OPERANT STUDIES ON THE BEHAVIOR OF PIGS AND ......Operant conditioning is a procedure that will reliably increase the frequency of occurrence of any behavior in the repertoire of an

OPERANT STUDIES ON THE BEHAVIOR OF PIGS AND SHEEP IN RELATION TO THE PHYSICAL ENVIRONMENT

B. A. Baldwin

Agricultural Research Council, Institute of Animal Physiology, Babrabam, Cambridge, England

Summary An outline is given of the use of operant

conditioning techniques in the study of the behavior of farm animals. Using operant methods, pigs and sheep placed in cold environments have been trained to perform simple responses in order to obtain radiant heat. Factors which influence this behavior such as level of nutri t ion, ambient temperature and the intensity of the radiant heat have been examined. The effect of warming or cooling the hypothalamus on the motivation to work for heat production has been studied. Experiments are also described in which, using operant methods, an a t tempt has been made to determine the il lumination preferences of pigs and ruminants. (Key Words: Operant Conditioning, Behavior, Thermoregulation, Illumination Preference, Sen- sory Reinforcement.)

Introduction

The environment of a farm animal may be considered to be composed of three main components which are indicated below.

1. The Physical Environment--temperature, illumination, type of floor, ventilation, et cetera.

2. The Social Environment--presence or absence of other animals, dominance hierarch- ies, group size and composit ion, et cetera.

3. The Managerial Environment--the diet, weaning systems, feeding regimen, et cetera.

Of the three main components mentioned above, the physical environment most readily lends itself to objective study by means of behavioral preference tests and measurements of the motivation to obtain a particular environmental situation. By means of operant conditioning techniques, the details of which are outlined below, it is possible to "ask the animal" what kind of physical environment it prefers. However, it must always be noted that

the environmental condit ions preferred by an animal may not be those that are economically feasible and it is also probable that the environmental preference of adults may be influ- enced by their previous environmental exper- ience as has been shown in rats (Roberts et aL, 1958). Before considering some of the experimental work concerned with the behavior of farm animals in relation to their physical environment, it is appropriate to explain some of the main features of the operant conditioning methods that are used in this work. For a more detailed exposit ion see Honig (1966).

In this paper, no a t tempt has been made to review the many observational studies on the behavior of pigs and sheep in relation to the physical environment (Hafez, 1968; Mount, 1968).

The Use of Operant Conditioning Methods in the Study of Animal Behavior. Operant conditioning is a procedure that will reliably increase the frequency of occurrence of any behavior in the repertoire of an animal. In very general terms, an animal's behavior may cause the appearance of an additional stimulus in its environment or the disappearance of some stimulus that was present in the environment.

If the appearance of a stimulus as a consequence of a behavioral act (response) produces an increased probabil i ty of that response occurring in the future, the stimulus is called a positive reinforcer and the process is termed positive reinforcement. If the removal of a stimulus as a consequence of a response results in an increased probabil i ty of that response occurring in the future, the stimulus is termed an aversive stimulus or negative reinforcer and the process is known as negative reinforcement. In everyday language, animals will learn to perform particular acts so as to obtain something they want or to remove something they dislike.

1125 JOURNAL OF ANIMAL SCIENCE, Vol. 49, No. 4, 1979

1126 BALDWIN

If we wish to study many types of behavior quantitatively we must first train our animals to perform a simple response, the occurrence of which we can conveniently measure. For example, if we place a hungry young pig in a small metal cage with a prominent panel switch mounted at one end of the cage, the pig will eventually, during its exploration of the cage, place its snout on or near the panel. When this happens the experimenter delivers a small amount of food to the animal by means of an electronically operated food dispenser. Soon the pig shows great interest in the panel switch and will usually "accidental ly" press it and thus operate the feeding device. Once this has occurred the pig usually begins to consistently press the panel in order to obtain food (figure 1). Using pigs that have been trained to press a panel to obtain food we can conventiently study their detailed pattern of food intake by fitting such a device in their pens and recording when the feeder is operated.

Provided an appropriate behavioral response is selected and the animals used are tame, the domesticated farm animals make good subjects for operant conditioning experiments. Goats, sheep and calves have been trained to press panels with their muzzle in order to obtain oats and, using a two panel system in which to obtain food they had to press the panel over which a particular odor was emitted, it has been possible to study their ability to make olfactory discriminations. It has been shown that these ruminants can distinguish between individuals of their own species on the basis of odors, particularly those from urine (Baldwin and Meese, 1977a; Baldwin, 1977).

Goats can be depleted of sodium by means of loss of saliva from a parotid fistula and animals prepared in this manner will learn to press panels in order to obtain salt solutions to drink. Operant methods have proved useful in the study of sodium appetite in goats (Baldwin, 1976) and sheep (Abraham et al., 1973).

Operant Studies on Behavioral Tbermoreg- ulation in Pigs and Sheep. Thermoregulatory behavior can conveniently be studied by means of operant techniques. In most experiments animals are placed in cold environments and perform a suitable response in order to obtain heat. In a few studies, rats exposed to hot environments have learned to press levers in order to obtain cool air or showers of water. There is an extensive literature on thermoregulatory behavior (reviewed by Cabanac, 1972;

Figure 1. The pig is about to push the panel switch in order to obtain pelleted food delivered into the bowl.

Corbit, 1970; Baldwin, 1974; Satinoff and Hendersen, 1977). Following the initial experiments of Weiss and Laties (1961) with rats, most of the work has been carried out on laboratory animals, but some experiments have been done on pigs and sheep.

The factors that influence thermoregulatory behavior in pigs have been examined in a series of experiments (Baldwin and Ingram, 1967a, b, 1968a, b; Baldwin and Lipton, 1973; Ingram, 1975). Young pigs that were exposed to cold soon learned to press panel switches with their snouts to obtain radiant heat from infrared heaters suspended above their cage (figure 2). When pigs, that had been trained to operate the heaters, were exposed at a range of ambient temperatures from -10C to +40C, it was found that the rate at which they turned on the heaters declined markedly at 25C (figure 3). It has also been shown that young pigs aged 8 to 14 weeks learned to operate infrared heaters at ambient temperatures at 20C, but not at 25C and this finding shows that a marked fall of subcutaneous temperature is not needed to ensure that learning occurs (Baldwin and Lipton, 1973).

The fact that young pigs cease to operate the heaters at ambient temperatures of 25C is of physiological interest because at about this temperature they enter their thermoneutral

BEHAVIORAL THERMOREGULATION AND ILLUMINATION PREFERENCE 1 1 2 7

REINFORCEMENT: INFRA RED HEATER ON FOR 3 SECONDS

Figure 2. The cage in which pigs were trained to work for infrared heat. The panel switch is shown at the end of the cage and each press delivered a few seconds of infrared heat. (Courtesy: Physiology and Be- havior).

zone in which the ambient temperature makes minimal metabolic demand upon the animal (Mount, 1968). Experimental evidence that pigs in cold environments operate radiant heaters so as to reduce their metabolic rate to levels similar to those observed when in their thermoneutral zone has been obtained (Baldwin and Ingram, 1967a; Ingram, 1975). There would be litt le physiological advantage in operating the heaters at ambient temperatures above 25C. The main point of behavioral interest in that operant methods enable the accurate determination of the ambient temperature at which trained pigs are no longer motivated to obtain extra radiant heat and may be assumed to be comfortable.

Young pigs that were kept on a low plane of nutr i t ion operated the radiant heaters more often than pigs that were on a high plane of nutri t ion when both groups were tested at ambient temperatures of 10C (figure 4) (Bald- win and Ingram, 1968b). These results suggest that hungry pigs feel the cold more markedly than well fed ones.

Behavioral thermoregulation has also been investigated using operant n~ethods, in a group of four young pigs aged 8 to 14 weeks that lived outside in a small hut with an adjoining yard (Baldwin and Ingram, 1968a). A 3 kW set of infrared heaters was suspended over a port ion of the yard and was turned on for 6 sec

8

z

z uJ

o z uJ

$ os _J

P

2 5 0 0 -

2000-

1500-

I000-

500" 1

0

PIG 1176 "1 KEY [ ] RESFaDNSES

�9 REINFORCEMENTS

TEMPERATURE IN C Figure 3. The response rate and re inforcements ob-

tained by a pig exposed at various environmental temperatures . As each reinforcement lasted several seconds the pig could press the panel during the period o f heat ing and thus responses could exceed reinforcements .

when a pig pressed a switch panel mounted on the wall of the pen near the heaters. The experiments were started during the winter and all pigs in the group learned to operate the heaters. The results obtained showed that the pigs operated the heaters during the daytime, but at night they preferred to huddle together in the hut even though the ambient temperatures were lower at night. This behavior of huddling at night and operating the heaters only during the dayt ime persisted even when the outside yard and the hut were continuously il luminated for over a week.

In the preceding experiments, in which the pigs spent the nights inside the hut, the possibil- i ty arose that, if the switch panel and heaters were placed in the hut, heat might be obtained during the night and daytime. This idea was tested using natural lighting and also when the hut and yard were continuously illuminated. It was found that, under natural illumination, the pigs only operated the heaters during the daytime. However, when the hut and yard were continuously il luminated some reinforcements

1128 BALDWIN

40C

:~ 30Q o

LU 20C

~ I�9 J ' f, ,/'

'd

I ~ IO t5J TRIALS

Figure 4. The amount of infrared heat obtained by young pigs on high and low planes of nutrition. The animals were tested at an ambient temperature of 10C. (Courtesy: Physiology and Behavior).

were obtained during the night and there was a tendency for the period during which most reinforcements were obtained to change so that the pigs started to operate the heaters later in the morning and continued to do so until about midnight. The above experiments illustrate the use by the young pigs of two types of thermoregulatory behavior, the inate tendency for them to huddle together and the operant behavior of turning on the radiant heaters. It is, of course, possible that there is a learned component in huddling as the piglets would rapidly sense the beneficial effects of their behavior. The physiological effects of huddling, in terms of reducing metabolic rate, have been fully investigated by Mount (1960).

In another series of experiments a group of young pigs that had learned to operate the heaters mounted in the small yard were offered access to a paddock 18 m • 8 meters. It was found that although the overall pat tern of responding was similar to that seen when they were in the small yard the number of heat reinforcements received by the pigs, which could move all over the paddock, was greatly reduced. This happened because social interac- tions or behavior such as rooting in the soil t ook precedence over operating the heaters. The experiment illustrates the fact that in a

complex environment activities such as thermoregulation do not have absolute priori ty but must compete for precedence with other types of behavior. In natural conditions animals have to seek food, avoid enemies, find mates, etc., and thermoregulatory behavior has to fit into the animals' lifestyle, its importance varying with the prevailing environmental conditions.

Ingram and Legge (1970)s tudied the thermoregulatory behavior of young pigs living in a natural environment and found that they showed no tendency to shelter and huddle until the ambient temperature fell below 5C. They noted that t h e pigs did avoid exposure to wind, but did not select areas of their environment that had the highest mean radiant temperature. These results again emphasize that thermoregulatory behavior does not take precedence over other activities until the environmental conditions begin to threaten the animals.

Recently, operant conditioning methods have been applied to the study of thermoregulatory behavior in sheep. In the initial experiments (Baldwin, 1975) a fully fleeced sheep was placed inside a cage fi t ted with a photoelectric beam so that interruption of the beam, by the sheep's muzzle, provided infrared heating for as long as the beam was interrupted. It was found that, when the sheep lived in the cage for 4 days at an ambient temperature of 5C it very seldom operated the heaters. How- ever, after shearing, the sheep soon learned to operate the heaters (figure 5). Using shorn sheep that had learned to operate the radiant heaters the effect of exposing them at various ambient temperatures has been examined and as can be seen in figure 6, at OC they had the heaters on for about 35% of each 24 hr, while at 25C they used the heaters for only a few minutes per 24 hours.

It has also been demonstrated that sheep can compensate very accurately for changes in the intensity of the radiant heat. The shorn sheep were trained to operate the infrared heaters by interrupting a photoelectric beam with their muzzles and were exposed for 48-hr periods at an ambient temperature of 10C with either 900 Watts of infrared heaters suspended above them or 1800 Watts. As~shown in figure 7, the sheep halved the duration of heating obtained when the intensity was doubled. This result illustrates the precision of the neural system controlling thermoregulatory behavior and implies that sheep have very effective peripheral tempera-

B E H A V I O R A L T H E R M O R E G U L A T I O N A N D I L L U M I N A T I O N P R E F E R E N C E 1129

z 3oo

200 -

i o o -

| o

1 +

1 2 3 4 s 6

DAYS

Figure 5. The effect of close shearing on the duration of intra-red heating obtained by sheep at an ambient temperature of + 5C. Mean results from three sheep exposed for 6 days. Vertical bars indicate -+ SE of mean. (Courtesy: J. Physiol.).

ture receptors despite the fact that most of them are usually covered by thick fleece.

Sheep can be chilled by rapidly loading the rumen with ice-cold water via a fistula. This procedure makes the sheep expend extra energy to raise the water temperature up to body temperature and the energy required is obvious-

6 0 0 -

SO0

z_.

4 0 0 _z

0 ~ 300

~ 200

I- Z I 0 0

2

O 110 2 0 25

A M B I E N T T E M P E R A T U R E C

Figure 6. The duration of infrared heating obtained by shorn sheep at ambient temperatures of 0, 10, 20 and 25C. Mean results from four sheep each exposed for 72 hr to each temperature. Vertical bars indicate -+ SE of means. (Courtesy: J. Physiol.).

- + -

5 0 0 -

4 0 0 '

z

8 _z 3oo .

o

_z ~ 2 0 0

~ 100

l -

IE 0 , 9 0 0 Watts 1800 Watts

INTENSITY OF | N F R A - RED HEATING

Figure 7. The duration of infrared heating obtained by shorn sheep exposed to two intensities of infrared heating.

ly related to the volume of water administered. Experiments have been carried out in which the effect on thermoregulatory behavior of loading the rumen with 1 or 2 liters of ice-cold water have been studied.

The experiments were done with the sheep restrained in a metal stand and operating the heaters by interrupting a photoelectric beam with their muzzle. They were allowed to operate the heaters for a 1-hr control period before the rumen was loaded with cold water. At the end of the control period the rumen was rapidly loaded using a hand pump, with 1 or 2 liters of water at 0 to 1C. The duration of infrared heating obtained in the 1-hr period following loading was recorded. The experiments were carried out at ambient temperatures of 0, 10, 20 and 30 C and in the case of the 2 liters loading also at 40C. The results obtained are shown in figures 8 and 9. At ambient temperatures of 0, 10 and 20 C the increment seen after intraruminal loading with 2 liters of water at 0 to 10C was almost exactly double that obtained with i liter loading (compare figures 8 and 9). At an ambient temperature of 30C loading the rumen with 1 liter of water did not result in the sheep increasing the duration of infrared heating, but a 2-liter load was effective. At an ambient temperature of 40C even the 2-liter load was ineffective.

1 1 3 0 BALDWIN

8OO-

e~ O

700-

z_ :E 6 0 0 -

- r

~ 5 0 0 - Z

N 4 0 0 -

0

~ 3 0 0 -

w

,~ 2oo-

g lOO-

O o IHr 2 H r O l H r 2Hr O IHr 2Hr O IHr 2Hr

I O C I § I § I + 3 0 C I

Figure 8. The effect of loading the rumen with 1 liter of water at 0 to 1C at ambient temperatures of O, 10, 20 and 30C on the duration of heating obtained by shorn sheep. The unhatched bars indicate the duration of heating obtained in the 1-hr control period preceding loading and the hatched bars indicate the duration of heating obtained in the 1-hr period following loading. Both 1-hr periods are divided into half- hour increments. Mean results from three sheep each tested three times at each temperature. Vertical bars indicate SE of mean (Courtesy: J. Physiol.).

a o

a .

< ua

Q ua _z

e~ o

z

o IF~ 2 ~ o IHr 2Hr O IHr 2Hr O IHr 2Hr O 1Ht 2Hi

I O C ! e l O C I ~ ' 20C I ~'30~ I e 4 0 C I

Figure 9. The effect of loading the rumen with 2 liters of water at 0 to 1C at ambient temperatures of 0, 10, 20, 30 and 40C on the duration of infrared heating obtained by shorn sheep. The unhatched bars indicate the duration of heating obtained in the 1-hr control period preceding loading and the hatched bars indicate the duration of heating in the 1-hr period following loading. Mean results from three sheep each tested three times at each temperature. Vertical bars indicate SE of mean. (Courtesy: J. Physiol).

The Neural Control o f Bebavioral Thermo- regulation. In studies on the neural control of thermoregulatory behavior the usual experimental method has been to apply localized thermal stimulation (warming or cooling) to those parts of the brain that are known to be concerned in temperature regulation and see whether this procedure altered the rate at which animals worked to obtain heat. Follow- ing the initial experiments of Satinoff (1964) with rats, experiments of this type have been carried out on a large number of species includ- ing monkeys (Adair et al., 1970), baboons (Gale et al., 1970), pigs (Baldwin and Ingram, 1967b), sheep (Baldwin and Yates, 1977), lizards (Hammet et al., 1967) and fish (Stromme et al., 1971). In addition to thermal stimulation

of localized brain regions the effect of brain lesions (Carlisle, 1969; Satinoff and Rut- stein, 1970) or chemical stimulation of the brain (Beckman and Carlisle, 1969; Avery and Penn, 1973; Crawshaw, 1973) on operant behavioral thermoregulation have also been studied. Much of the recent literature on the neural control of thermoregulation is summar- ized by Satinoff and Hendersen (1977), and in the present paper the emphasis is on studies in which sheep or pigs were used.

The effect of heating and cooling the preoptic region of the hypothalamus on the rate at which trained pigs operated infrared heaters has been investigated (Baldwin and Ingram, 1967b). The pigs had metal "thermodes" implanted in the preoptic region of the hypothalamus so that

B E H A V I O R A L T H E R M O R E G U L A T I O N A N D I L L U M I N A T I O N P R E F E R E N C E 1 13 1

this region could be cooled or warmed during operant thermoregulatory behavior. The effects of cooling the preoptic region were examined at various ambient temperatures between OC and 35C. When the pigs were at low ambient temperatures the rate at which the heaters were turned on increased during central cooling, while in a thermoneutral environment at 25C the pigs often only operated the heaters during central cooling. At the higher ambient temperatures of 30 and 35C cooling the preoptic region was an ineffective stimulus in about one-half the trials. Warming this region de- creased the rate at which the heaters were tumed on at low ambient temperatures, but the effects were not as marked as the increases seen during central cooling. The effects on operant thermoregulatory behavior of cooling the anterior spinal cord in the pig have been studied by Carlisle and Ingrain (1973). They found that spinal cooling increased the response rate, but the effects were not as marked as those seen during cooling of the preoptic region.

The neuraJ control of thermoregulatory behavior in sheep has been examined by Bald- win and Yates (1977). Shorn sheep were trained to press panels with their muzzles in order to obtain infrared heat. At ambient temperatures of 5, 15, 25 and 35 C cooling the preoptic region by means of an implanted thermode for periods of 20 min resulted in a marked increase in the rate at which the heaters were used. The relationship for trials conducted at 15 and 25C is shown in figure 10. At ambient temperatures of 25 and 35C the absolute size of the response to cooling was less than that seen at 15C and this reduction indicates that a warm peripheral input can reduce the behavioral effects of hypothalamic cooling.

The effect of warming the preoptic region by means of an implanted thermode has also been studied and at ambient temperatures of 5 and 15C warming the hypothalamus caused a considerable reduction in the response rate (figure 11).

An inherent problem, when using implanted metal thermodes to alter hypothalamic temperature, is that the neural tissue adjacent to the thermode is subjected to abnormal temperatures when the thermode is heated or cooled. It is therefore possible that some of the effects on behavioral thermoregulation are due to unphysiological levels of thermal st imulation applied to temperature sensitive neurons. In

A

22-

20:

16-

11-

12-

10-

o ' 1~' ~o' 3'o' ~o' 5'o'

2 B z

~ 22-

~ 20-

~ 18

z 16:

14: 12

8

TA15 C

60

MINUTES

TA25 C

4 �84

2 , . . . . . . . . . . . . . . . . i,i,i,ii 10 20 30 40 50 60 70 80 90 100 110 120 130 140

MINUTES

Figure 10. The effect of cooling the hypothalamus at ambient temperatures of 15C (A) and 25C (B). The black bars indicate thermode cooling to 15C. Mean results + SE of mean from four sheep each tested three times at each temperature. (i.e., each sheep was cooled nine times). (Courtesy: J. Physiol.).

order to determine whether lowering brain temperature by 1C could produce changes in thermoregulatory behavior similar to those obtained using thermodes, a method has been devised for the intracarotid injection of cold saline while the sheep pressed a panel to obtain infrared heat.

At ambient temperatures of 15, 25 and 35C the cephalic region was cooled by intracarotid injections of cold normal saline for periods of 15 minutes. Hypothalamic temperature fell by about 1C and increases were seen in the rate at which the heaters were used which were similar to those observed when the thermodes were cooled. To elicit large increases in the rate at which the radiant heaters were used, it was not

1132 BALDWIN

22

20

18

16

14

12

10

z

z 4 -

~_ 2- z

A TA5 C

10 20 30 40 50 60 70 80 90 100 110 120 130 140

MINUTES

B TA15 C

2 2 .

~ 18-

16-

14-

12

10-

8-

6 -

4 -

2 -

10 211 3 0 4 0 5 0 6 0 70 8 0 9 0 100 110 12(I 130 140

MINUTES

Figure 11. The effect of warming the hypothalamus at ambient temperatures of 5C (A) and 15C (B). The black bars indicate thermode warming to 43C.

necessary to lower hypotha lamic t empera tu re outs ide the normal range. Intracarot id injec- t ions o f saline at body tempera tu re had no ef fec t on response rate (figure 12).

It is known that thermal s t imulat ion of the medul lary region can inf luence behavioral thermoregula t ion in rats (Lipton, 1971, 1973) but exper iments of this type have no t been carried out on o ther species.

The study of the neural control of behavioral thermoregula t ion provides an excel lent mode l for the investigation o f the physiological basis o f mot iva t ional processes as the relevant biological s t imulus can be applied to the central and peripheral receptor systems and the effect on behavior observed.

Studies on Sensory Reinforcement and Illumination Preference in Pigs and Ruminants. Rats and mice placed in darkness learn to

Z l a .

16.

z 14,

12"

u o

A T A 1 5 C

.-|~ .... ! ~- 38.o

4 o . o o

3q . s

39.0

~ 3 5 0 , , , , , , , , , , , , , , , , , . . . . , , , io ~o 3o 4o so 6o 7o so 9o ioo 11o 12o

~- MINUTES

B m m TA15 C

z

"| ~ - - 6 -

~ z "

~ .

Z .

v c

3 9 o

31t5

38 0

a o o v 3 ~ 5

a * 0

0 m 20 30 40 so ~ 70 ~ 90 i ~ 110 120

MINUTES

Figure 12. A. Effect of intracarotid injection of cold saline (about 2 to 3 C) at an ambient temperature of 15 C. Black bars indicate injections. Mean results -+ SE of mean from three sheep each tested on six occasions (i.e., each sheep received 12 injections). B. Effect of intracarotid injection of saline at body temperature (about 39 to 40 C) at an ambient temperature of 15 C. Black bars indicate injections. Mean results +- SE of mean from three sheep each tested six times (ie., each sheep received 12 injections). (Courtesy: J. Physiol.).

per form operant responses in order to obtain i l luminat ion and the l i terature on this behavior has been reviewed by Kish (1966). The te rm sensory re in forcement is usually applied to the demons t ra t ion of the reinforcing propert ies of st imuli , such as light, that are unrelated to pr imary needs such as food , water or warmth .

BEHAVIORAL THERMOREGULAT1ON AND ILLUMINATION PREFERENCE 1 133

In intensive husbandry systems the level of il lumination provided for pigs and calves is of interest to those concerned with animal welfare. Recently, operant methods have been used to determine illumination preferences in pigs (Baldwin and Meese, 1977b) and also sheep and calves (Baldwin and Start, 1978).

In the experiments on illumination preferences in pigs the animals were individually housed in a metal pen situated inside a sound proof room. The pigs were placed in the pen in darkness. In one wall of the pen were two adjacent slits through which the pig could place its snouth. Both slits contained infrared beams, and it was arranged such that when one beam was interrupted the room lights came on and remained on until the other beam was interrupted (the equivalent of an on/off switch). The pigs remained in the pen for a 2-day period and the position of the "on" and "off" beams were alternated each day in order to avoid position preferences. The total duration of light obtained each hour was rated to the nearest second. As can be seen in figure 13, the pigs tended to keep the light on for a mean of 72% of the time and there was no obvious tendency for the pigs to leave the lights off for prolonged periods. The results indicate a definite preference for light over darkness.

In the experiments on the reinforcing properties of light, the pigs were placed in the darkness in the pen containing two slits, and it was arranged that when one beam was interrupted the room lights came on for as long as the beam was interrupted. When the pig put its snout through the other beam, it had no effect on the lights but the duration of interruption of the "control" beam was noted. It was found that the pigs only had the lights on for about .5% of the possible time. However, they interrupted the "stimulus" beam for significantly longer periods than the "control" beam (mean values per 24 hr of 133 -+ 26 sec for the stimulus beam compared with 57 + 17 sec for the control beam, P<.001). Results indicate that the light was only a weak reinforcer to pigs kept in darkness. It appears that, although pigs prefer light to darkness, they are not highly motivated to perform operant responses to obtain light.

Similar techniques have been used to study illumination preferences in sheep and calves (Baldwin and Start, 1978). The sheep lived in the test pen for a 5-day test period and were provided, using the infrared beam system, with

4 0 0 0 - -

3000-

o

O. 2000-

m o o -

o~

0

I 1 I 21 3 I 4 I 5 I s I 7 I 8 I 9110111112113114 Its1161171181191z01211221231241

TIME (HOUR I

Figure 13. Illustrates the duration of lighting obtained in each hour by pigs in the light-darkness preference test. Mean results from six pigs each exposed for 48 hours. At 10 hr each day they were fed and cleaned and left in darkness. Bars indicate -+ SEM. (Courtesy: Anim. Behav.).

an "on/off" switch. The mean duration of lighting obtained per 24 hr was 1,175 + 100 min (82% of 24 hr). No obvious circadian pattern of lighting was seen.

The above examples illustrate the value of operant methods in the evaluation of illumination preferences in farm animals. They can provide an objective means of "asking the animal" what its environmental preferences are, and the information can be of value in the design of husbandry systems.

Literature Cited

Abraham, S., R. Baker, D. A. Denton, F. Kraintz, L. Kraintz and L. Purser. 1973. Components in the regulation of salt balance: salt appetite studied by operant behaviour. Australian J. Exp. Biol. Med. Sci. 51:65.

Adair, E. R., J. V. Casby and J. A. J. Stolwijk 1970. Behavioural temperature regulation in the squirrel monkey. Changes induced by shifts in hypothalamic temperature. J. comp. Physiol. Psychol. 72:17.

Avery, D. D. and P. E. Penn. 1973. Effects of intrahypothalamic injections of adrenergic and cholinergic substances on behavioural thermoregulation and associated skin temperature levels in rats. Pharmacol. Biochem. and Behav. 1:159.

Baldwin, B. A. 1974. Behavioural thermoregulation. In J. L. Monteith and L. E. Mount (Ed.) Heat Loss from Animals and Man. Assessment and Control. London, Butterworths.

Baldwin, B. A. 1975. The effects of intraruminal loading with cold water on thermoregulatory behaviour in sheep. J. Physiol. 249:139.

1134 BALDWIN

Baldwin, B. A. 1976. Effects of intracarotid and intraruminal injections of NaC1 or NaHCO 3 on sodium appetite in goats. Physiol. Behav. 16:59.

Baldwin, B. A. 1977. Ability of goats and calves to distinguish between conspecific urine samples using olfaction. App1. Anim. Ethol. 3:145.

Baldwin, B. A. and D. L. Ingram. 1967a. Behavioural thermoregulation in pigs. Physiol. Behav. 2:15.

Baldwin, B. A. and D. L. Ingrain. 1967b. The effect of heating and cooling the hypothalamus on behavioural thermoregulation in the pig. J. Phy- siol. 191:375.

Baldwin, B. A. and D. L. Ingram. 1968a. Factors influencing behavioural thermoregulation in pigs. Physiol. Behav. 3:409.

Baldwin, B. A. and D. L. Ingram. 1968b. The effects of food intake and acclimatization to temperature on behavioural thermoregulation in pigs and mice. Physiol. Behav. 3:395.

Baldwin, B. A. and J. M. Lipton. 1973. Central and peripheral temperatures and EEG changes during behavioural thermoregulation in pigs. Acta Neurobiol. Exp. 33:433.

Baldwin, B. A. and G. B. Meese. 1977a. The ability of sheep to distinguish between conspecifics by means of olfaction. Physiol. Behav. 18:803.

Baldwin, B. A. and G. B. Meese. 1977b. Sensory reinforcement and illumination preference in the domesticated pig. Anita. Behav. 25:497.

Baldwin, B. A. and J. O. Yates. 1977. The effect of hypothalamic temperature variation and intracarotid cooling on behavioural thermoregulation in sheep. J. Physiol. 265:705.

Baldwin, B. A. and I. B. Start. 1978. Methods for the study of illumination preferences in sheep and calves. J. Physiol. 284:13P.

Beckman, A. L. and H. Carlisle. 1969. Effect of intrahypothalamic infusion of acetylcholine on behavioural and physiological thermoregulation in the rat. Nature 221:561.

Cabanac, M. 1972. Thermoregulatory behaviour. P 19-36. In J. Bligh and R. Moore (Ed.) Essays on temperature regulation. North Holland Pub. Co., Amsterdam.

Carlisle, H. J. 1969. The effects of preoptic and anterior hypothalamic lesions on behavioural thermoregttlation in the cold. J. Comp. Physiol. Psychol. 69:391.

Carlisle, H. J. and D. L. Ingram. 1973. The effects of heating and cooling the spinal cord and hypothalamus on thermoregulatory behaviour in the pig. J. Physiol. 231:353.

Corbit, J. D. 1970. Behavioural regulation of body temperature. P 777-800. In J. D. Hardy, A. P. Gagge and J. A. J. Stolwijk (Ed.) Physiological and Behavioural temperature regulation. Charles C Thomas, Springfield IL.

Crawshaw, L. I. 1973. Effect of intracranial acetylcholine injections on thermoregulatory responses in the rat. J. Comp. Physiol. Psychol. 83:32.

Gale, C. C., M. Mathews and J. Young. 1970. Behav- ioural thermoregulatory responses to hypothalamic cooling and warming in baboons. Physiol. Behav. 5:1.

Hafez, E. S. E. 1968. Adaptation of Domestic Animals. P 202-214. Lea & Febiger, Philadelphia, PA.

Hammel, H. T., F. T. Caldwell and R. M. Abrams. 1967. Regulation of body temperature in the blue tongued lizard. Science 156:1260.

Honig, W. K. 1966. Operant behaviour. Areas of Research and Application. Appleton-Century- Crofts, New York.

Ingram, D. L. 1975. The efficiency of operant thermo- -regulatory behaviour in pigs as measured by the rate of oxygen consumption. Pfl. Arch. Ges. Physiol. 353:139.

Ingram, D. L. and K. F. Legge 1970. The thermoregulatory behaviour of young pigs in a natural environment. Physiol. Behav. 5:981.

Kish, G. B. 1966. Studies of sensory reinforcement. P 109-159. In W. K. Honig (Ed.). Operant behaviour, Appleton-Century-Crofts, New York.

Lipton, J. M. 1971. Behavioural temperature regulation in the rat. Effects of thermal stimulation of the medulla. J. Physiol. (Paris) 63: 325.

Lipton, J. M. 1973. Thermosensitivity of medulla oblongata in control of body temperature. Amer. J. Physiol. 224:890.

Mount, L. E. 1960. The influence of huddling and body size on the metabolic rate of the young pig. J. Agr. Sci. Camb. 55:101.

Mount, L. E. 1968. The climatic physiology of the pig. P 271. Edward Arnold, London.

Roberts, C. L., M. H. Marx and G. Collier. 1958. Light onset and offset as reinforcers for the albino rat. J. Comp. Physiol. Psychol. 51:575.

Satinoff, E. 1964. Behavioural thermoregulation in response to local cooling of the rat brain. Amer. J. Physiol. 206, 1389.

Satinoff, E. and J. Rutstein. 1970. Behavioural thermoregulation in rats with anterior hypothalamic lesions. J. Comp. Physiol. Psychol. 71:77.

Satinoff, E. and R. Hendersen. 1977. Thermoregula- tory behaviour. P 153-173. In W. K. Honig and J. E. R. Stadden (Ed.). Handbook of Operant Behaviour. Prentice Hall, Inc., Englewood Cliffs, NJ.

Str~mme, S. B., K. Myhre and H. T. Hammel. 1971. Forebrain temperature activates behavioural thermoregulatory responses in arctic sculpins. J. Physiol. (Paris) 63:433.

Weiss, B. and V. G. Laties. 1961. Behavioural thermoregulation. Science 133 : 1338.

Documents

OPERANT STUDIES ON THE BEHAVIOR OF PIGS AND ......Operant conditioning is a procedure that will reliably increase the frequency of occurrence of any behavior in the repertoire of an