1996b), and the authors suggested that there may have been some rewarding elements for the pigs in these handling bouts, such as the pres-ence of the handler and the opportunity to closely approach and interact with the handler before and after injection. Previous positive handling has been shown to improve ease of handling and reduce heart rates during loading of calves for transport (Lensink et al., 2001a,b), reduce heart rate and salivary cortisol con-centrations in lambs following tail docking (Tosi and Hemsworth, 2002), and reduce heart rates, kicking and restless behaviour in dairy cows during rectal palpation (Waiblinger et al., 2004). These data indicate that positive relation-ships with humans may ameliorate aversive experiences in farm animals when in the presence of humans. Further research is clearly required to understand the effects of the animal’s relationship with humans on how animals respond to stressful situations in the presence of humans because of the implications of these effects on animal welfare and ease of inspection and handling.

As Hemsworth and Gonyou (1997) have suggested, there are opportunities to reduce or even eliminate human involvement in some animal management procedures that are aversive to the animal. Examples include robotic shearing of sheep, robotic milking of cows and automated handling facilities for sheep (Syme et al., 1981) and pigs (Barton Gade et al., 1992). The effect of eliminating humans from such handling procedures on animal responses is well illustrated by

Fig. 8.2. Fear of humans can be reduced in pigs through regular exposure to humans in a neutral context, such as a human standing stationary and near the animals, followed by the opportunity for positive behaviours, such as talking and patting, as the animal approaches (photograph courtesy of Wageningen UR Communication Services).

160 Chapter 8

research on mechanical and manual harvesting of broiler chickens (Duncan et al., 1986). While maximum heart rates of birds caught by either method were similar, the rates remained high for longer in manually caught birds than in birds caught by a specially designed machine. In tonic immobility tests, manually caught birds showed a longer response than did machine-caught birds, possibly indicative of greater fear. Research on shearing sheep also indicates the welfare implications of eliminating humans from the procedure. Shearing appears to produce one of the most marked acute stress responses of all non-surgical hus-bandry procedures. The cortisol response to shearing has been found to be greater than many procedures such as stop-start transport, steady transport, shearing noise, isolation, jetting, yarding and blood sampling (Kilgour and de Langen, 1970; Fulkerson and Jamieson, 1982; Fell and Shutt, 1988; Harg-reaves and Hutson, 1990a,b). Using aversion learning techniques, in which ani-mals learn to associate a location with a specifi c treatment, Rushen and Congdon (1986) found that sheep developed an aversion to simulated shearing that was similar to the aversion developed to electro-immobilization (immobilizing the animal with a pulsed low-voltage electric current passed along the spine). Sur-prisingly, Mears et al. (1999) found that the cortisol and β-endorphin responses to shearing were of greater magnitude and of longer duration in previously shorn ewes than in naïve sheep, suggesting that rather than habituating to shearing the sheep may have been sensitized to shearing. Although the contribution of humans to the aversiveness of these procedures is generally unknown, these results indicate that the stressfulness of some procedures may be reduced by eliminating humans from the procedure or at least improving the human contact associated with the procedure. Similar opportunities may exist with other man-agement and health procedures and, clearly, research on these procedures should be conducted to determine the effects of the component of the procedure involving human contact on the animals’ responses. In situations where the human contact component is highly aversive or even injurious to the animal, procedures that eliminate human involvement or changes in the behaviour of the human should be sought. For example, because the method of catching laying hens in cages affects the incidence of bird injuries (Gregory et al., 1993), those techniques which minimize injury should be identifi ed and adopted.

Another potential problem for farm animals that have been deprived of human contact is the fact that if human contact is required, perhaps in an emer-gency situation, this interaction will be highly fear provoking and aversive because of the unfamiliarity with the handler. One could also question whether or not humans are viewed as social partners in the environment and whether or not they provide environmental variation, something that may be minimal in a totally automated system if not specifi cally addressed. If this were the case, auto-mated systems would present an impoverished environment relative to less automated systems. Indeed, several authors have proposed that zoo visitors may provide zoo animals with environmental enrichment (Hosey, 2008).

On balance, there appears to be a strong case for stockpeople to continue to interact with livestock. Even if an opposing case could be mounted, no cost–benefi t analysis of eliminating the stockperson has been carried out and, given the capital cost of many alternatives, it is unlikely that such elimination would be

Conclusion 161

practical. Therefore in order to maximize opportunities to benefi t from the stock-person’s input into farm production, the next two sections deal with managing this important resource in agriculture: training and selecting stockpeople to improve human–animal interactions.

8.4 Training Stockpeople to Improve their Interactions with their Animals

In the past, there have been limited opportunities for stockpeople to receive formal training in any aspect of their work. The new stockperson is t ypicallygiven a brief o rientation and is th en placed in the work environmen t where he o r she is expected t o learn ‘ on the j ob’. O n smal ler independent fa rms, the owner/operator will usually per form thi s func tion. On l arger farms, i t is u suallythe immediate s upervisor who will o versee this process whil e co-w orkers p rovideday-to- day feedback. We would sugges t that thi s practice refl e cts the belief t hatthe stockperson has a set of mec hanical functi ons to per form and, so l ong asthese are d one well, the farm w ill prod uce well. In general, where sy stematictraining is emplo yed, it us ually targets management rat her th an stockpeop le.

In A ustralia, fo r exam ple, the re have been m any attempts to deve lop te chni-cal training cours es s uitable for stoc kpeople. However, in an at tempt to s et a standard for su ch tra ining, the A ustralian National T raining Informat ion Service (NTIS) provides nationall y re cognized units of competency for a s pecific indu s-try, indu stry secto r or enterprise (NTIS, 2010). Th ese competencies focus on knowledge and skills as well a s language, literacy and numeracy, and occupa-tional health and safety requirements. The increasing units of competency in Australia lead to the qualifi cations of Certifi cate I, Certifi cate II, Certifi cate III, Certifi cate IV, Diploma or Advanced Diploma. One of the limiting factors for training stockpeople is that they usually have a restricted educational back-ground, may not handle for mal trai ning p articula rly well and, in s ome cases, may h ave lite racy problems. The fact that stockpeople m ay oft en come fr omsuch di sadvantaged grou ps means t hat training needs to be c arefully targeted and al so needs to be deliv ered in a wa y that i s accessible t o such individ uals.Appropriate targeting requires that th e key skills tha t a stockperson need s must be identifi ed a nd trainin g programmes devel oped that specifi cal ly relat e to those skills. A ccessible delivery means that th e train ing p rogramme s hould not be too for mally pr esented, should not be too the oret ical in cont ent and sh ould provide content to which stockpeople can ea sily relate.

An exampl e of su ch a tra ining program me is the on e deve loped by theauthors an d d escribed i n the pr evious cha pter ( Sections 7.4.1 and 7.4.2). Th isprogramme was spec ifi call y desig ned to target th ose attitudes an d b ehavioursof th e st ockperson that had a dir ect effect on p ig behaviour and pr oductivity. Tra ining stockpeople b y targeting (for i mprovement), both the key beliefs and behaviours tha t are infl uent ial in regulat ing hu man–animal interactions in a commerc ial setting offers the animal industries opportunities to improve animal performance and welfare. In the dairy and pig industries, these key human characteristics have been identifi ed and an intervention procedure based on

162 Chapter 8

cognitive–behavioural principles has been shown to be effective in improving human–animal interactions in these livestock industries.

The intervention procedure used as an experimental tool in the research described above was commercialized by the authors as a training package for the pig industry called ‘ProHand’ (‘Professional Handling of Pigs Program’). The ProHand approach has been extended into a package for dairy stockpeople and, following recent research at abattoirs, into packages for pig abattoir and red meat abattoir stockpeople in Australia. Training packages based on the ProHand prin-ciples have been developed by the authors and colleagues in Austria, France and the Netherlands for pig, dairy cattle, beef cattle and laying hen stockpeople under the label ‘Welfare Quality’. The main source of material for stockpeople undergoing these programmes is an interactive computerized program. A CD-ROM contains a detailed series of reviews and discussions on the subject of human–animal interactions, utilizing data and observations from key studies, and emphasizing the important relationships between stockperson attitude, stockperson and animal behaviour, and animal stress and productivity. Critical advice on changing attitudes and behaviour and maintaining these changes is also provided. A key feature of these CD programmes is that they provide inte-grated video material and a complete soundtrack so that there is not a str ongemphasis on the read ing skills of the intend ed participan ts. Typically, a dditionalmaterial inc ludes a book let that summarize s the ke y fa cts and recommenda tions(par ticularly on c hanging and main taining atti tudes and beha viour) and a video that shows exampl es of appropriate and inapprop riate behaviou rs by stockpeo-ple and the accomp anying behavioural responses by farm animals. This video and written material, together with a series of posters, is given to participants to use for reinforcement and revision at a later date. A video of a ‘question and answer’ session is also used by the trainers in the training sessions to address the most common questions raised by stockpeople. These programmes avoid the use of didactic classroom situations and attempt to personalize the learning experience and to make it accessible to people with limited classroom skills. For example, not only do stockpeople work through the programmes at their own pace, but they receive personalized feedback, relative to their industry peers, based on their responses to a built-in attitude questionnaire.

The most diffi cult aspect of the utilization of these programmes relates to their uptake in the relevant industries. As indicated earlier in this chapter (Section 8.1), there are numerous barriers to the uptake of these kinds of training packages. Despite the fact the original pig ProHand package was developed in 1996, it is only relatively recently that there has been widespread uptake in Aus-tralia and this has depended on strong support and coordination from the main pig body, Australian Pork Limited, and a willingness by the authors to invest substantial effort in promoting the package and training trainers. Much of the recent success of the programme may be a result of the introduction in Australia of a Code of Practice which requires that stockpeople should undergo formal training and/or be trained on the job within the fi rst 6 months of employment and that training should be conducted on a regular basis. Furthermore, there has been a progressive cultural change in the Australian pig industry which embraces the principle that good stockmanship is essential for good pig welfare.

Conclusion 163

In other countries, the USA for example, there has been an unwillingness to accept the credentials of the ProHand packages without local evaluations, despite the fact that although the US pig industry is much larger than the Australian pig industry, in most relevant respects it is similar. In the EU (European Union), where livestock production is on a smaller scale and often uses different housing systems from the Australian counterparts, the Welfare Quality programmes were developed as a way of addressing the differences in production approaches. Interestingly, the EU Welfare Quality programmes are strikingly similar to the Australian ProHand packages.

Perhaps the main barrier to the uptake of these training programmes relates to the animal welfare culture and the regulatory frameworks of the different coun-tries. In the EU, there is a well-established regulatory framework used to set animal welfare standards. There are EU directives for housing and transport, some of which were introduced in the late 1990s, which impose mandatory standards on member countries. In the USA, there is no consistent regulatory framework in regard to animal welfare, although, several initiatives were passed in Florida in 2002 and Arizona in 2004 prohibiting the use of gestation stalls for sows in both states and the use of veal calf crates in Arizona. A voter referendum was passed in 2008 in California that prohibits from the 1 January 2015 the confi nement of pregnant sows, veal calves and hens. Prior to these recent changes in state legisla-tion, food retailers were the primary drivers of animal welfare standards in the USA (Mench, 2008). While there is a culture of duty of care towards animals emerging in the EU and a regulatory framework emerging in Australia, the main pressure for improvement in animal welfare in the USA comes from animal rights groups, and usually results in changes to housing or welfare audits with no obvious emphasis on stockperson training, despite the fact that most welfare incidents that are reported in the media are a result of inappropriate behaviour by stockpeople.

There is clearly a signifi cant need for national initiatives to promote duty of care towards farm animals and to ensure that stockpeople are well schooled in their role and responsibilities. The rationale for this is as follows. The increasing interest by society in animal welfare has resulted in a corresponding scrutiny of animal use. We generally recognize that the relationships that develop between humans and most domestic animals in society are inevitably unequal, with basi-cally two principles applying to the management of animals in a range of animal uses, from individual pets to livestock production (Hemsworth et al., 2009). These principles are: (i) management to comply with the objectives of human profi t, benefi ts or pleasure; and (ii) management responsibilities refl ecting a duty of humane care of animals. In relation to livestock production and specifi cally to stockmanship, animal productivity is a key objective and consequently stock-people have an explicit responsibility to care for and manage their livestock to achieve effi cient animal performance. The second principle, that of manage-ment responsibilities refl ecting a duty of humane care of animals, is based on the widely held view in many societies that the use of animals by humans is accept-able provided that such use is humane (Mellor and Littin, 2004). Implicit in this view is that stockpeople have a responsibility to handle and care for their live-stock in a humane manner (Hemsworth et al., 2009). While technical skills and knowledge are important attributes of the work performance of stockpeople,

164 Chapter 8

as shown in this book, attitudes and behaviours of stockpeople towards farm animals are also important attributes.

8.5 Selecting Stockpeople to Improve their Interactions with their Animals

The possible benefi ts of selecting stockpeople appropriate to work with animals were discussed in the fi rst chapter (Section 1.6.2). As indic ated at the beginn ingof this chapte r, sel ection tools may hav e limited utility where the pool of availablere cruits is also limite d, and it m ay be appropriate to use selection tools to assist in identifying the nature and extent of training that may be necessary to ensure that the recruit is suited to the role of a stockperson. However, whether a selection tool is used to make a decision about the employability of a person or about the areas in which an applicant may need specifi c training, there is a need to have a well-validated procedure that can be used to improve the stockperson workforce.

There is a wide variety of tests that focus on personality, vocational prefer-ence, work motivation, etc. The specifi c needs for selecting stockpeople, how-ever, mean that many of these tests are not appropriate for the subject population involved. Further, there are no well-validated tests for empathy towards animals and no published tests at all for stockperson attitudes and beliefs. In choosing tests for use in selection of stockpeople, issues which need to be considered include:

● Is the test designed as a pre-employment tool? Tests that specifi cally refer to a current job or require specifi c knowledge of a particular agricultural industry are clearly inappr opriate for selecting stockpeople to w ork in an agricultural industry for th e fi rst time.

● Is the languag e appropriate for stockpeople? Tests t hat us e lo cal vernacular or u se sophisticated terms may n ot be appropriate f or a ta rget population from which stockpe ople are likely to come.

● What is the length of the test? If a pplicants are to complete a number of test s, then no o ne test should b e undu ly long. A total tes t batter y th at t akes more than an hou r or so to fi ll out is pr obably inappropriate for s electing s tockpeople.

● What d oes the test measu re? Care shou ld be t aken to ensure that th e te stmeasur es vari ables that are ap propriate for sele ction p urposes. There are many publi shed tests tha t may be usefu l for providing em ployers with ad-vice, for ident ifying a reas in wh ich training migh t be appropria te or f or char-acterizing the rang e of qualities t hat people i n particular employm ent sectorsmight have, but such tests may n ot be approp riate tools for selection. Te stsused for select ion pu rposes must be able t o d iscriminate with acce ptableaccuracy those people who will perform well in the job for which they a re selected from those who will no t perfor m well.

● Is t he answer format simple? Tests chose n for us e in selecting sto ckpeoplemay not give reli able res ults if some people are likely to have diffi cult y in responding to questio ns.

● How is t he test presente d? Gi ven that there m ay be some literacy p roblemswithin the targ et population, t he us e of a mu ltimedia format t o pre sent t he

Conclusion 165

tests that is somewhat similar to t hat used i n the training prog ramme de-scribed above may be appropriate.

In s ummary, procedu res to s elect people to work as stockpeo ple need to ta keinto account the pa rt icular character istics of the target p opulation as wel l as t hosecharacteristics that make a p erson well suite d to work with a nimals in the live-stock industries.

The authors’ r esearch ha s shown that there are a spects of stockperson atti-tudes and behaviour that aff ect animal welfare an d pr oductivity in livestock pro-duction. Wh ere pe ople have had previo us experience wo rking in a particular industry, our research has shown that the person’s attitude towards working with animals in that industry is a good predictor of the ir behaviour and, ultimate ly, of farm p roduction. While o ther stu dies have suggested th at p ersonality m ay b erele vant and that empath y may be importa nt, th ese vari ables need t o be s tudiedfurther. As d escribed e arlier i n this book (Secti on 1 .6.2), our r esearch us ing a pre-employment test batt ery that targets those s pecifi c attitudes and behavi ours as well as those generic characte ristics that a re predictive of work performance doeshave the capaci ty to pred ict th e subsequent work perf ormance of pig stockpeople.

Because there is increasing recognition of the need to employ people who will be adaptable and conscientious and who will treat animals well, a demand will develop for a selection procedure that can be widely used. At present the research reported here has yet to be implemented as a practical tool for industry use. There is a real need to extend this research into the other livestock industries and to develop practical stockperson selection tools for the livestock industries.

One other benefi t of having tools that reliably predict stockperson perfor-mance is that such tools might also be used to audit stockmanship. Hemsworth et al. (2009) argued that, owing to the strong relationships between stockperson attitudes and behaviours and animal fear responses, as well as the relationships between attitudes and other job-related characteristics, attitudes, together with empathy, work motivation and technical knowledge and skills, should be the prin-cipal focus of measuring stockmanship in on-farm welfare monitoring schemes.

While large organizations would be able to use the selection procedures described above by utilizing their existing human resources s taff, smalle r opera-tors would nee d to have ac cess t o appropri ate pe ople to conduct selection ontheir beh alf. A s imilar situa tion e xists for traini ng stockpeople on si te. There appears to be a clea r opportunity for tr aining and s election t o be p rovided to the livestock indus tries by consultants. In Australia, th e ProH and training pac kagedescribed above is be ing delivere d by designa ted training offi cers in each state. In ot her count ries, a strategy s imilar to th is might be appr opriate, or perhap sother pro fessionals s uch as agricu ltural pr oduct companies or finance c ompaniesmay provide t he resource.

8.6 Th e B enefi ts for Sto ckpeople of W orking with Animals

Working in intensive farming industries off ers a n opportun ity, fr om a d win-dling range of opportunities, for people to obtain employm ent in rural areas.

166 Chapter 8

This facilitates people remaining with f riends a nd relatives, living in a community with which they identify and having the fi nancial base to live and raise a family in the district where they choose to live. While the better-educated youth from rural areas probably enter rural businesses or migrate to the city, those from rural areas with limited educational opportunities or achievements have less choice. Thus, there may be quite a strong motivation to succeed in jobs that require limited formal training.

In addition, while stockpeople may be dissatisfi ed with their jobs (English et al., 1992), one can question whether most stockpeople actually dislike working with animals. Keeping pets is common in most families and, while the potential benefi ts, such as promoting the development of social competency and responsi-bility (Edney, 1992), are well recognized for children, it is highly likely that adults gain considerable satisfaction and enjoyment from keeping pets. Surveys have shown that most people own pets for largely emotional reasons, which include companionship and the provision of love and affection (Leslie et al., 1994). It is

Fig. 8.3. In addition to the responsibility and satisfaction derived from successfully caring for farm animals, stockpeople may also enjoy interacting with their livestock.

Conclusion 167

also considered that pet ownership provides a range of other rewards for owners, such as companionship, support and a form of interest outside themselves (Edney, 1992).

Interviews of several hundred stockpeople in the pig and dairy industries in Australia by the authors indicate, surprisingly, that while many expressed a dis-like for various aspects of the job, a clear majority of stockpeople (86% and 76% of pig and dairy stockpeople, respectively) enjoyed working with their animals. Therefore, it is not unreasonable to suggest that working with animals provides stockpeople with a number of benefi ts, such as companionship and a commit-ment and interest, and offers both responsibility and a sense of satisfaction for the health and welfare of lives other than those of themselves or their families. Indeed, successfully caring for farm animals may provide greater rewards for people than those gained by successfully working on a production line producing inanimate objects.

8.7 Conclusion

Human–animal interactions can have profound effects on farm animals. Regular negative interactions by stockpeople can result in the animals developing stimulus-specifi c fear responses to humans. As a result of stress responses, high levels of fear for humans can depress both the welfare and performance of farm animals. Our knowledge of human–animal interactions is still limited, but several avenues for improving human–animal interactions to improve animal perfor-mance and welfare exist.

Tactile behaviours may be either positive or negative in nature, and stock-people should be aware that it is important not only to make all behaviours as positive as possible, but also to ensure that the proportion of negative behav-iours is kept low. If some procedures involve negative behaviours, it may be possible to eliminate the procedure altogether, accomplish the procedure mechanically and thus remove the human association, or compensate for the negative behaviours by additional positive interactions. Rewarding experiences at the time may also alleviate the aversiveness of the situation. Further, there is limited evidence that farm animals which have positive relationships with humans may show a reduced stress response in stressful situations in which humans are present. Evidence in a number of livestock industries has shown that the stockperson’s behaviour is ultimately a consequence of his or her beliefs about handling and interacting with farm animals. Therefore, in order to infl u-ence stockperson behaviour, stockpeople have to be exposed to information that will produce changes in their beliefs about handling and interacting with animals. Alternatively, we may wish to select stockpeople to work in the livestock indus-tries on the basis of their beliefs and thus their behaviour towards farm animals. Research has shown that targeting these key attitudes and behaviour can indeed improve animal productivity via reductions in fear and stress.

The human–animal relationship also has immediate and long-term implica-tions for the stockperson. The human–animal interactions may affect the stock-person to the extent that job-related characteristics, such as job satisfaction,

168 Chapter 8

motivation and commitment, may be stimulated with implications for the job performance and career prospects of the stockperson. This, in turn, is also likely to affect the performance and welfare of the livestock.

Therefore, selection and training procedures that target the attitudes and behaviour of the stockperson offer considerable opportunity to improve animal welfare and performance. This is the new direction for industries in which live-stock regularly interact with stockpeople. Much has been done to improve genet-ics, nutrition, health and housing, but efforts to target the stockperson, who performs such a key function, have just begun.

169

References

Abbott, T.A., Hunter, E.J., Guise, H.J. and Penny, R.H.C. (1997) The effect of experi-ence of handling pigs on willingness to move. Applied Animal Behaviour Science54, 371–375.

Adams, B.M. (1968) Effect of cortisol on growth and uric acid excretion in the chick. Jour-nal of Endocrinology 40, 145–151.

Adorno, T., Frenkel-Brunswik, E., Levinson, D. and Sanford, N. (1950) The Authoritarian Personality. Norton, New York.

Ajzen, I. (1985) From intentions to actions: a theory of planned behaviour. In: Kuhl, J. and Beckman, J. (eds) Action-control: from Cognition to Behavior. Springer Verlag, Heidelburg, Germany, pp. 11–39.

Ajzen, I. (1988) Attitudes, Personality and Behaviour. Open University Press, Milton Keynes, UK.

Ajzen, I. and Fishbein, M. (1977) Attitude-behaviour relations: a theoretical analysis and review of empirical research. Psycho-logical Bulletin 84, 888–918.

Ajzen, I. and Fishbein, M. (1980) UnderstandingAttitudes and Predicting Social Behaviour.Prentice-Hall Inc., Englewood Cliffs, New Jersey.

Albarracín, D., Zanna, M.P., Johnson, B.T. and Kumkale, G.T. (2005) Attitudes: introduc-tion and scope. In: Albarracin, D., Johnson, B.T. and Zanna, M.P. (eds) The Handbook

of Attitudes. Psychology Press, New York, pp. 3–19.

Allport, G.W. (1935) Attitudes. In: Murchison, C. (ed.) Handbook of Social Psychology.Clark University Press, Worchester, Massa-chusetts, pp. 798–844.

Allport, G.W. (1937) Personality; Characters and Characteristics. Henry Holt, New York.

Andersen, I.L., Berg, S., Bøe, K.E. and Edwards, S. (2006) Positive handling in late pregnancy and the consequences for maternal behaviour and production in sows. Applied Animal Behaviour Science99, 64–76.

Appleby, M.C., Hughes, B.O. and Elson, H.A. (1992) Poultry Production Systems. CABInternational, Wallingford, UK.

Appleby, M.C., Mench, J.A. and Hughes, B.O. (2004) Poultry Behaviour and Welfare.CAB International, Wallingford, UK.

Arave, C.W., Mickelsen, C.H. and Walters, J.L. (1985) Effect of early rearing experience on subsequent behavior and production of Holstein heifers. Journal of Dairy Science68, 923–929.

Arave, C.W., Albright, J.L., Armstrong, D.V., Foster, W.W. and Larson, L.L. (1992) Effects of isolation of calves on growth, behavior and fi rst lactation milk yield of production of Holstein heifers. Journal of Dairy Science 75, 3408–3415.

170 References

Aronson, E. (1969) The theory of cognitive dissonance: a current perspective. In: Berkowitz, L. (ed.) Advances in Experi-mental Social Psychology 4, 1–34.

Asch, S.E. (1956) Studies of independence and conformity: I. A minority of one against a unanimous majority. Psycho-logical Monographs: General and Applied 70, 547–549.

Aureli, F. and Schaffner, C.M. (2002) Relation-ship assessment through emotional media-tion. Behaviour 139, 393–420.

Bagozzi, R.P. (1980) Causal Models in Market-ing. John Wiley & Sons, New York.

Barnett, J.L. (2003) Studying stress to assess animal welfare. In: Paterson, J.E. (ed.) Manipulating Pig Production IX. Australasian Pig Science Association, Werribee, Victoria, Australia, pp. 107–114, 116–120.

Barnett, J.L. and Hemsworth, P.H. (1989) Fear of humans by laying hens in different tiers of a battery, behavioural and physiologi-cal responses. British Poultry Science 30, 497–502.

Barnett, J.L. and Hemsworth, P.H. (2003) Sci-ence and its application in assessing the welfare of laying hens. Australian Veteri-nary Journal 81, 615–624.

Barnett, J.L. and Hemsworth, P.H. (2009) Wel-fare monitoring schemes: using research to safeguard the welfare of animals on the farm. Journal of Applied Animal Welfare Science 12, 114–131.

Barnett, J.L., Hemsworth, P.H. and Hand, A.M. (1983) The effect of chronic stress on some blood parameters in the pig. AppliedAnimal Ethology 9, 273–277.

Barnett, J.L., Hemsworth, P.H., Cronin, G.M., Winfi eld, C.G., McCallum, T.H. and New-man, E.H. (1988) The effects of genotype on physiological and behavioural responses related to the welfare of pregnant pigs. Applied Animal Behaviour Science 20, 287–296.

Barnett, J.L., Hemsworth, P.H., Newman, E.A., McCallum, T.H. and Winfi eld, C.G. (1989) The effect of design of tether and stall housing on the behavioural and physio-logical responses related to the welfare of pregnant pigs. Applied Animal Behaviour Science 24, 1–12.

Barnett, J.L., Hemsworth, P.H., Cronin, G.M., Newman, E.A. and McCallum, T.H. (1991) Effects of design of individual cage-stalls on the behavioural and physiological responses related to the welfare of preg-nant pigs. Applied Animal Behaviour Sci-ence 32, 23–33.

Barnett, J.L., Hemsworth, P.H. and Newman, E.A. (1992) Fear of humans and its rela-tionships with productivity in laying hens at commercial farms. British Poultry Science33, 699–710.

Barnett, J.L., Hemsworth, P.H. and Jones, R.B. (1993) Behavioural responses of commercial farmed laying hens to humans: evidence of stimulus generalization. Applied Animal Behaviour Science 37, 139–146.

Barnett, J.L., Hemsworth, P.H., Hennessy, D.P., McCallum, T.M. and Newman, E.A. (1994) The effects of modifying the amount of human contact on the behavioural, physiological and production responses of laying hens. Applied Animal Behaviour Science 41, 87–100.

Barnett, J.L., Hemsworth, P.H., Cronin, G.M., Jongman, E.C. and Hutson, G.D. (2001) A review of the welfare issues for sows and piglets in relation to housing. Australian Journal of Agricultural Research 52, 1–28.

Barnett, J.L., Edge, M.K. and Hemsworth, P.H. (2008) The place of quality assurance in managing animal welfare during long dis-tance transport. In: Adams, D.B. and Thornber, P.M. (eds) Welfare aspects of long distance transportation of animals. Veterinaria Italiana 44, 121–131.

Barrick, M.P. and Mount, M. (1991) The big fi ve personality dimensions and job perfor-mance: a meta-analysis. Personnel Psychol-ogy 44, 1–26.

Barton Gade, P., Blaabjerg, L. and Christensen, L. (1992) New lairage system for slaughter pigs – effects on behaviour and quality characteristics. In: Proceedings of the 38th International Congress of Meat Science and Technology, Vol. II. Danish Meat Institute, Roskilde, Denmark, pp. 161–164.

Bartov, I., Jensen, L.S. and Veltman, J.R. (1980) Effect of corticosterone and prolactin on fattening in broiler chicks. Poultry Science59, 1328–1334.

References 171

Bellamy, D. and Leonard, R.A. (1965) Effect of cortisol on the growth of chicks. GeneralComparative Endocrinology 5, 402–410.

Bentler, P.M. and Speckart, G. (1979) Models of attitude–behavior relations. Psychological Review 86, 452–464.

Beveridge, L.M. (1996) Studies on the infl u-ence of human characteristics and training on stockperson work performance and farm animal behaviour. PhD thesis. The University of Aberdeen, UK.

Beynon, N.M. (1991) Analysis of stockmanship. Pig Veterinary Journal 26, 67–77.

Blake, R. and Dennis, W. (1943) Development of stereotypes concerning the Negro. Journal of Abnormal Social Psychology 38, 525–531.

Blecha, F. (2000) Immune system response to stress. In: Mench, J.A. and Moberg, G. (eds) Biology of Animal Stress. CAB Inter-national, Wallingford, UK, pp. 111–122.

Blumberg, M. and Pringle, C.D. (1982) The missing opportunity in organisational research: some implications for a theory of work performance. Academy of Manage-ment Review 7, 560–569.

Bodsworth, W. and Coleman, G.J. (2001) Child–companion animal attachment bonds in single and two-parent families. Anthrozooz 14, 216–223.

Boissy, A. and Bouissou, M.F. (1988) Effects of early handling on heifers’ subsequent reactivity to humans and to unfamiliar situations. Applied Animal Behaviour Science 20, 259–273.

Boivin, X., Le Neindre, P. and Chupin, J.M. (1992) Establishment of cattle–human relationships. Applied Animal Behaviour Science 32, 325–335.

Boivin, X., Tournadre, H. and Le Neindre, P. (2000) Hand-feeding and gentling infl u-ence early-weaned lambs’ attachment responses to their stockperson. Journal of Animal Science 78, 879–884.

Boivin, X., Nowak, R. and Garcia, A.T. (2001) The presence of the dam affects the effi -ciency of gentling and feeding on the early establishment of the stockperson–lamb relationship. Applied Animal Behaviour Science 72, 89–103.

Boivin, X., Boissy, A., Nowak, R., Henry, C., Tournadre, H. and Le Neindre, P. (2002)

Maternal presence limits the effects of feed-ing and gentling on the lambs’ socialisation to the stockperson. Applied Animal Behav-iour Science 77, 311–328.

Boivin, X., Lensink, J., Tallet, C. and Veissier, I. (2003) Stockmanship and farm animal welfare. Animal Welfare 12, 479–492.

Boivin, X., Marcantognini, L., Boulesteix, P., Godet, J., Brulé, A. and Veissier, I. (2007) Attitudes of farmers towards Limousin cat-tle and their handling. Animal Welfare 16, 147–151.

Bolles, R. (1981) Emotion. In: McFarland, D.J. (ed.) The Oxford Companion to Animal Behaviour. Oxford University Press, Oxford, UK, pp. 149–152.

Borofsky, G.L. and Smith, M. (1993) Reductions in turnover, accidents, and absenteeism: the contribution of a pre-employment screening inventory. Journal of Clinical Psychology 49, 109–116.

Bouissou, M.F. and Vandenheede, M. (1995) Fear reactions of domestic sheep confronted with either a human or a human-like model. Behavioural Processes 43, 81–92.

Brambell, F.W.R., Barbour, D.S., Barnett, M.B., Ewer, T.K., Hobson, A., Pitchforth, H., Smith, W.R., Thorpe, W.H. and Winship, F.J.W. (1965) Report of the Technical Com-mittee to Enquire into the Welfare of Ani-mals Kept Under Intensive Husbandry Systems. Her Majesty’s Stationery Offi ce, London, UK.

Bredbacka, P. (1988) Relationships between fear, welfare and productive traits in caged White Leghorn hens. In: Unshelm, J., Van Putten, G., Zeeb, K. and Eksebo (eds)Proceedings of the International Congress on Applied Ethology in Farm Animals.Research Institute of Animal Production, Prague, Czech Republic and Institute of Animal Biochemistry and Genetics, Slovak Academy of Sciences, Slovakia, pp. 74–89.

Brehm, J. and Cohen, A.R. (1962) Explorationsin Cognitive Dissonance. Wiley, New York.

Breuer, K., Hemsworth, P.H. and Coleman, G.J. (1997) The infl uence of handling on the behaviour and productivity of lactating heif-ers. In: Hemsworth, P.H., Spinka, M. and Košt’ál, L. (eds) Proceedings of the 31st International Congress of the International

172 References

Society for Applied Ethology. ISAE, Prague, Czech Republic.

Breuer, K., Hemsworth, P.H., Barnett, J.L., Matthews, L.R. and Coleman, G.J. (2000) Behavioural response to humans and the productivity of commercial dairy cows. Applied Animal Behaviour Science 66, 273–288.

Breuer, K., Hemsworth, P.H. and Coleman, G.J. (2003) The effect of positive or negative handling on the behavioural responses of nonlactating heifers. Applied Animal Behaviour Science 84, 3–22.

Broom, D.M. (1986) Indicators of poor welfare. British Veterinary Journal 142, 524–526.

Broom, D.M. and Johnson, K.G. (1993) Stress and Animal Welfare. Chapman and Hall, London.

Bruckmaier, R.M. and Blum, J.W. (1998) Oxyto-cin release and milk removal in ruminants. Journal of Dairy Science 81, 939–949.

Bruckmaier, R.M., Schams, D. and Blum, J.W. (1993) Milk removal in familiar and unfamiliar surroundings: concentra-tions of oxytocin, prolactin, cortisol and β-endorphin. Journal of Dairy Research60, 449–456.

Bruckmaier, R.M., Wellnitz, O. and Blum, J.W. (1997) Inhibition of milk ejection in cows by oxytocin receptor blockade, α-adrenergic receptor stimulation and in unfamiliar sur-roundings. Journal of Dairy Research 64, 315–325.

Buckland, R.B., Goldrosen, A. and Bernon, D.E. (1974) Effect of blood sampling by cardiac puncture on subsequent body weight of broilers and S.C. White Leghorn replacement pullets. Poultry Science 53,1256–1258.

Bud Williams Stockmanship (2001) Teaching low stress livestock handling methods. Available at: www.stockmanship.com (accessed 29 April 2009).

Budd, R.J., North, D. and Spencer, C. (1984) Understanding seat-belt use: A test of Bentler and Speckart’s extension of the ‘theory of reasoned action’. European Journal of Social Psychology 14, 69–78.

Burrow, H.M. and Corbet, N.J. (2000) Genetic and environmental factors affecting tem-perament of zebu and zebu-derived beef

cattle grazing at pasture in the tropics. Australian Journal of Agricultural Research 51, 155–162.

Burrow, H.M. and Dillion, R.D. (1997) Rela-tionship between temperament, live-weights and commercial carcass traits of Bos indicus crossbreds. Australian Journal of Experimental Agriculture 37, 407–411.

Cabanac, M. (1979) Sensory pleasure. Quar-terly Review of Biology 54, 1–29.

Cacioppo, J.T., Klein, D.J., Berston, G. and Hatfi eld, E. (1993) The psychophysiology of emotion. In: Lewis, M. and Haviland, J.M. (eds) Handbook of Emotions. Guild-ford Press, New York, pp. 119–142.

Cain, A.O. (1991) Pets and the family. HolisticNursing Practice 5, 58–63.

California Secretary of State (2007) Informa-tion for voters. Available at: www.sos.ca.gov (accessed 10 July 2009).

Cammann, C., Fichmann, M., Jenkins, D. and Klesh, J. (1979) The Michigan Organiza-tional Assessment Questionnaire. The University of Michigan, Ann Arbor, Michigan.

Campbell, D.T. (1963) Social attitudes and other acquired behavioural dispositions. In: Koch, S. (ed.) Psychology: a Study of a Science.McGraw-Hill, New York, pp. 94–172.

Campion, M.A. (1991) Meaning and measure-ment of turnover: comparison of alterna-tive measures and recommendations for research. Journal of Applied Psychology76, 199–212.

Cannon, W.B. (1914) The emergency function of the adrenal medulla in pain and the major emotions. American Journal of Physiology 33, 356–372.

Carless, S.A., Fewings-Hall, S., Hall, M., Hay, M., Hemsworth, P. and Coleman, G.J. (2007) Selecting unskilled and semi-skilled blue-collar workers: the criterion-related validity of the PDI-Employment Inventory. International Journal of Selection and Assessment 15, 335–340.

Chester Jones, I. and Henderson, I.W. (1976) General, Comparative and Clinical Endo-crinology of the Adrenal Cortex, Vol. 1. Academic Press, London.

Chlopan, B.E., McCain, M.L., Carbonell, J.L. and Hagen, R. (1985) Empathy: review of

References 173

available measures. Journal of Personality and Social Psychology 48, 635–653.

Chusmir, L.H. (1982) Job commitment and th eorganisational woman. Aca demy of M an-agement Re view 7, 595– 60 2.

C l arke, I.J., Hem sworth, P.H., Barnett, J.L. and Tilbrook, A.J. (1992) Stress and reproduc-tion in farm animals. In: Sheppard, K.E., Boublik, J.H. and Funder, J.W. (eds) Stress and Reproduction. Serono Symposium Publications, Vol. 86. Raven Press, New York, pp. 239–251.

Cleary, G.V. (1990) Personnel management and staff training. In: Gardner, J.A.A., Dunkin, A.C. and Lloyd, L.C. (eds) Per-sonnel Management and Record Keeping.Pig Production in Australia. Australian Pig Research Council, Canberra, Australian Capital Territory, pp. 284–291.

Coleman, G.J. (2001) Selection of stockpeople to improve productivity. Fourth Industrial and Organisational Psychology Confer-ence, Sydney, New South Wales, p. 30.

Coleman, G.J. (2004) Personnel management in livestock industries. In: Benson, G.J. and Rollin, B.E. (eds) The Well-Being of Farm Animals: Challenges and Solutions. Black-well Publishing, Ames, Iowa, pp.167–181.

Coleman, G.J. (2008) Public perceptions of animal pain and animal welfare. OIE Technical Series (World Organisation for Animal Health) 10, 26–37.

Coleman, G.J. (2010) Educating the public: information or persuasion? Journal of Vet-erinary Medical Education 37, 74–82.

Coleman, G.J., Hemsworth, P.H. and Hay, M. (1998) Predicting stockperson behaviour towards pigs from attitudinal and job-related variables and empathy. AppliedAnimal Behaviour Science 58, 63–75.

Coleman, G.J., Hemsworth, P.H., Hay, M. and Cox, M. (2000) Modifying stockperson attitudes and behaviour towards pigs at a large commercial farm. Applied Animal Behaviour Science 66, 11–20.

Coleman, G.J., McGregor, M., Hemsworth, P.H., Boyce, J. and Dowling, S. (2003) The rela-tionship between beliefs, attitudes and observed behaviours of abattoir personnel in the pig industry. Applied Animal Behav-iour Science 82, 189–200.

Costa, P.T. and McCrae, R.R. (1992) NEO PI-R.Professional Manual. Psychological Assess-ment Resources Inc., Odessa, Florida.

Cote, S. (2004) Stockmanship. A Powerful Tool for Grazing Lands Management. Natural Resources Conservation Service, and the Butte Soil and Water Conservation District, Arco, Idaho.

Cox, R.P. (1993) The human–animal bond as a correlate of family functioning. ClinicalNursing Research 2, 224–231.

Cransberg, P.H., Hemsworth, P.H. and Coleman, G.J. (2000) Human factors affecting the behaviour and productivity of commercial broiler chickens. British Poul-try Science 41, 272–279.

Creel, S.R. and Albright, J.L. (1988) The effects of neonatal social isolation on the behaviour and endocrine function of Holstein calves. Applied Animal Behaviour Science 21, 293–306.

Cummings, R.B. (1994) Opportunities for whole grain feeding. In: Proceedings of the Ninth European Poultry Conference,Glasgow, UK. World Poultry Science Asso-ciation, UK Branch, pp. 219–222.

Danbury, T.C., Weeks, C.A. and Chambers, J.P. (2000) Self-selection of the analgesic drug carprofen by lame broiler chickens. Veteri-nary Record 146, 307–311.

Dantzer, R. (1988) Les Emotions. Collection Que Sais-je? Presse Universitaires de France, Paris.

Dantzer, R. and Mormède, P. (1983) Stress in farm animals: a need for reevaluation. Journal of Animal Science 57, 6–18.

Dawkins, M. (1976) Towards an objective method of assessing welfare in domestic fowl. Applied Animal Ethology 2, 245–254.

Dawkins, M. (1977) Do hens suffer in battery cages? Environmental preferences and wel-fare. Animal Behaviour 25, 1034–1046.

Dawkins, M. (1980) Animal Suffering: the Sci-ence of Animal Welfare. Chapman and Hall, New York.

Dawkins, M. (1983) Battery hens name their price: consumer demand theory and the measurement of animal needs. Animal Behaviour 31, 1195–1205.

Dawkins, M.S. (2003) Behaviour as a tool in the assessment of animal welfare. Zoology106, 383–387.

174 References

Dawkins, M.S. (2006) Through animal eyes: what behaviour tells us. Applied Animal Behaviour Science 100, 4–10.

Day, J.E.L., Spoodler, H.A.M., Burfoot, A., Chamberlain, H.L. and Edwards, S.A. (2003) The separate and interactive effects of handling and environmental enrichment on the behaviour and welfare of growing pigs. Applied Animal Behaviour Science75, 177–192.

De Alencar, M. do C.B., Nääs, I.A., Gontijo, L.A. and Salgado, D.A. (2007) Effects of labor motivation in poultry production. Brazilian Journal of Poultry Science 9, 213–217.

de Passille, A.M. and Rushen, J. (2005) Can we measure human–animal interactions in on-farm animal welfare assessment? Some unresolved issues. Applied Animal Behav-iour Science 92, 193–209.

de Passille, A.M., Rushen, J., Ladewig, J. and Petherick, C. (1996) Dairy calves’ discrimination of people based on previous handling. Journal of Animal Science 74, 969–974.

Dewsbury, D.A. (1992) Studies of rodent–human interactions in animal psychology. In: Davis, H. and Balfour, A.D. (eds) TheInevitable Bond – Examining Scientist–Animal Interactions. Cambridge University Press, Cambridge, UK, pp. 27–43.

Duan, C. and Hill, C. (1996) The current state of empathy research. Journal of Counsel-ing Psychology 43, 261–274.

Duncan, I.J.H. (1978) The interpretation of preference tests in animal behaviour. Applied Animal Ethology 4, 197–200.

Duncan, I.J.H. (2004) A concept of welfare based on feelings. In: Benson, G.J. and Rollin, B.E. (eds) The Well-Being of Farm Animals: Challenges and Solutions. Black-well Publishing, Ames, Iowa, pp. 95–101.

Duncan, I.J.H. and Fraser, D. (1997) Under-standing animal welfare. In: Appleby, M.C. and Hughes, B.O. (eds) Animal Welfare.CAB International, Wallingford, UK, pp. 19–31.

Duncan, I.J.H. and Petherick, J.C. (1991) The implications of cognitive processes for ani-mal welfare. Journal of Animal Science 69, 5017–5022.

Duncan, I.J.H., Slee, G.S., Kettlewell, P., Berry, P. and Carlisle, A.J. (1986) Com-parison of the stressfulness of harvesting broiler chickens by machine and by hand.British Poultry Science 27, 109–114.

Eagly, A.H. and Chaiken, S. (1993) The Psychology of Attitudes. Harcourt Brace Jovanovich College Publishers, Orlando, Florida.

Echabe, A.E., Rovir, D.P. and Garate, J.V. (1988) Testing Ajzen and Fishbein’s attitude model: the prediction on voting. European Journal of Social Psychology 18, 181–189.

Edney, A.T.B. (1992) Companion animals and human health. Veterinary Record 130, 285–288.

Edwards, L.E. (2009) The human–animal rela-tionship in the laying hen. PhD thesis. The University of Melbourne, Victoria, Australia.

EFSA (European Food Safety Authority) (2005) Welfare Aspects of Various Systems for Keeping Laying Hens. Annex to The EFSA Journal 197, 1–23. EFSA-Q-2003–92.

Eldridge, G.A. and Knowles, H.M. (1994)Improving Meat Quality through Improved Pre-slaughter Management. Pig Research & Development Corporation Research Report (DAV91P), Victorian Institute of Animal Science, Werribee, Victoria, Australia.

Elsasser, T.H., Klasing, K.C., Filiov, N. and Thomp-son, F. (2000) The metabolic consequences of stress: targets for stress and priorities of nutrient use. In: Mench, J.A. and Moberg, G. (eds) Biology of Animal Stress. CAB Interna-tional, Wallingford, UK, pp. 77–110.

English, P., Burgess, G., Segundo, R. and Dunne, J. (1992) Stockmanship: Improving the Care of the Pig and Other Livestock.Farming Press Books, Ipswich, UK.

Estep, D.Q. and Hetts, S. (1992) Interactions, relationships, and bonds: the conceptual basis for scientist–animal relations. In: Davis, H. and Balfour, A.D. (eds) TheInevitable Bond – Examining Scientist–Animal Interactions. Cambridge University Press, Cambridge, UK, pp. 6–26.

European Commission (2007) Attitudes of EU Citizens towards Animal Welfare. Special Eurobarometer 270/Wave 66.1.

Eysenck, H.J. (1966) Personality and experi-mental psychology. Bulletin of the British Psychological Society 19, 1–28.

References 175

Faure, J.M. (1981) Bidirectional sectional for open-fi eld activity in young chicks. Behav-iour Genetics 11, 135–143.

FAWC (Farm Animal Welfare Council) (1993) Second Report on Priorities for Research and Development in Farm Animal Welfare.London.

Fell, L.R. and Shutt, D.A. (1988) Salivary cortisol and behavioural indicators of stress in sheep. Animal Production in Australia 17, 186–189.

Festinger, L. (1957) A Theory of Cognitive Dissonance. Stanford University Press, Polo Alto, California.

Festinger, L. and Carlsmith, J.M. (1959) Cogni-tive consequences of forced compliance. Journal Abnormal and Social Psychology58, 203–210.

Fishbein, M. and Ajzen, I. (1974) Attitudes towards objects as predictors of single and multiple behavioral criteria. Psychological Review 81, 59–74.

Fordyce, G., Wythes, J.R., Shorthose, W.R., Underwood, D.W. and Shepherd, R.K. (1988) Cattle temperaments in extensive beef herds in Northern Queensland. 2. Effect of temperament on carcass and meat quality. Australian Journal of Experimental Agriculture 28, 689–693.

Fraser, A.F. and Broom, D.M. (1990) Farm Ani-mal Behaviour and Welfare. CAB Interna-tional, Wallingford, UK.

Fraser, D. (2003) Assessing animal welfare at the farm and group level: the interplay of science and values. Animal Welfare 12, 433–443.

Fraser, D. (2008) Understanding Animal Welfare: the Science in its Cultural Context. Wiley-Blackwell, Chichester, West Sussex, UK.

Fraser, D. and Duncan, I.J.H. (1998) ‘Plea-sures’, ‘pains’ and animal welfare: toward a natural history of affect. Animal Welfare7, 383–396.

Fraser, D. and Matthews, L.R. (1997) Prefer-ence and motivation testing. In: Appleby, M.C. and Hughes, B.O. (eds) Animal Wel-fare. CAB International, Wallingford, UK, pp. 159–174.

Freeman, B.M. and Manning, A.C.C. (1979) Stressor effects of handling on the immature fowl. Research in Veterinary Science 26, 223–226.

Fulkerson, W.J. and Jamieson, P.A. (1982) Pattern of cortisol release in sheep following adminis-tration of synthetic ACTH or imposition of various stressor agents. Australian Journal of Biological Science 35, 215–222.

Galef, B.G. Jr (1970) Aggression and timidity: responses to novelty in feral Norway rats. Journal of Comparative and Physiological Psychology 70, 370–381.

Geverink, N.A., Kappers, A., van de Burgwal, J.A., Lambooij, E., Blokhuis, H.J. and Wiegant, V.M. (1998) Effects of regular moving and handling on the behavioural and physiological responses of pigs to pre-slaughter treatment and consequences for subsequent meat quality. Journal of Ani-mal Science 76, 2080–2085.

Gonyou, H.W., Hemsworth, P.H. and Barnett, J.L. (1986) Effects of frequent interactions with humans on growing pigs. AppliedAnimal Behaviour Science 16, 269–278.

Grandin, T. (1980) Livestock behavior as related to handling facilities design. International Journal for the Study of Animal Problems 1, 33–52.

Grandin, T. (1993) Behavioural principles of cattle handling under extensive conditions. In: Grandin, T. (ed.) Livestock Handling and Transport. CAB International, Walling-ford, UK, pp. 43–57.

Grandin, T. (2007) Behavioural principles of handling cattle and other grazing animals under extensive conditions. In: Grandin, T. (ed.) Livestock Handling and Transport,3rd edn. CAB International, Wallingford, Oxon, UK, pp. 44–64.

Grandin, T., Curtis, S.E. and Taylor, I.A. (1987) Toys, mingling and driving reduce excitability in pigs. Journal of Animal Science 65 (Suppl. 1), 230 (abstract).

Gray, J.A. (1987) The Psychology of Fear and Stress, 2nd edn. Cambridge University Press, Cambridge, Massachusetts.

Gregory, N.G., Wilkins, L.J., Alvey, D.M. and Tucker, S.A. (1993) Effect of catching method and lighting intensity on the prevalence of broken bones and on the ease of handling of end of lay hens. Veterinary Record 132, 127–129.

Gross, W.B. and Siegel, P.B. (1979) Adaptation of chickens to their handlers and

176 References

experimental results. Avian Diseases 23, 708–714.

Gross, W.B. and Siegel, P.B. (1980) Effects of early environmental stresses on chicken body weight, antibody response to RBC antigens, feed effi ciency and response to fasting. Avian Diseases 24, 549–579.

Gross, W.B. and Siegel, P.B. (1981) Socializa-tion as a factor in resistance to infection, feed effi ciency and response to antigen in chickens. American Journal of Veterinary Research 43, 2010–2012.

Gross, W.B. and Siegel, P.B. (1982) Infl uences of sequences of environmental factors on the responses of chickens to fasting and to Staphylococcus aureus infection. American Journal of Veterinary Research 43, 137–139.

Gross, W.B. and Siegel, H.S. (1983) Evaluation of the heterophil/lymphocyte ratio as a measure of stress in chickens. Avian Dis-ease 27, 972–979.

Harbuz, M.S. and Lightman, S.L. (1992) Stress and the hypothalamo-pituitary-adrenal axis: acute, chronic and immunological activation. Journal of Endocrinology 134, 327–339.

Hargreaves, A.L. and Hutson, G.D. (1990a) Changes in heart rate, plasma cortisol and haematocrit of sheep during a shearing procedure. Applied Animal Behaviour Sci-ence 26, 91–101.

Hargreaves, A.L. and Hutson, G.D. (1990b) An evaluation of the contribution of isolation, p-ending and wool removal to the stress response to shearing. Applied Animal Behaviour Science 26, 103–113.

Hargreaves, A.L. and Hutson, G.D. (1990c) The effect of gentling on heart rate, fl ight distance and aversion of sheep to a handling procedure. Applied Animal Behaviour Science 26, 243–252.

Harrison, R. (1964) Animal Machines. Vincent Stuart Ltd, London.

Hearnshaw, H., Barlow, R. and Want, G. (1979) Development of a temperament or handling diffi culty score for cattle. Proceedings of the Inaugural Conference of Australian Animal Breed Genetics 1, 164–166.

Hediger, H. (1964) The animal’s relationship with man. In: Hediger, H. Wild Animals in Captivity. Dover Publications Inc., New York.

Hemsworth, P.H. (2000) Behavioural principles in pig handling. In: Grandin, T. (ed.) Live-stock Handling and Transport. CAB Inter-national, Wallingford, UK, pp. 255–274.

Hemsworth, P.H. (2007a) Ethical stockman-ship. Australian Veterinary Journal 85, 194–200.

Hemsworth, P.H. (2007b) Behavioural princi-ples of pig handling. In: Grandin, T. (ed.)Livestock Handling and Transport, 3rd edn. CAB International, Wallingford, UK, pp. 214–227.

Hemsworth, P.H. and Barnett, J.L. (1987) Human–animal interactions. In: Price, E.O. (ed.) The Veterinary Clinics of North America, Food Animal Practice, Vol. 3. W.B. Saunders, Philadelphia, pp. 339–356.

Hemsworth, P.H. and Barnett, J.L. (1989) Relationships between fear of humans, productivity and cage position of laying hens. British Poultry Science 30, 505–518.

Hemsworth, P.H. and Barnett, J.L. (1991) The effects of aversively handling pigs either individually or in groups on their behaviour, growth and corticosteroids. Applied Animal Behaviour Science 30, 61–72.

Hemsworth, P.H. and Barnett, J.L. (1992) The effects of early contact with humans on the subsequent level of fear of humans by pigs. Applied Animal Behaviour Science 35, 83–90.

Hemsworth, P.H. and Coleman, G.J. (2009) Animal welfare and management. In: Smulders, F.J.M. and Algers, B. (eds) Food Safety Assurance and Veterinary Public Health, Volume 5. Welfare of Pro-duction Animals: Assessment and Manage-ment of Risks. Wageningen Academic Publishers, Wageningen, The Netherlands, pp. 133–147.

Hemsworth, P.H. and Gonyou, H.W. (1997) Human contact. In: Appleby, M.C. and Hughes, B.O. (eds) Animal Welfare. CAB International, Wallingford, UK, pp. 205–305.

Hemsworth, P.H., Barnett, J.L. and Hansen, C. (1981a) The infl uence of handling by humans on the behaviour, growth and corticosteroids in the juvenile female pig. Hormones and Behavior 15, 396–403.

Hemsworth, P.H., Brand, A. and Willems, P. (1981b) The behavioural response of sows

References 177

to the presence of human beings and its relation to productivity. Livestock Pro-duction Science 8, 67–74.

Hemsworth, P.H., Gonyou, H.W. and Dzuik, P.J. (1986a) Human communication with pigs: the behavioural response of pigs to specifi c human signals. Applied Animal Behaviour Science 15, 45–54.

Hemsworth, P.H., Barnett, J.L., Hansen, C. and Gonyou, H.W. (1986b) The infl uence of early contact with humans on subsequent behavioural response of pigs to humans. Applied Animal Behaviour Science 15, 55–63.

Hemsworth, P.H., Barnett, J.L. and Hansen, C. (1986c) The infl uence of handling by humans on the behaviour, reproduction and corticosteroids of male and female pigs. Applied Animal Behaviour Science15, 303–314.

Hemsworth, P.H., Barnett, J.L. and Hansen, C. (1987) The infl uence of inconsistent handling on the behaviour, growth and corticosteroids of young pigs. AppliedAnimal Behaviour Science 17, 245–252.

Hemsworth, P.H., Barnett, J.L., Tilbrook, A.J. and Hansen, C. (1989a) The effects of handling by humans at calving and during milking on the behaviour and milk cortisol concentrations of primiparous dairy cows. Applied Animal Behaviour Science 22, 313–326.

Hemsworth, P.H., Barnett, J.L., Coleman, G.J. and Hansen, C. (1989b) A study of the relationships between the attitudinal and behavioural profi les of stockpeople and the level of fear of humans and the reproduc-tive performance of commercial pigs. App-lied Animal Behaviour Science 23, 301–314.

Hemsworth, P.H., Barnett, J.L. and Coleman, G.J. (1993) The human–animal relationship in agriculture and its consequences for the animal. Animal Welfare 2, 33–51.

Hemsworth, P.H., Coleman, G.J. and Barnett, J.L. (1994a) Improving the attitude and behaviour of stockpeople towards pigs and the consequences on the behaviour and reproductive performance of commer-cial pigs. Applied Animal Behaviour Science 39, 349–362.

Hemsworth, P.H., Coleman, G.J., Cox, M. and Barnett, J.L. (1994b) Stimulus generali-sation: the inability of pigs to discriminate between humans on the basis of their previous handling experience. AppliedAnimal Behaviour Science 40, 129–142.

Hemsworth, P.H., Coleman, G.C., Barnett, J.L. and Jones, R.B. (1994c) Behavioural responses of humans and the producti-vity of commercial broiler chickens. Applied Animal Behaviour Science 41, 101–114.

Hemsworth, P.H., Barnett, J.L., Breuer, K., Coleman, G.C. and Matthews, L.R. (1995) An Investigation of the Relationships Bet-ween Handling and Human Contact and the Milking Behaviour, Productivity and Welfare of Commercial Dairy Cows.Research Report on Dairy Research and Development Council Project, Attwood, Victoria, Australia.

Hemsworth, P.H., Coleman, G.C., Cransberg, P.H. and Barnett, J.L. (1996a) Human Fac-tors and the Productivity and Welfare of Commercial Broiler Chickens. Research Report on Chicken Meat Research & Deve-lopment Council Project, Attwood, Victoria, Australia.

Hemsworth, P.H., Barnett, J.L. and Campbell, R.G. (1996b) A study of the relative aversiveness of a new daily injection procedure for pigs. Applied Animal Behaviour Science 49, 389–401.

Hemsworth, P.H., Price, E.O. and Bogward, R. (1996c) Behavioural responses of domes-tic pigs and cattle to humans and novel stimuli. Applied Animal Behavioural Sci-ence 50, 43–56.

Hemsworth, P.H., Verge, J. and Coleman, G.J. (1996d) Conditioned approach avoidance responses to humans: the ability of pigs to associate feeding and aversive social experiences in the presence of humans with humans. Applied Animal Behaviour Science 50, 71–82.

Hemsworth, P.H., Pedersen, V., Cox, M., Cronin, G.M. and Coleman, G.J. (1999) A note on the relationship between the behavioural response of lactating sows to humans and the survival of their piglets. Applied Animal Behaviour Science 65, 43–52.

178 References

Hemsworth, P.H., Coleman, G.J., Barnett, J.L. and Borg, S. (2000) Relationships between human–animal interactions and productiv-ity of commercial dairy cows. Journal of Animal Science 78, 2821–2831.

Hemsworth, P.H., Coleman, G.J., Barnett, J.L., Borg, S. and Dowling, S. (2002) The effects of cognitive behavioral intervention on the attitude and behavior of stockper-sons and the behavior and productivity of commercial dairy cows. Journal of Animal Science 80, 68–78.

Hemsworth, P.H., Barnett, J.L., Rickard, M. and Coleman, G.J. (2007) Australia’s research and development capacity in animal wel-fare. Farm Policy Journal 4 (4, November Quarter 2007), 23–31.

Hemsworth, P.H., Rice, M., Calleja, L., Karlen, G.A., Coleman, G.J. and Barnett, J.L. (2008) The relationships between handling prior to slaughter and stress in sheep and cattle at abattoirs. In: Boyle, L., O’Connell, N. and Hanlon, A. (eds) Proceedings of the 42nd International Congress of the Inter-national Society for Applied Ethology,Dublin, Ireland. Wageningen Academic Publishers, Wageningen, The Netherlands, p. 75 (abstract).

Hemsworth, P.H., Barnett, J.L. and Coleman, G.J. (2009) The integration of human-animal relations into animal welfare monitoring schemes. Animal Welfare 18, 335–345.

Hill, J.D., McGlone, J.J., Fullwood, S.D. and Miller, M.F. (1998) Environmental enrich-ment infl uences on pig behaviour, perfor-mance and meat quality. Applied Animal Behaviour Science 57, 51–68.

Hinde, R.A. (1970) Behaviour: a Synthesis of Ethology and Comparative Psychology.McGraw-Hill Kogakusha Ltd, Tokyo, Japan.

Hinde, R.A. (1976) Interactions, relationships and social structure. Man 11, 11–17.

Hogan, J.A. (2008) Motivation. In: Bolhuis, J.J. and Giraldeau, L.-C. (eds) The Behaviour of Animals: Mechanisms, Function and Evolution. Blackwell Publishing, Malden, Massachusetts, pp. 41–70.

Hosey, G.T. (2008) A preliminary model of human–animal relationships in the zoo. Applied Animal Behaviour Science 109, 105–127.

Hovland, C.I., Janis, I.L. and Kelley, H.H. (1953) Communication and Persuasion: Psycho-logical Studies of Opinion Change. Yale University Press, New Haven, Connecticut.

Hughes, B.O. (1975) Spatial preference in the domestic fowl. British Veterinary Journal131, 560–564.

Hughes, B.O. (1976) Preference decisions of domestic hens for wire or litter fl oors. Applied Animal Behaviour Science 2, 155–165.

Hughes, B.O. and Black, A.J. (1973) The pref-erence of domestic hens for different types of battery cage fl oor. British Journal of Poultry Science 14, 615–619.

Hughes, B.O. and Black, A.J. (1976) The infl u-ence of handling on egg production, egg shell quality and avoidance behaviour of hens. British Poultry Science 17, 135–144.

Hutson, G.D. (1981) Sheep movement on slated fl oors. Australian Journal of Experi-mental Agriculture and Animal Husbandry 21, 474–479.

Hutson, G.D. (1982) ‘Flight distance’ in Merino Sheep. Animal Production 35, 231–235.

Hutson, G.D. (1984) Animal welfare and con-sumer demand theory: are preference tests a luxury we can’t afford? Animal Behav-iour 32, 1260–1261.

Hutson, G.D. (1985) The infl uence of barley food rewards on sheep movement through a handling system. Applied Animal Behaviour Science 14, 263–273.

Ivanov, Yu.K. and Patrushev, V.D. (1976) The impact of labor conditions on work satisfac-tion of agricultural workers. Sotisiologicheskie-Issledovaniya 3 (Jul–Sep), 60–70.

Jago, J.G., Krohn, C.C. and Matthews, L.R. (1999) The infl uence of feeding and han-dling on the development of the human–animal interactions in young cattle. AppliedAnimal Behaviour Science 64, 137–151.

Janis, I.L. and Hovland, C.I. (1959) An over-view of persuasibility research. In: Hov-land, C.I. and Hanis, I.L. (eds) Personality and Persuasibility. Yale University Press, New Haven, Connecticut, pp. 1–28.

Jewell, L.N. and Siegall, M. (1990) Contempo-rary Industrial/Organizational Psychology,2nd edn. West Publishing, St Paul, Minne-sota.

References 179

Johns, G. (1987) The great escape: the whys and wherefores of absenteeism add up to a $30 billion question for American business. Psychology Today 21, 30–33.

Jones, R.B. (1983) Fear responses in domestic chicks as a function of the social environ-ment. Behavioural Processes 8, 309–325.

Jones, R.B. (1987) Fear and fear responses: a hypothetical consideration. Medical Science Research 15, 1287–1290.

Jones, R.B. (1993) Reduction of the domestic chick’s fear of humans by regular handling and related treatments. Animal Behaviour46, 991–998.

Jones, R.B. (1994) Regular handling and the domestic chick’s fear of human beings: generalisation of response. Applied Animal Behaviour Science 42, 129–143.

Jones, R.B. and Faure, J.M. (1981) The effects of regular handling on fear responses in the domestic chick. Behavioural Processes 6, 135–143.

Jones, R.B. and Hughes, B.O. (1981) Effects of regular handling on growth in male and female chicks of broiler and layer strains. British Poultry Science 22, 461–465.

Jones, R.B. and Waddington, D. (1992) Modifi cation of fear in domestic chicks, Gallus gallus domesticus via regular handling and early environmental enrichment. Animal Behaviour 43, 1021–1033.

Jones, R.B. and Waddington, D. (1993) Attenuation of the domestics chick’s fear of human beings via regular handling: in search of a sensitive period. Applied Animal Behaviour Science 36, 185–195.

Jones, R.B., Duncan, I.J.H. and Hughes, B.O. (1981) The assessment of fear in domestic hens exposed to a looming human stimulus. Behavioural Processes 6, 121–133.

Jones, R.B., Mills, A.D. and Faure, J.M. (1991) Genetic and experimental manipulation of fear-related behaviour in Japanese quail chicks (Coturnix coturnix japonica). Journal of Comparative Psychology 105, 15–24.

Kaltas, G.A. and Chrousos, G.P. (2007) The neuroendorinology of stress. In: Cacioppo, J.T., Tassinary, L.G. and Berntson, G.G. (eds) Handbook of Psychophysiology.Cambridge University Press, Cambridge, UK, pp. 303–318.

Kanekar, S. (1976) Observational learning of attitudes: a behavioural analysis. European Journal of Social Psychology 6, 5–24.

Katz, D. (1960) The functional approach to the study of attitude. Public Opinion Quarterly 24, 163–204.

Kehrli, M.E., Burton, J.L., Nonnecke, B.J. and Lee, E.K. (1999) Effect of stress on leuko-cyte traffi cking and immune responses: implications for vaccinations. Advances in Veterinary Medicine 41, 61–81.

Kendall, P.C. and Hollon, S.D. (1979) Cognitive-Behavioural Interventions. Academic Press, London.

Kennedy, J.M. and Baldwin, B.A. (1972) Taste preferences in pigs for nutritive and non-nutritive sweet solutions. Animal Behav-iour 20, 706–718.

Kerpelman, J.P. and Himmelfarb, S. (1971) Partial reinforcement effects in attitude acquisition and counterconditioning. Journal of Person-ality and Social Psychology 19, 301–305.

Kilgour, R. and de Langen, H. (1970) Stress in sheep resulting from management prac-tices. Proceedings of the New Zealand Society of Animal Production 30, 65–76.

Kline, P. and Lapham, S. (1991a) The validity of the PPQ: a study of its factor structure and its relationship to the EPQ. Personality and Individual Differences 6, 631–635.

Kline, P. and Lapham, S. (1991b) The validity of the V scale of the PPQ. Personality and Individual Differences 6, 637–641.

Kondos, A.C. (1983) Human resources. In: Pro-ceedings of the 2nd National Pig Produc-tion and Marketing Review Conference, Hobart. Standing Committee on Agri-culture, Department of Primary Industries and Energy, Canberra, Australian Capital Territory.

Krohn, C.C., Jago, J.G. and Boivin, X. (2001) The effects of early handling on the social-ization of young calves to humans. AppliedAnimal Behaviour Science 74, 121–133.

Krohn, C.C., Boivin, X. and Jago, J.G. (2003) The presence of the dam during handling prevents the socialization of young calves to humans. Applied Animal Behaviour Sci-ence 80, 263–275.

La Piere, R.T. (1934) Attitudes vs. actions. Social Forces 13, 230–237.

180 References

Ladewig, J. and Matthews, L.R. (1996) The role of operant conditioning in animal welfare research. Acta Agriculturae Scandinavica Section A: Animal Science 27 (Suppl.), 64–68.

Lambton, S.L., Knowles, T.G., Yorke, C. and Nicol, C.J. (2010) The risk factors affecting the development of gentle and severe feather pecking in loose house laying hens. Applied Animal Behaviour Science 123,32–42.

Lawrence, A.B. and Illius, A.W. (1997) Measur-ing preferences and the problems of identi-fying proximate needs, animal choices. Animal Science 20, 19–26.

LeDoux, J.E. (1996) The Emotional Brain.Simon & Schuster, New York.

Lensink, B.J., Boissy, A. and Veissier, I. (2000a) The relationship between attitude and behaviour towards calves, and productivity of veal units. Annales de Zootechnie 49, 313–327.

Lensink, B.J., Boivin, X., Pradel, P., Le Neindre, P. and Veissier, I. (2000b) Reducing veal calves’ reactivity to people by providing additional human contact. Journal of Ani-mal Science 78, 1213–1218.

Lensink, B.J., Fernandez, X., Boivin, X., Pradel, P., Le Neindre, P. and Veissier, I. (2000c) The impact of gentle contacts on ease of handling, welfare, and growth of calves and on quality of veal meat. Journal of Animal Science 78, 1219–1226.

Lensink, B.J., Fernandez, X., Cozzi, G., Florand, L. and Veissier, I. (2001a) The infl uence of farmers’ behaviour on calves’ reactions to transport and quality of veal meat. Journalof Animal Science 79, 642–652.

Lensink, B.J., Raussi, S., Boivin, X., Pyykko-nen, M. and Veissier, I. (2001b) Reactions of calves to handling depend on housing condition and previous experience with humans. Applied Animal Behaviour Sci-ence 70, 187–199.

Lensink, B.J., Veissier, I. and Florland, L. (2001c) The farmers’ infl uence on calves’ behaviour, health and production of a veal unit. Animal Science 72, 105–116.

Lensink, B.J., Leruste, H., De Bretagne, T. and Bizeray-Filoche, D. (2009) Sow behav-iour towards humans during standard

management procedures and their rela-tionship to piglet survival. Applied Animal Behaviour Science 119, 151–157.

Leslie, B.E., Meek, A.H., Kawash, G.F. and McKeown, D.B. (1994) An epidemiological investigation of pet ownership in Ontario. Canadian Veterinary Journal 35, 218–222.

Levy, N. (2004) What Makes Us Moral? Cross-ing the Boundaries of Biology. Oneworld Publications, Oxford, UK.

Lloyd, D.H. (1974) Effective staff management. In: Freedman, B.M. and Boorman, K.N. (eds) Economic Factors Affecting Egg Production. British Poultry Science Ltd, Edinburgh, UK, pp. 221–251.

Lorenz, K.Z. (1970) Companions as factors in the bird’s environment. In: Lorenz, K.Z. (ed.) Studies on Animal and Human Behaviour. Harvard University Press, Cambridge, Massachusetts.

Low Stress Stockhandling (2005) Attitude is everything. Available at: www.lss.net.au/index.htm (accessed 25 April 2010).

Lynne, D.G. and Rola, L.R. (1988) Improving attitude–behavior prediction models with economic variables: farmer actions toward soil conservation. Journal of Social Psy-chology 128, 19–28.

Lyons, D.M. (1989) Individual differences in temperament of dairy goats and the inhibition of milk ejection. Applied Animal Behaviour Science 22, 269–282.

Lyons, D.M., Price, E.D. and Moberg, G.P. (1988) Individual differences in tempera-ment of domestic dairy goats: constancy and change. Animal Behaviour 36, 1323–1333.

Maio, G.R., Esses, V.M. and Bell, D.W. (1994) The formation of attitudes toward new immigrant groups. Journal of Applied Social Psychology 24, 1762–1776.

Mandler, G. (1981) Emotion. In: Gregory, R.L. (ed.) The Oxford Companion to the Mind.Oxford University Press, Oxford, UK, pp. 1219–1220.

Marchant, J.N. and Broom, D.M. (1994) Effects of housing system on movement and leg strength in sows. Applied Animal Behav-iour Science 41, 275–276.

Martel, Y. (2002) Life of Pi. Canongate Books Ltd, Edinburgh, UK, p. 16.

References 181

Maslow, A.H. (1970) Motivation and Personality,2nd edn. Harper and Row, New York.

Mateo, J.M., Estep, D.Q. and McCann, J.S. (1991) Effects of differential handling on the behaviour of domestic ewes (Ovisaries). Applied Animal Behaviour Science 32, 45–54.

Matteri, R.L., Carroll, J.A. and Dyer, C.J. (2000) Neuroendocrine response to stress. In: Mench, J.A. and Moberg, G. (eds) Biologyof Animal Stress. CAB International, Wallingford, UK, pp. 43–76.

Matthews, L.R. (1993) Deer handling and transport. In: Grandin, T. (ed.) LivestockHandling and Transport. CAB International, Wallingford, UK, pp. 253–272.

Matthews, L.R. and Ladewig, J. (1994) Envir-onmental requirements of pigs measured by behavioural demand functions. AnimalBehaviour 47, 713–719.

McFarland, D. (1981) The Oxford Companion to Animal Behaviour. Oxford University Press, Oxford, UK.

McGuire, W.J. (1964) Inducing resistance to persuasion: some contemporary approa-ches. In: Berkowitz, L. (ed.) Advances in Experimental Social Psychology, Vol. 1. Academic Press, New York.

Mears, G.J., Brown, F.A. and Redmond, L.R. (1999) Effects of handling, shearing and previous exposure to shearing on cortisol and β-endorphin responses in ewes. Cana-dian Journal of Animal Science 79, 35–38.

Mellor, D.J. and Littin, K.E. (2004) Using science to support ethical decisions pro-moting humane livestock slaughter and vertebrate pest control. Animal Welfare 13, 127–132.

Mellor, D.J. and Stafford, K.J. (2000) Acute cas-tration and/or tailing distress and its allevia-tion in lambs. New Zealand Veterinary Journal 48, 33–43.

Mellor, D.J., Cook, C.J. and Stafford, K.J. (2000) Quantifying some responses to pain as a stressor. In: Mench, M. and Moberg, G.O. (eds) Biology of Animal Stress. CAB International, Wallingford, UK, pp. 171–198.

Mench, J.A. (2008) Farm animal welfare in the USA: farming practices, research, educa-tion, regulation and assurance programs.

Applied Animal Behaviour Science 113, 298–312.

Mendl, M. (1999) Performing under pressure: stress and cognitive function. Applied Ani-mal Behaviour Science 65, 221–244.

Miller, D. (1995) Evaluating stockmanship capabilities in employees. Professional Swine Management Certifi cation Series, Swine Breeding Herd Management, Module 3 Stockmanship, Part H. Iowa Pork Industry Center, Ames, Iowa, pp. 1–4.

Mills, A.D. and Faure, J.M. (1990) Panic and hysteria in domestic fowl: a review. In: Zayan, R. and Dantzer, R. (eds) SocialStress in Domestic Animals. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 248–272.

Mills, A.D., Beilharz, R.G. and Hocking, P.M. (1997) Genetic selection. In: Appleby, M.C. and Hughes, B.O. (eds) Animal Welfare.CAB International, Wallingford, UK, pp. 205–217.

Ministry of Agriculture, Fisheries and Food (MAFF) (1983) British Codes of Recommendations for the Welfare of Livestock. Her Majesty’s Stationery Offi ce, London.

Moberg, G.P. (2000) Biological response to stress: implications for animal welfare. In: Mench, J.A. and Moberg, G. (eds) Biologyof Animal Stress. CAB International, Wallingford, UK, pp. 1–21.

Moe, R.O., Nordgreen, J., Janczak, A.M., Spruijt, B.M., Zanella, A.J. and Bakken, M. (2009) Trace classical conditioning as an approach to the study of reward-related behaviour in laying hens: a methodological study. Applied Animal Behaviour Science121, 171–178.

Mounier, L., Dubroeucq, L., Andanson, S. and Veissier, I. (2006) Variations in meat pH of beef bulls in relation to conditions of trans-fer to slaughter and previous history of the animals. Journal of Animal Science 84, 1567–1576.

Munksgaard, L., de Passille, A.M., Rushen, J., Thodberg, K. and Jensen, M.B. (1995) The ability of dairy cows to distinguish between people. Proceedings of the 29th International Congress of the International

182 References

Society of Applied Ethology, pp. 19–20 (abstract).

Murphey, R.M., Moura Duarte, F.A. and Torres Penendo, M.C. (1981) Responses of cattle to humans in open spaces: breed comparisons and approach–avoidance relationships. Behaviour Genetics 2, 37–47.

Murphy, L.B. (1976) A study of the behavioural expression of fear and exploration in two stocks of domestic fowl. PhD thesis. Edinburgh University, Edinburgh, UK.

Murphy, L.B. (1978) The practical problems of recognizing and measuring fear and exploration behaviour in the domestic fowl. Animal Behaviour 26, 422–431.

Murphy, L.B. and Duncan, L.J.H. (1977) Attempts to modify the responses of domestic fowl towards human beings. 1. The association of human contact with a food reward. Applied Animal Ethology 3, 321–334.

Murphy, L.B. and Duncan, L.J.H. (1978) Attempts to modify the responses of domestic fowl towards human beings. II. The effect of early experience. AppliedAnimal Ethology 4, 5–12.

Murray, R.K., Granner, D.K., Mayes, P.A. and Rodwell, V.W. (2003) Harper’s Illustrated Biochemistry. McGraw-Hill, New York.

Myers, I. Briggs and Myers, P.B. (1995) Gifts Differing: Understanding Personality Type. Davies-Black Publishing, Palo Alto, California.

NTIS (National Training Information Service) (2010) Comprehensive information for the training sector specialist. Available at: www.ntis.gov.au (accessed 28 April 2010).

OIE (World Animal Health Organization) (2008) Terrestrial Animal Health Code, 17th edn. Oxford University Press, New York.

Olson, J.C. and Mitchell, A.A. (1975) The pro-cess of attitude acquisition: the value of a developmental approach to consumer atti-tude research. Advances in Consumer Research 2, 249–264.

Olson, J.M. and Stone, J. (2005) The infl uence of behavior on attitudes. In: Albarracin, D., Johnson, B.T. and Zanna, M.P. (eds) TheHandbook of Attitudes. Psychology Press, New York, pp. 223–271.

Pajor, E.A., Rushen, J. and de Passille, A.M.B. (2000) Aversion learning techniques to

evaluate dairy cattle handling practices. Applied Animal Behaviour Science 69, 89–102.

Panksepp, J. (1998) Affective Neuroscience.The Foundation of Human and Animal Emotions. Oxford University Press, New York.

Panksepp, J. (2005) Affective consciousness: core emotional feelings in animals and humans. Consciousness and Cognition 14, 30–80.

Parker, K.J., Schatzberg, A.F. and Lyons, D.M. (2003) Neuroendocrine aspects of hyper-cortisolism in major depression. Hormonesand Behaviour 43, 60–66.

Paterson, A.M. and Pearce, G.P. (1989) Boar-induced puberty in gilts handled pleasantly or unpleasantly during rearing. Applied Animal Behaviour Science 22, 225–233.

Paterson, A.M. and Pearce, G.P. (1992) Growth, response to humans and corticosteroids in male pigs housed individually and subjected to pleasant, unpleasant or minimal handling during rearing. AppliedAnimal Behaviour Science 34, 315–328.

Pavlov, I.P. (1960) Conditioned Refl exes: an Investigation of the Physiological Activity of the Cerebral Cortex. Dover Publications Inc., New York.

Pearce, G.P., Paterson, A.M. and Pearce, A.N. (1989) The infl uence of pleasant and unpleasant handling and the provision of toys on the growth and behaviour of male pigs. Applied Animal Behaviour Science23, 27–37.

Pedersen, L.J., Jensen, M.B., Hansen, S.W., Munksgaard, L., Ladewig, J. and Matthews, L. (2002a) Social isolation affects the motivation to work for food and straw in pigs as measured by operant con-ditioning techniques. Applied Animal Behaviour Science 77, 295–309.

Pedersen, L.J., Moeller, N.H. and Jeppesen, L.L. (2002) Behavioural and physiological effects of post-weaning handling and access to shelters in farmed blue foxes (Alopex lagopus). Applied Animal Behav-iour Science 77, 139–154.

Pedersen, L.J., Damm, B.I. and Kongsted, A.G. (2003) The infl uence of adverse or gentle

References 183

handling procedures on sexual behaviour in fearful and confi dent sows. Applied Ani-mal Behaviour Science 83, 277–290.

Pedersen, V. (1993) Effects of difference post-weaning handling procedures on the later behaviour of silver foxes. Applied Animal Behaviour Science 37, 239–250.

Pedersen, V. and Jeppesen, L.L. (1990) Effects of early handling on better behaviour and stress responses in the silver fox (Vulpes vulpes). Applied Animal Behaviour Science 26, 383–393.

Pedersen, V., Barnett, J.L., Hemsworth, P.H., Newman, E.A. and Schirmer, B. (1998) The effects of handling on behavioural and physiological responses to housing in tether-stalls in pregnant pigs. Animal Wel-fare 7, 137–150.

Personnel Decisions Inc (1996) PDI Employ-ment Inventory. Minneapolis, Minnesota.

Petherick, J.C., Holroyd, R.G., Doogan, V.J. and Venus, B.K. (2002) Productivity, car-cass and meat quality of lot-fed Bos indicuscross steers grouped according to tempera-ment. Australian Journal of Agricultural Research 42, 389–398.

Petherick, J.C., Doogan, V.J., Venus, B.K., Hol-royd, R.G. and Olsson, P. (2009a) Quality of handling and holding yard environment, and beef cattle temperament: 1. Relation-ships with fl ight speed and fear of humans. Applied Animal Behaviour Science 120, 18–27.

Petherick, J.C., Doogan, V.J., Venus, B.K., Hol-royd, R.G. and Olsson, P. (2009b) Quality of handling and holding yard environment, and beef cattle temperament: 2. Conse-quences for stress and productivity. AppliedAnimal Behaviour Science 120, 28–38.

Phillips, J.C. and Benson, J.E. (1983) Some aspects of job satisfaction in the Soviet Union. Personnel Psychology 36,633–645.

Phillips, P.A., Thompson, B.K. and Fraser, D. (1988) Preference tests of ramp designs for young pigs. Canadian Journal of Animal Science 68, 41–48.

Phillips, P.A., Fraser, D. and Thompson, B.K. (1991) Preference by sows for a partially enclosed farrowing crate. Applied Animal Behaviour Science 32, 35–43.

Phillips, P.A., Fraser, D. and Thompson, B.K. (1996) Sow preference for types of fl ooring in farrowing crates. Canadian Journal of Animal Science 76, 485–489.

Pianka, E.R. (1974) Evolutionary Ecology.Harper and Row, New York.

Pollard, J.C., Littlejohn, R.P. and Suttie, J.M. (1994) Responses of red deer to restraint in a Y maze preference test. Applied Animal Behaviour Science 39, 63–71.

Price, E.O. (1984) Behavioral aspects of animal domestication. Quarterly Review of Biol-ogy 59, 1–32.

Price, E.O. (2002) Animal Domestication and Behaviour. CAB International, Wallingford, UK.

Provenza, F.D. (1995) Postingestive feedback as an elementary determinant of food prefer-ence and intake in ruminants. Journal of Range Management 48, 2–17.

Purcell, D., Arave, C.W. and Walters, J.L. (1988) Relationship of three measures of behaviour to milk production. AppliedAnimal Behaviour Science 21, 307–313.

Ravel, A., D’Allaire, S.D., Bigras-Poulin, M. and Ward, R. (1996) Personality traits of stockpeople working in farrowing units on two types of farms in Quebec. In: Proceedings of the 14th Congress of the International Pig Veterinary Society, C,Bologna, Italy. Faculty of Veterinary Medicine, Unversity of Bologna, Bologna, p. 514.

Reber, A.S. (1988) Dictionary of Psychology.Penguin Books, London.

Reichmann, K.G., Barram, K.M., Brock, I.J. and Standfast, N.F. (1978) Effects of regu-lar handling and blood sampling by wing vein puncture on the performance of broi-lers and pullets. British Poultry Science 19, 97–99.

Reisberg, D. and Heuer, F. (1995) Emotion’s multiple effects on memory. In: McGaugh, J.L., Weiberger, N. and Lynch, G. (eds) Brain and Memory: Modulation and Medi-ation of Plasticity. Oxford University Press, New York, pp. 84–92.

Rokeach, M. (1960) The Open and Closed Mind: Investigations into the Nature of Belief Systems and Personality Systems.Basic Books, New York.

184 References

Ruis, M., Coleman, G.J., Waiblinger, S., Winschnurer, I. and Boivin, X. (2010) A multimedia- based cognitive-behavioural intervention program improves attitudes and handling behaviours of stockpeople in lives-tock farming. In: Proceedings of the British Society of Animal Science and Agricultural Research Forum. Cambridge University Press, Cambridge, UK, p. 175.

Rusbult, C.E., Farrell, D., Rogers, G. and Main-ous, A.G. III (1988) Impact of exchange variables on exit, voice, loyalty, and neglect: an in tegrative model of respo nsesto declining jo b sati sfaction. Academy of Management Jou rnal 31, 599–627.

Rushen, J. (1 986) Avers ion of sheep to handling treatments: paired-choice studies. Applied Animal Behaviour Science 16, 363–370.

Rushen, J. and Congdon, P. (1986) Relative aversion of sheep to simulated shearing with and without electroimmobilization. Australian Journal of Experimental Agri-culture 26, 535–537.

Rushen, J., Munksgaard, L., de Passille, A.M.B., Jensen, M.B. and Thodberg, K. (1995) Location of handling and dairy cows’ abil-ity to discriminate between gentle and aversive handlers. In: Proceedings of the 29th International Congress of the Interna-tional Society for Applied Ethology, Exeter, UK. University Federation of Animal Wel-fare, Potter’s Bar, UK.

Rushen, J., de Passille, A.M.B. and Munks-gaard, L. (1999) Fear of people by cows and effects on milk yield, behaviour and heart rate at milking. Journal of Dairy Sci-ence 82, 720–727.

Rushen, R., de Passille, A.M., Munksgaard, L. and Tanida, H. (2001) People as social actors in the world of farm animals. In: Keeling, L.J. and Gonyou, H.W. (eds) Social Behaviour in Farm Animals. CAB International, Wallingford, UK, pp. 353–372.

Rutter, S.M., Orr, R.J., Yarrow, N.H. and Cham-pion, R.A. (2004) Dietary preference of dairy heifers grazing ryegrass and white clover, with and without and anti-bloat treatment. Applied Animal Behaviour Sci-ence 85, 1–10.

Saadoun, A., Simon, J.R. and Leclercq, B. (1987) Effect of exogenous corticosterone

in genetically fat and lean chickens. BritishPoultry Science 28, 519–528.

Sandøe, P., Forkman, F. and Christiansen, S.B. (2004) Scientifi c uncertainty – how should it be handled in relation to scientifi c advice regarding animal welfare issues? AnimalWelfare 13, S121–S126.

Sapolsky, R.M. (1992) Neuroendocrinology of the stress-response. In: Becker, J.B., Breed-love, S.M. and Crews, D. (eds) BehavioralEndocrinology. MIT Press, Cambridge, Massachusetts, pp. 287–324.

Schaefer, T. (1968) Some methodological implications of the research on “early handling” in the rat. In: Newton, G. and Levine, S. (eds) Early Experience and Behaviour: the Psychobiology of Develop-ment. Charles C. Thomas Publisher, Springfi eld, Illinois.

Schmied, C., Waiblinger, S., Scharl, T., Leisch, F. and Boivin, X. (2008) Stroking of different body regions by a human: effects on behaviour and heart rate of dairy cows. Applied Animal Behaviour Science 109, 25–38.

Scott, J.P. (1992) The phenomenon of attachment in human–non human relationships. In: Davis, H. and Balfour, A.D. (eds) The Inevitable Bond – Examining Scientist–Animal Interactions. Cambridge University Press, Cambridge, UK, pp. 72–92.

Seabrook, M.F. (1972a) A study to determine the infl uence of the herdsman’s personality on milk yield. Journal of Agriculture Labour Science 1, 45–59.

Seabrook, M.F. (1972b) A study of the infl uence of the cowman’s personality and job satisfaction on milk yield of dairy cows. In: Joint Conference of the British Society for Agriculture Labour Science and the Ergonomics Research Society, September 1972. National College of Agricultural Engineering, Silsoe, UK.

Seabrook, M.F. (1982) Motivation and perfor-mance. Ergonomics 25, 65–72.

Seabrook, M. (1994) The effect of production systems on the behaviour and attitudes of stockpersons. In: 4th Zodiac Symposium.Publication No. 67, Wageningen Academic Publishers, Wageningen, The Netherlands, pp. 252–258.

References 185

Seabrook, M. (1996) The Role of the Human Factor in Increasing Performance and Profi tability in Pig Production. Filozzo Rhone Poulenc Conference, Moderna, Italy, pp. 2–13.

Seabrook, M.F. and Bartle, N.C. (1992) The practical implications of animals’ responses to man. In: Winter Meeting, British Society of Animal Production, 23–25 March 1992, Scarborough, UK, Paper Number 34.

Segundo, R.C. (1989) A study of stockpeople and managers in the pig industry with special emphasis on the factors affecting their job satisfaction. MSc thesis. University of Aberdeen, Aberdeen, Scotland.

Selye, H. (1946) The general adaptation syndrome and the diseases of adaptation. Journal of Clinical Endocrinology 6, 117–230.

Selye, H. (1976) Stress in Health and Disease.Butterworth, Boston, Massachusetts.

Serpell, J.P. (2004) Factors affecting human atti-tudes to animals and their welfare. AnimalWelfare 13, S145–S151.

Sherwin, C.M. (2003) Social context affects the motivation of laboratory mice, Mus muscu-lus, to gain access to resources. AnimalBehaviour 66, 649–655.

Siegel, H.S. and van Kampen, M. (1984) Energy relationships in growing chickens given daily injections of corticosterone. British Poultry Science 25, 477–485.

Singer, P. (1975) Animal Liberation. Avon Books, New York.

Singh, S. (1983) Effect of motivation, values, cognitive factors and child rearing among Punjab farmers. Journal of Social Psychol-ogy 120, 273–278.

Skinner, B.F. (1969) Contingencies of Reinforce-ment: a Theoretical Analysis. Prentice-Hall Inc., Englewood Cliffs, New Jersey.

Stagner, R. (1961) Psychology of Personality.McGraw-Hill, New York.

Stevens, B., Barnett, J.L., Tilbrook, A. and Hemsworth, P.H. (2009) Effects of depriva-tion of a preferred resource (feed or social contact) on the biological functioning of pigs. In: van Barneveld, R.J. (ed.) Manipu-lating Pig Production XII. Australasian Pig Science Association, Werribee, Victoria, Australia, p. 28.

Suarez, S.D. and Gallup, G.G. (1982) Open-fi eld behaviour in chickens: the experimenter is a predator. Journal of Comparative and Physiological Psychology 96, 432–439.

Sutton, J.R. and Casey, J.H. (1975) The adrenocortical response to competitive athletics in veteran althletes. Journal of Clinical Endocrinology and Metabolism40, 135–138.

Syme, L.A., Durham, I.H. and Elphick, G.R. (1981) Microprocessor control of sheep movement. In: Proceedings of the 2nd Con-ference on Wool Harvesting Research and Development. Australian Wool Corporation, Sydney, New South Wales, pp. 237–245.

Szechtman, H., Lambrou, P.J., Caggiula, A.R. and Redgate, E.S. (1974) Plasma corticoste-rone levels during sexual behaviour in male rat. Hormones and Behaviour 5, 191–200.

Tallet, C., Veissier, I. and Boivin, X. (2005) Human contact and feeding as rewards for the lambs’ affi nity to their stockperson. Applied Animal Behaviour Science 94, 59–73.

Tanida, H., Miura, A., Tanaka, T. and Yoshimoto, T. (1995) Behavioural response to humans in individually handled weanling pigs. Applied Animal Behaviour Science 42, 249–259.

Teutsch, G.M. (1987) Intensive farm animal management seen from an ethical stand-point. In: von Loeper, E., Martin, G., Muller, J., Nabholz, A. and van Putten, G. (eds) Ethical, Ethological and Legal Aspects of Intensive Farm Animal Management.Tierhaltung, 18, 9–40.

Thompson, C.I. (1976) Growth in the Hubbard broiler: increase size following early handling. Developmental Psychobiology 9, 459–464.

Thorpe, W.H. (1965) The assessment of pain and distress in animals. In: Brambell, F.W.R. (ed.) Report of the Technical Committee to Enquire into the Welfare of Animals kept under Intensive Livestock Systems. Com-mand paper 2836. Her Majesty’s Stationery Offi ce, London, pp. 71–79.

Tilbrook, A.J., Turner, A.I. and Clarke, I.J. (2000) Effects of stress on reproduction in non-rodent mammals: considerations of the role of glucocorticoids and sex differ-ences. Reviews of Reproduction 5, 105–113.

186 References

Toates, F.M. (1980) Animal Behaviour – A Systems Approach. John Wiley & Sons, Chichester, West Sussex, UK.

Toates, F. (2004) Cognition, motivation, emo-tion and action: a dynamic and vulnerable interdependence. Applied Animal Behav-iour Science 86, 173–204.

Tosi, M.V. and Hemsworth, P.H. (2002) Stockperson–husbandry interactions and animal welfare in the extensive livestock industries. In: Proceedings of the 36th Con-gress of the International Society for Applied Ethology. Utrecht, The Nether-lands, p. 129 (abstract).

Turk, D.C. and Salovey, P. (1985a) Cognitive struc-tures, cognitive processes, and cognitive-behavior modifi cation: I. Client issues.Cognitive Therapy and Research 9, 1–17.

Turk, D.C. and Salovey, P. (1985b) Cognitive structures, cognitive processes, and cognitive-behavior modifi cation: II. Judge-ments and inferences of the clinician. Cog-nitive Therapy and Research 9, 19–33.

UK Farm Animal Welfare Council (1993) Sec-ond Report on Priorities for Research and Development in Farm Animal Welfare.Ministry of Agriculture, Fisheries and Food (MAFF), London.

Vandenheede, M. and Bouissou, M.F. (1993) Sex differences in fear reactions in sheep. Applied Animal Behaviour Science 37, 39–55.

Vielle-Thomas, R.K. and Signoret, J.P. (1992) Pheromonal transmission of aversive experiences in domestic pigs. Journal of Chemical Endocrinology 18, 1551 (abstract).

Voith, V.L. (1985) Attachment of people to companion animals. Veterinary Clinics of North America 15, 289–295.

Waiblinger, S., Menke, C. and Coleman, G. (2002) The relationship between attitudes, personal characteristics and behavior of stockpeople and subsequent behavior and production of dairy cows. Applied Animal Behaviour Science 79, 195–219.

Waiblinger, S., Menke, C. and Folsch, D.W. (2003) Infl uences on the approach and avoidance behaviour of dairy cows at 35 farms. Applied Animal Behaviour Science84, 23–39.

Waiblinger, S., Menke, C., Korff, J. and Bucher, A. (2004) Previous handling and gentle interactions affect behaviour and heart rate of dairy cows during a veterinary procedure. Applied Animal Behaviour Science 85, 31–42.

Waiblinger, S., Boivin, X., Pedersen, V., Tosi, M.-V., Janczak, A.M., Visser, E.K. and Jones, R.B. (2006) Assessing the human–animal relationship in farmed species: a critical review. Applied Animal Behaviour Science 101, 185–242.

Walker, S. (1987) Animal Learning: an Introduction. Routledge and Kegan Paul Ltd, London.

Walsh, F. (2009) Human–animal bonds II: the roles of pets in family systems and family therapy. Family Process 48, 481–489.

Warburton, H. and Mason, G. (2003) Is out of sight out of mind? The effects of resource cues on motivation in mink, Mustela vison.Animal Behaviour 65, 755–762.

Waynert, D.F., Stookey, J.M., Schwartzkopf-Genswein, K.S. and Webster, A.J.F. (1999) The response of beef cattle to noise during handling. Applied Animal Behaviour Sci-ence 62, 27–42.

Wells, D.L. (2007) Domestic dogs and human health: an overview. British Journal of Health Psychology 12, 145–156.

Westphal, U. (1971) Steroid–Protein Interac-tions. Springer-Verlag, Berlin.

Widowski, T.M. and Hemsworth, P.H. (2008) Housing hens to suit their ‘needs’. In: Pro-ceedings of the XXIII World Poultry Con-gress. World’s Poultry Science Journal and the World’s Poultry Science Association, The Netherlands, p. 241.

Windschnurer, I., Boivin, X. and Waiblinger, S. (2009a) Reliability of an avoidance dis-tance test for the assessment of animals’ responsiveness to humans and a prelimi-nary investigation of its association with farmers’ attitudes on bull fattening farms. Animal Behaviour Science 117, 117–127.

Windschnurer, I., Boivin, X., Coleman, G., Ruis, M., Mounaix, B., Brule, A. and Waiblinger, S. (2009b) A multimedia-based cognitive-behavioural intervention program improves attitudes and handling behaviours of

References 187

stockpeople on dairy farms. In: Proceedings from the International Society for Applied Ethology, Regional Meeting East and West Central Europe. International Society for Applied Ethology (ISAE), Vienna.

Wu, Y.B. and Ravindran, V. (2004) Infl uence of whole wheat inclusion and xylanase supplementation on the performance, digestive tract measurements and carcass characteristics of broiler chickens. AnimalFeed Science and Technology 116, 129–139.

Zulkifl i, I. and Siti Nor Azah, A. (2004) Fear and stress reactions, and the performance of commercial broiler chickens subjected to regular pleasant and unpleasant contacts with human beings. Applied Animal Behaviour Science 88, 77–87.

Zulkifl i, I., Gilbert, P.K., Liew, J. and Ginos, J. (2002) The effects of regular visual contact with human beings on fear, stress, antibody and growth responses in broiler chickens. Applied Animal Behaviour Science 79,103–112.

This page intentionally left blank

189189

Index

16 PF test 94adrenocorticotrophic hormone (ACTH)

32–34, 76affective states 37–42Ajzen and Fishbein 8, 90–92, 95–96,

122, 154animal–human interactions see

stockperson behaviour–animal behaviour relationship

animal welfare, see welfareattitude 26, 84, 85–86, 88–92, 95,

128–134, 164–165affect 86, 90–92, 94–95, 136–137change 87–88, 95, 137–139,

140–142cognition 86, 88, 89, 90–92,

96–102, 121–122, 141–151cognitive–behavioural intervention

techniques 138–139, 140–152, 161–162

cognitive dissonance theory 88, 137–140

conation 86, 89, 90–92, 96–102, 121–122

development 87–88, 95, 136function 87handling animals 92, 95–102maintenance 135measurement 86, 95–98modifi cation 100, 129, 136–139,

142–152

productivity 7, 128–130theory of reasoned action 8, 90–92,

95–96, 122, 154see also personality

attitudes and behaviour 88–92, 95–102auditory contact 73–74, 96–102,

113–114, 126, 156automation 154, 156, 159–161

behaviourattitudes and behaviour 88–92,

95–102, 121–122, 128–134, 135

auditory contact 73–74, 96–102, 113–114, 126, 156

development 135–136effects on fear see handlingexpression of normal or ‘natural’

behaviours 43–45modifi cation 129, 138–139,

140–152, 161–162olfactory contact 73–74prediction 88–92stockperson attitude–behaviour

relationship 88–92, 95–102stockperson behaviour

effects of stockperson training 142–151

effects of stockperson selection 3, 154

190 Index

behaviour continuedstockperson behaviour continued

measurement 95–102variation in farm animals 129

stress response 36–37tactile contact 48, 53, 62–70, 77, 80,

96–102, 108–114, 121–126, 127, 157

visual contact 48, 62, 70–73, 77, 124–126, 156

see also stockperson behaviour animal behaviour relationship

behaviour modifi cation 38–39, 138–139, 142–151

cognitive–behavioural intervention techniques 138–139, 140–152, 161–162

behavioural responses to humans 36–37, 50–51, 55–58, 75, 78, 107–116, 117–119, 126–128, 142–151, 157–159

biological functioning 29–37

catecholamines 32, 56, 75–76, 127adrenalin 31, 38, 54, 56, 76, 107noradrenalin 31, 54, 56, 107

cognitive–behavioural intervention techniques 138–139, 140–152, 161–162

cognitive–behavioural training 138–139, 140–152, 161–162

cognitive dissonance theory 88, 137–140corticosteroids 32–34, 56, 76

corticosterone 32, 72, 79, 107cortisol 3–4, 32, 34, 56, 59, 79–80,

107, 127, 159, 160

emotions 35–36, 37–42, 54–55empathy 9, 25, 94–95, 131–133employment 6–14, 154, 164–165

fear 35–36general fearfulness 50–51, 81fear of humans see fear of humans

fear of humans 50, 53–54, 82–83, 120, 133–134

behavioural responses 36–37, 50–51, 55–58, 75, 78,

107–116, 117–119, 126–128, 142–151, 157–159

defi nition 54–55effects of handling 51–52, 142–144effects of handling during rearing

51–52, 65–70, 71–72effects on productivity 74, 116–117,

126–128beef cattle 81–82dairy cattle 79–81, 128goats 81pigs 57–58, 74–76, 143–145poultry 58, 76–79, 127

effects on welfare 82–83, 117, 126–128

levels in farm animals 50–53, 108–114

measurement of 58physiological responses 52, 54,

55–58, 74, 159–160reduction 52–53, 55, 63–70, 71–72,

76–77, 80, 110–111, 120, 167

Festinger 88, 137–140fl ight

distance 50–51, 60, 99, 124, 149speed 81–82

handling 3–4, 62–63, 117–119, 120, 122–126, 157–161, 167

attitudes 92, 95–102auditory contact 73–74, 96–102,

113–114, 126, 156effects on fear 51–52, 142–144negative 7, 48, 52–53, 63–64, 71,

74–75, 77, 79–80, 96–102, 108–114, 114–116, 118–119, 121–124, 126–128, 142–144

olfactory contact 73–74positive 52–53, 63–70, 71–72,

76–77, 80, 96–102, 108, 110–113, 114–116, 118–119, 123–124, 142–144

tactile contact 48, 53, 62–70, 77, 80, 96–102, 108–114, 121–126, 127, 157

visual contact 48, 62, 70–73, 77, 124–126, 156

homeostasis 30–31, 34human–animal interactions 1–2, 3–4,

Index 191

21–24, 47–50, 82, 84, 103–104, 108–114, 120, 122–126, 128–133, 153–154, 167

animal discrimination of humans 80, 114–116

behavioural responses of animals 36–37, 50–51, 55–58, 75, 78, 107–116, 117–119, 126–128, 142–151, 157–159

ease of handling 117–119conditioned responses to humans

51–52classical conditioning 49–50,

51–53, 62–63, 73, 79–77, 105, 136–137, 157–159

conditioned approach–avoidance response 50–53, 59–60, 63–64, 74, 78, 79–80, 112–113, 122–126, 144

empathy 9, 25, 94–95, 131–133habituation to humans 52, 70–72,

125–126see also learning

human behaviours 3–4, 62–63, 117–119, 120, 122–126, 157–161, 167

auditory 73–74, 96–102, 113–114, 126, 156

effects on fear 51–52, 142–144negative 7, 48, 52–53, 63–64,

71, 74–75, 77, 79–80, 96–102, 108–114, 114–116, 118–119, 121–124, 126–128, 142–144

olfactory 73–74positive 52–53, 63–70, 71–72,

76–77, 80, 96–102, 108, 110–113, 114–116, 118–119, 123–124, 142–144

tactile 48, 53, 62–70, 77, 80, 96–102, 108–114, 121–126, 127, 157

visual 48, 62, 70–73, 77, 124–126, 156

model 128–133, 135reciprocal 135, 137–138sequential 128–129, 135

human behaviour 9–10, 49–50, 62–63

attitudes and behaviour 88–92, 95–102, 121–122, 128–134, 135

auditory contact 73–74, 96–102, 113–114, 126, 156

behaviour modifi cation 38–39, 129, 138–139, 142–151

cognitive–behaviouralintervention techniques 138–139, 140–152, 161–162

development 135–136effects on fear see handlingolfactory contact 73–74tactile contact 48, 53, 62–70, 77, 80,

96–102, 108–114, 121–126, 127, 157

visual contact 48, 62, 70–73, 77, 124–126, 156

see also stockperson and stockperson behaviour–animal behaviour relationship

human–farm animal interactions 47–49effects of stockperson selection, 6–14,

154effects of stockperson training

142–151models 128–133necessity of 156–161see also handling and stockperson

behaviour–animal behaviour relationship

human resourcesabsenteeism 11–14evaluating performance 14–16recruitment 6–14, 154, 164–165training 2–3, 16–19, 129, 131,

151–152, 153–154, 161–164cognitive–behavioural training/

intervention 138–139, 140–152, 161–162

low stress handling training 151hypothalamic–pituitary–adrenal (HPA)

axis 31–37, 75–76

immune function 76–77, 127–128

job satisfaction 12–14, 130–131, 148, 165–167

192 Index

learning 103–104classical conditioning 49–50, 51–53,

62–63, 73, 79–77, 105, 136–137, 157–159

conditioned approach–avoidance response 50–53, 59–60, 63–64, 74, 78, 79–80, 112–113, 122–126, 144

habituation 52, 70–72, 125–126instrumental conditioning 53, 63, 64,

72–73, 105–106, 124motivation 104–105operant conditioning, see learning,

instrumental conditioningreinforcement 104–105, 136reinforcers

negative 52–53, 63, 104, 108positive 52–53, 104, 108primary 53, 107secondary 107

sensitization 72, 160stimulus 49–50, 54, 58–59, 60–62

conditioned 51, 59–60, 105–106

generalization 107, 114–116unconditioned 51–52, 56,

105–106stimulus-response 51–52, 56, 58,

104–105legislation 25, 26, 27, 162–163low stress handling 151

modelshuman–animal interactions 128–133job commitment 12theory of reasoned action 8, 90–92,

95–96, 122, 154motivation 24, 54, 104–105

work 9–10, 130Myer-Briggs type indicator (MBTI) 8,

93–94

negative fear–productivity relationship see fear of humans, effects on productivity

negative handling 7, 48, 52–53, 63–64, 71, 74–75, 77, 79–80, 96–102, 108–114, 114–116, 118–119, 121–124, 126–128, 142–144

olfactory contact 73–74opinions see attitude

PDI employment inventory (PDI-EI)PDI-EI performance 15PDI-EI tenure 15–16

personality 7, 95, 154, 164–16516 PF test 94authoritarianism 87, 139‘big fi ve’ measures 8, 16dogmatism 139effect on productivity 8–9, 93–95,

154empathy 9, 25, 94–95, 131–133extraversion 8, 93fi ve factor theory 8introversion 8, 93Myer-Briggs type indicator (MBTI) 8,

93–94see also attitude

personality traits see personalitypositive handling 52–53, 63–70, 71–72,

76–77, 80, 96–102, 108, 110–113, 114–116, 118–119, 123–124, 142–144

preference testing 39–42productivity 1, 22–23

effect of attitudes and behaviour 7, 128–130

effect of personality 8–9, 93–95, 154effect of stockperson 3–4, 20,

126–128fear of humans 74, 116–117,

126–128beef cattle 81–82dairy cattle 79–81, 128goats 81pigs 57–58, 74–76, 143–145poultry 58, 76–79, 127

public opinion 21, 22–23, 27–28

reproduction 74, 127, 144, 145–146

selection 6–14, 154, 164–165sixteen personality factor questionnaire

(16PF) 94skills 4–5, 7, 129–131socialisation 66–70

Index 193

stereotypes 87–88stockperson 1–2, 153–154

absenteeism 11–14attitude 95–102, 128–133, 135

see also attitudebehaviour

effects of stockperson selection 3, 154

effects of stockperson training 142–151

measurement 95–102variation in farm animals 129see also human behaviour

effects on productivity 3–4, 93see also productivity

evaluating performance 14–16gender 15–16, 98image and self-esteem 2–3, 6, 155job characteristics 4–6, 167–168job commitment 11–12job satisfaction 12–14, 130–131,

148, 165–167job turnover 4, 11–14, 15–16, 148knowledge 4–5, 7, 129–131, 154personality 7, 95, 154, 164–165

authoritarianism 87, 139dogmatism 139effect on productivity 8–9,

93–95, 154empathy 9, 25, 94–95, 131–133extraversion 8, 93introversion 8, 93

role 2–4, 156–157selection 6–14, 154, 164–165skills 4–5, 7, 129–131training 2–3, 16–19, 129, 131,

151–152, 153–154, 161–164cognitive–behavioural training/

intervention 138–139, 140–152, 161–162

low stress handling training 151work benefi ts 87, 130–131, 165–167work conditions 5–6, 129–130, 131work motivation 9–10, 130

stockperson attitude–behaviour relationship 88–92, 95–102

stockperson behaviour–animal behaviour relationship 3–4, 21–24, 47–50, 82, 84, 103–104, 108–114, 120, 122–126, 128–133, 153–154, 167

animal discrimination of humans 80, 114–116

behavioural responses of animals 36–37, 50–51, 55–58, 75, 78, 107–116, 117–119, 126–128, 142–151, 157–159

ease of handling 117–119classical conditioning 49–50, 51–53,

62–63, 73, 79–77, 105, 136–137, 157–159

conditioned approach–avoidance response 50–53, 59–60, 63–64, 74, 78, 79–80, 112–113, 122–126, 144

conditioned responses to humans 51–52

empathy 9, 25, 94–95, 131–133habituation to humans 52, 70–72,

125–126see also learning

human behaviours 3–4, 62–63, 117–119, 120, 122–126, 157–161, 167

auditory 73–74, 96–102, 113–114, 126, 156

effects on fear 51–52, 142–144negative 7, 48, 52–53, 63–64,

71, 74–75, 77, 79–80, 96–102, 108–114, 114–116, 118–119, 121–124, 126–128, 142–144

olfactory 73–74positive 52–53, 63–70, 71–72,

76–77, 80, 96–102, 108, 110–113, 114–116, 118–119, 123–124, 142–144

tactile 48, 53, 62–70, 77, 80, 96–102, 108–114, 121–126, 127, 157

visual 48, 62, 70–73, 77, 124–126, 156

model 128–133, 135reciprocal 135, 137–138sequential 128–129, 135

stress 30–31, 82–83acute stress response 33, 34, 37,

56–57, 79, 117catecholamines 32, 56, 75–76, 127

194 Index

stress continuedcatecholamines continued

adrenalin 31, 38, 54, 56, 76, 107

noradrenalin 31, 54, 56, 107chronic stress response 33–34,

57–58, 75–76, 79, 116–117, 126–128

corticosteroids 32–34, 56, 76corticosterone 32, 72, 79, 107cortisol 3–4, 32, 34, 56, 59,

79–80, 107, 127, 159, 160

effects on animal productivity 36–37, 74, 116–117, 126–128

beef cattle 81–82dairy cattle 79–81, 128goats 81pigs 57–58, 74–76, 143–145poultry 58, 76–79, 127

effects on animal welfare 34–37fi ght-fl ight syndrome 32, 56hypothalamic-pituitary-adrenal (HPA)

axis 31–37, 75–76measurement 34–37, 58–62

sympathetic-adrenal-medullary (SAM) axis 31–33, 37

see also stress

tactile contact 48, 53, 62–70, 77, 80, 96–102, 108–114, 121–126, 127, 157

theory of planned behaviour 92, 154behavioural intention 90–91perceived behavioural control 92subjective norms 90–91

theory of reasoned action 8, 90–92, 95–96, 122, 154

training 2–3, 16–19, 129, 131, 151–152, 153–154, 161–164

cognitive–behavioural training/intervention 138–139, 140–152, 161–162

low stress handling training 151turnover 4, 11–14, 15–16, 148

visual contact 48, 62, 70–73, 77, 124–126, 156

welfare 1–4, 24–25, 45–46, 82–83, 117animal rights 45–46community views 21, 22–23, 25–28concepts

affective states 37–42biological functioning 29–37expression of normal or ‘natural’

behaviours 43-45effects of the stockperson 20, 21–22,

23–24, 126–128, 133–134, 154–155

exploitation of animals 22–23measurement 26–27, 28–29, 39–42standards 25, 26, 27see also fear of humans