Toward a Social Psychology of Loneliness (pág. 31...)

7/30/2019 Toward a Social Psychology of Loneliness (pg. 31...)

1/13

In $>\ ~ u c k R. G i h o u r ( E d s . ), Perso n a lRela t ions hips in Disorder. London: Academic

Press , 1981.

CHAPTER 2

Toward a Social Psychology ofLoneliness

Daniel Perlman and Letitia Anne Peplau

Loneliness is a common experience - probably few people avoid beinglonely at some time in their life. It is also a distressing experience as manyindividual accounts bear out. For example, in interviews with journalistSuzanne Gordo n (1976) one retired surgeon commented on the lonelinessofold age, that,"You are alone, people have died and you look in the mirror

and you look awful . . ."; while a middle-aged woman describing herunhappy marriage noted that "There was no one to talk to . . . To me,loneliness and depression were absolutely synonymous". Yet, despite the

pervasiveness and importance of the phenomenon, it is only recently that

social scientists have attempted a suitably empirical, theoretically-derivedstudy of loneliness (e.g. see Hartog et at. , 1980; Peplau and Perlman, inpress).

The intention of the present chapter is to review the literature, drawing

mainly on current empirical research, and to provide a conceptual perspec-

tive on loneliness. We start, therefore, by considering a more formaldefinition of the concept; in our view loneliness is the unpleasant experience

that occurs when a person's network of social relations is deficient in someimportant way, either quantitatively or qualitatively; and although

31

32 DANIEL PERLMAN AND

loneliness may at times reach pathological propo

concerned with"normal" ranges of loneliness amonthis definition there are three general points to be nshared by other definitions that have been offered (sein press): first, loneliness results from deficiencies

relations; second, loneliness is a subjective phenomsarily synonymous with objective isolation, so tha

without being lonely); third, loneliness is unpleasant

Conceptually, we draw upon an attributional appr1979) and view loneliness as a discrepancy betwachieved levels of social relations. One advantage of

draws attention to the levels of social contact that peo

important set of conditions producing loneliness,

social scientists have ignored this aspect of the prob

on the low levels of social contact that people ac

crepancy perspectivethus gives a more comprehensithat contribute to loneliness and helps us to understmight otherwise be anomalous.

A second major advantage of the discrepancy-atthat it takes account of cognitive factors mediating

deficiency and emotional response. Cognitive proce

butions and perceived control are seen as affecting

situation subjectively. Most traditional views of lonPerlman, in press), however, emphasize our humanthat loneliness is seen as the inevitable direct consatisfy these needs and any intervening cognitive

entirely ignored.In developing the approach indicated above the

into six parts: (1) the forms and measurement of lo

tations of loneliness, (3) antecedents of loneliness,that modulate the lonely experience, ( 5 ) how peopl

others , and (6) coping with loneliness.

FormsandMeasurementof Loneliness

Forms

Various typologies have been used to distinguish d

liness (see de Jong-Gierveld and Raadschelders, in plying factors have each been used in articulating type


2/13

SOCIAL PSYCHOLOGY OF LONELINESS 33The first factor uked in classifying types of loneliness can be seen in the

writing of Moustakas (1961). He distinguished between loneliness anxietyand existential loneliness. According to him, loneliness anxiety is aversiveand results from "a basic alienation between man and man", whereas

existential loneliness is an inevitable part of the human experience, involvingperiods of self-confrontation and providing an avenue for self-growth.While it can be painful, it can also lead to "triumphant creation". Thus,

Moustakas, like others , sees a positive-negative dimension running throughloneliness experiences. Loneliness anxiety is the negative form; existentialloneliness is the positive form. In this chapter, we will primarily be con-cerned with what Moustakas callsloneliness anxiety.

Time has been used as a second basis of classification schemes. Loneliness

can be seen as a temporary "state" perhaps linked to specific events such asmoving to a new community; or, it can be seen as a more chronic"trait". Theindividual can have a short-term loneliness "experience", or s/he can be a"lonely person".

A th ird way of categorizing forms of loneliness has been on the basis of the

social deficiency involved. Weiss (1973) distinguished emotional loneliness(based on the absence of a personal, intimate relationship) from social

loneliness (based on the lack of social "connectedness"or sense of com-munity). He believes emotional loneliness is a more acutely painful form of

isolation; social loneliness is experienced as a mixture of feeling rejected orunacceptable, together with a sense of boredom.

In all these forms, loneliness is assumed to be an emotionally intenseexperience. Empirical work by de Jong-Giemeld and Raadschelders (inpress) has identified yet one other type of loneliness. People in this group are

passively resigned to their fate. Although they may lack both an intimatepartner and friendshipsand theymaysee no end to their condition, they accept

their social deprivations as unavoidable and are apathetic in their response.

Measuring loneliness

Researchers have used single items, uni-dimensional scales, and multi-dimensional approaches to measure loneliness (Russell, in press). In all

cases, paper and pencil techniques (or verbal questioning) have been used,

probably the most thoroughly developed and widely accepted of which is the

University of California, Los Angeles (UCLA) Scale. This consists oftwenty statements such as "I lack companionship"and "I am an outgoing

person" and respondents taking the scale are instructed: "Indicate howoften (never, rarely , sometimes or often) each of the following statements

describes you

7

'. Scoring is done in an uni-dimensional manner.

34 DANIEL PERLMAN ANDThe UCLA scale performs well on traditional psyc

a coefficient alpha of0-94 and reasonably high tesearlier form of the UCLA Scale correlated 0-74 anloneliness measures (Bradley's and Ellison and Palrisk" groups of individuals (divorced adults, priso

seeking help with their social skills) have been adminexpected, scored high on loneliness. The scale also h

that it correlates with activities (e.g. time alone persad) that theorists have linked with loneliness.

Finally, Russell and his colleagues have demonstrdiscriminant validity. It measures loneliness, per s

such as depression, anxiety, or self-esteem. This measuring loneliness but also for providing confiden

research findings in this area. If loneliness were inwith another concept such as depression, one woulddoubts that the presumed effects of loneliness were

the other factor. In real life it is probable that lotogether with depression, anxiety and/or low self-e

the UCLA Scale in combination with careful resea

the identification of lonelinessper se and its unique c

Manifestations of Loneliness

Several manifestations of loneliness can be ident

section, manifestations of loneliness are dividedaffective, motivational, behavioural and social proloneliness, of which the emotional o r affective mani

most thoroughly studied.

Affective manifestations

Virtually by definition, loneliness is an unpleasa

Reichmann (1959) described it as "painful and

clinicians have commented on the frequent associ

depression. Further writers have associated lonelindissatisfaction (Rubenstein et al., 1979), anxieboredom (Weiss, 1973) and interpersonal hostility (

Research provides empirical support for manemotional correlates: for example, Sermat (1980) a

reported data linking loneliness with hostility. In a


3/13

SOCIAL PSYCHOLOGY OF LONELINESS 35Russell et at . , (1978) found lonely students were apt to feel "angry", "self-enclosed","empty", and "awkward". These students also described them-selves as tense, restless and anxious. Similar results were obtained among a

sample of senio r citizens studied by Perlman, Gerson and Spinner (1978).Another consistent finding has been that lonely individuals have a

basically negative outlook: for instance, lonely respondents report being less

happy, less satisfied, and more pessimistic (Russell etal. , 1978; Perlman etal . , 1978). When ask ed to list and then rate ten activities they did over theproceeding weekend, lonely respondents were less satisfied with how theyspent their time (Perlman eral., 1979).

Motivational and cognitive manifestations

Two seemingly contr adictory viewpoints have been expressed concerningthe motivational aspects of loneliness. On the one hand, some authors

consider loneliness arousing: for instance, Sullivan (1953) believed lone-liness was a "driving" force. He observed that loneliness motivates

individuals to in itiate social interaction despite th e anxiety such interactionshold for lonely people. On the other hand, some authors believe that

loneliness decreases motivation: for instance, Fromm-Reichmann (1959)contended tha t tru e loneliness creates a sense of"paralyzing hopelessnessand unalterable futility". Similarly, Weiss (1973) claimed that for lonelypeople, tasks lose thei r meaning.

In on e unpublished survey by Perlman, the answers of lonely respondentsindicated apathy : fo r instance, lonely individuals endorsed such items as"A ttimes1feel worn o ut for n o special reason

", and

"My strength often seems to

drain away from me", but rejected the statement "I have a lot of energy". In

another study (Loucks, 1974), lonely students were found lacking in"vigor". Naturally, the despondency apparent in this evidence contrasts

with the hyperactivity which can be engendered by anxiety.

Several factors may be helpful in resolving the apparently paradoxicalmotivational properties of loneliness. First, loneliness may arouse motiva-

tion for interpersonal contact but diminish motivation for other tasks.Secondly, loneliness may be arousing yet interfere with the effective

channeling of one's energies to complete tasks. Thirdly, loneliness may have

different motivational properties over time. Perhaps having perceivedcontrol o ver one' s loneliness motivates people to seek ways of alleviating

their exper ience. La st, but equally plausible, loneliness may influence thefluctuation in one's motivational state more than its "average" level. In

oth er word s, lonely individuals may alternate between periods of high and

low motivational arousal.

36 DANIEL PERLMAN AND Some of the motivational manifestations of

cognitive processes. Perhaps the most salient of theinterpersonal relationships. Weiss (1973, p. 21) commfollows:

"The individual isforever appraising others for their potneeded relationships, a nd forever appraising situations ifor making the needed relationships available . . . (Loversensitivity to minimal cues and a tendency to misintehostile or affectionate intent of others."

Todate, very little systematic evidence has been presclaims. W hat has , however, b een established in severaal. , in press) is that lonely respondents are high in self-they dwell on their actions, as well as the impressiomaking on others.

Besides this vigilance, clinicians have observed toften have difficulty concentrating. The Manitoba

study) suppo rt this insight: lonely respondents were m

"easily distracted fr om a task", and, indeed, under

lonely subjects in a lab experiment made more erropaired associates than did non-lonely subjects (Perlm

were n o differences between the two groups in a non-d

Behavioural manifestations

In thinkin g about th e behavioural manifestations of ldifficult t o distinguish behaviours that accompany that lead t o loneliness in the first place, and behaviour

with loneliness. In this discussion, we consider soccontext of antec edents of loneliness, and we conside

such as attempting to meet new people, in asectionon cThr ee possible behavioural manifestations of lon

tion. First, to the extent that loneliness creates anxietindividuals may exhibit some of the characteris

frequentl y accompany these states. Second, evidencis correlated with a lack of assertiveness (Jones etal. ,of causality her e, is of course, open t o debate: whilepredispose people to loneliness, lonely people ma

mobilizing assertive behaviours. Th ird, it has been su

Reichmann, 1959) that lonely people have difficuloneliness with others.


4/13

SOCIAL PSYCHOLOGY OF LONELINESS

Finally, while some research has been done on manifestations of

loneliness, several crucial questions remain unanswered. For instance, dothese manifestations inevitably accompany loneliness? Do the variousmanifestations occur together in one or more cohesive clusters of

symptoms? If there is more than one cluster, what are these patterns or typesof loneliness? While one tempting research strategy is to identify lonelypeople via their symptoms, such a technique appears premature. Possibly,

such a technique will never be practical.

Social and medical problems

Popular writers have associated loneliness with a variety of social problemssuch as suicide, alcoholism and even illness: for instance, it has been notedthat the death rate for surviving marital partners is atypically high in the

period following their spouse's death; and some observers regard this as a

consequence of loneliness. In his book, The Broken Heart: The MedicalConsequences of Loneliness, James Lynch (1977) argued that loneliness alsomakes peop le susceptible to serious illness and promotes overuse of medicalservices. Lynch provided provocative evidence in behalf of his thesis, but itwas mostly based on people who were socially isolated and, from our

perspective, these individuals need not necessarily be lonely. Rubenstein

and Shaver (1980) report a strong relationship between loneliness and achecklist of psychosomatic symptoms such as headaches, poor appetite andfeeling tired. However, this checklist combines medical symptoms withcognitive problems (e.g., "worrying7') and feelings of self worth.

Further evidence for the link between loneliness and social problems has

been reported by Brennan and Auslander (1979). Their study was based onsecondary analyses o f several large scale surveys of American adolescents,

and they found that loneliness was associated with poor grades, expulsionfrom school, running away from home, and engaging in delinquent acts suchas theft, gambling and vandalism.

Antecedents of Loneliness

The possible antecedents of loneliness are numerous and it is useful to

distinguish events that precipitate the onset of loneliness from factors thatpredispose individuals to become lonely or to persist in being lonely overtime. Based on our definition of loneliness, precipitating events may be

broadly categorized into changes in a person's achieved social relations and

changes in a person's desired or expected social relations. Predisposing

38 DANIEL PERUVIAN AND

factors include the usual quantity and quality of onecharacteristics of the individual (e.g. personality, phmore general characteristics of a given situation orvariables are typically enduring aspects of the person

From a discrepancy viewpoint, most of these predconceptualized as variables underlying the amount ofindividual typically achieves and/or desires. Predispoand limit how people react t o life changes that migh

desired o r achieved levels of social contact. Thusfactors as putting people at risk of being lonely, bunecessarily the immediate cause of loneliness. Howe

the way one conceptualizes the causal chain leading t

expect predisposing variables to be statistically assoc

Changes in achieved social relations

Loneliness is frequently precipitated by changes in a pships that lead to a sub-optimal level of achieved sochanges may affect a single relationship, or may network of social relations.

Termination. The ending of a close emotional relcause of loneliness so that, fo r example, widowhoowith loneliness by several researchers (e.g. Weissreports that 48%of a random sample of urban wido

the major problem in widowhood, while an addit

loneliness in conjunction with other problems. Divcommon phenomenon which is also associated with

1973; Gordon, 1976); and at least one study (Hill etbreakup of dating relationships, too, is accompani

liness and depression.

Physical separation. In a mobile society, separ

friends is a common occurrence. Separation reduces

action, ma kes the satisfactions provided by a relationmay raise fears that the relationship will be weakeevents as moving to a new community, going away

camp or to university, or spending extended period

hospitals or prisons all affect social relationships, iments of work o ften impinge on social relations outsof business trips, extended hours spent working over

moving as part of career advancement. Evidence th


5/13

SOCIAL PSYCHOLOGY OF LONELINESS 39puts people at risk for loneliness is readily available: for example Weiss

(1973) has noted the difficulties experienced by wives forced to move bytheir husband's work. However, Rubenstein et al. (1979) maintain that theloneliness passes quickly for most people who move to a new community.

Status change. An individual's position within a group or organization hasconsiderable impact on interaction with others both inside and outside the

group. A s a result, changes in status may lead to loneliness. For example,promotion in a business may weaken ties with former peers, and create

loneliness until new peer relations are established. Persons who complete a

term as president o r chair of a group may also find that their contacts withothers are reduced. Similarly, role loss through retirement or unemploy -

ment typically disrupts social ties with former co -workers and so mayprecipitate loneliness (see Rubenstein e ta l . ,1979). Bart (1972) documentedthe distress felt by mothers when their grown children leave home, and

indicated that women who had invested the most in the maternal rolesuffered t he most from having an"empty nest". The acquisition of new rolescan also disrupt established social networks. For young adults, both

marriage and parenthood may lead to major and often unanticipated

changes in contact with friends and relatives (see Dickens and Perlman,1981).

Changes in desired social relations

Loneliness may be precipitated when an increase in a person's desired social

relations is not accompanied by an increase in achieved social relations.

Developmen tal changes. Age-related changes in a person's capacities anddesire for social relations may precipitate loneliness. Sullivan (1953) positeda developmental sequence in which children of different ages have different

needs and social skills. In his view, loneliness first becomes possible during

the pre-adolescent era, in which a "need for intimacy" is added to earlier

needs for tenderness, for peers and for acceptance. A rather different

developmental approach is provided by cognitive psychologists whoemphasize the growing child's changing intellectual capacities, such as role-taking ability (see Dickens and Perlman, 1981).

Developmental changes in desired social relations undoubtedly occur afteradolescence as well. For example, Gail Sheehy (1976, p. 415) suggested that

for many professionally successful people, "midlife may be a time to relax .. .and put more . . . into cultivating friendships, being a companion . . . ,beingmore activein the community". Other experiences, such as psychotherapy orconsciousness-raising groups, may also encourage individuals to re-assess the

importance and the quality of their social relations.

40 DANIEL PERLMAN AN

Situational changes. A person's desire to be withInstead, it fluctuates frequently depending on the tathe person's mood and the like. Middlebrook (197

students preferred being alone when tired or emba

companions when happy. Schachter's (1959) classthat situations of s tress or uncertainty can influence

Even holidays and seasonal changes can be importan

Changes in expectations. A person's desired levtempered, to some extent, by expectations about thare possible or likely in a given situation. In some

about future social contact help to prevent or m

example, a woman entering hospital for surgery mreduced social contacts, and so moderate her desirethat situation. In other instances, however, expecta

likelihood of loneliness. A young boy going awaypriately expect to make friends quickly, and so rairelationships to unrealistically high levels. By affec

social contact, expectations may influence the exten

experiences.

The quantity and quality of social contacts

Quantity. Perhaps the most obvious determinant of a person's social relationships. Changes in socia

been treated as a precipitating factor in loneliness. Hlevels of contact, per se, as another causal inNaturally, in cross-sectional surveys, reports of ocan reflect recent changes in one's situation: butbelieve such reports reflect on-going levels of one'

regard the quantitative aspects of one's social relatiofactor in loneliness.

There are several indications that lonely people ha

than do other people (see Jones, in press). For in

have been found to date less, and report fewer socia

more time alone; whilst lonely senior citizens havwith their friends (Perlrnanet al.,1978).

TWOinterest ing anomalies in the overall pattenoting. First, in some surveys where global indicesand non-lonely respondents have reported a similar

This may be because lonely respondents have aacquaintances but aren't actually very close to these


6/13

SOCIAL PSYCHOLOGY OF LONELINESS41

a study where college students recorded their social interactions in a diaryfor two days, loneliness was not related to the total number of interactionsthe students had. The lonely diary-keepers did, however, report moreinteractions with strangers and casual acquaintances and fewer interactionswith family and fr iends. Thus, even if lonely people have a number of brief

superficialcontacts, the overall pattern of data suggests their social contactsare defic ient as one would suspect.

The quality of relationships. Loneliness is affected not only by theexistence of social relationships and the frequency of social interaction, but

also by the quality of relationships and the needs that they meet. Forexample, among senior citizens, marital dissatisfaction was associated with

greater loneliness (Perlmanel

al., 1978). Similarly, in Cutrona's (in press)study of UCLA students, dissatisfaction with one's friendships, dating life,and family relationships were all significant predictors of loneliness. Sennat(1980) suggested that loneliness is fostered by poor communication.

Our contact must also satisfy our needs. Weiss (1973) has delineated six6 provisions" supplied by social relationships, which include feelings of

personal attachment (as in intimate relations), social integration, the oppor-tunity to receive nurturance, re-assurance of one's worth, and guidance. InWess' view, no one relationship is apt to satisfy all these needs and, instead,

different kinds of relationships are apt to satisfy different needs. In the

aforementioned UCLA study (Cutrona, in press), students rated how welltheir current relationships supplied them with each of Weiss' six provisions.As predicted, students whose needs were well met tended to be less lonely:

in particu lar, having a set of relationships that provided social integration, asense of worth, and guidance helped students avoid being lonely.

Thus. both the quantity and quality of social contacts do contribute toloneliness. However, it is worth re-iterating that according to our viewpoint,it is not achieved levels of contact per se that are crucial: rather, therelationship of achieved to desired (or needed) levels of contact should be

taken into consideration. More will be said about this later in the chapter.

Personal factors contributing to loneliness

Individual characteristics that make it difficult for a person to establish ormaintain sati sfactory relationships increase the likelihood of loneliness.

Such characteristics as shyness, self-esteem and physical attractiveness mayaftect loneliness in several related ways. First, characteristics that reduce aperson's social desirability may limit the person's opportunities for social

relations; secondly, personal characteristics influence a person's own

behaviour in social situations; thirdly, personal qualities may determine how

42 DANIEL PERMAN AND

a person reacts to changes in his or her achieved soinfluence how effective the person is in avoiding, miloneliness. This section discusses characteristics that prloneliness.

Shyness. Shyness, definedas a"tendency to avoid sofail to participate appropriately in social situations" (P

an important contributor to loneliness. Significant corrreports of shyness and loneliness have been found byJones ei al. (in press). Recent work by Pilkonis (1977) ways in which shy people's verbal and nonverbal behavinteraction , for instance, by not taking the initiative in

Sermat (1980) has indicated that lonely men are lowerisk-taking: while Cutrona's research (in press) hasstudents are introver ted and lacking in assertiveness. Tfactors - shyness, low social risk-taking, lack of asciousness in social situations-may well contribute t

Self-esteem. There is considerable evidence that hand in hand with greater loneliness (Loucks, 197

Jones er al. (in press) found a significant correlation UCLA loneliness scale and on Coopersmith's self(1980) reported that lonely individuals scored lowself-actualization and inner-directedness subscales ofOrientation Inventory. Eddy (1961) found a significaloneliness and an indirect measure of self-esteem, ththe person's actual and ideal self concepts.

The link between self-esteem and loneliness is re

self-esteem (and correlated factors such as shyness ansocial risks) may foster loneliness but, at the same

self-esteem may blame themselves for social "failulevels of social contact, and thus reinforce their own

Socialskiffs.Weiss (1973) and others have suggesskills, perhaps stemming from childhood, may be ass

In some instances, people with adequate social skillsperforming effectively by anxiety or shyness. In othe

may not have learned essential social skills. Whatstudents (see Horowitz et ul., in press) report"inhibthey report problems making friends, introducing th

in groups, enjoying parties, making phone calls to and the like.

The argument here is that people with poor social

satisfying social relationships, and so experience


7/13

SOCIAL PSYCHOLOGY OF LONELINESS 43difficulty with the reasoning is evidence that loneliness is not invariably

correlated with objective characteristics of a person's social life. Forinstance, young adults appear to have mote contacts with friends than dosenior citizens (see Dickens and Perlman, 1981) yet loneliness is moreprevalent in young adulthood than in old age, Several factors may operate to

produce these results. First, measures of "objective" social relationships,corresponding t o the achieved level of social relations in our definition, do

not consider the individual's desires for the number and kind of relationships

to have. Perhaps seniors have fewer social needs than young adults. Ourposition suggests that objective indices of frequency of interaction are less

appr opri ate predictors of loneliness than are indications of the discrepancy

between achieved and desired levels of social interaction. In addition, it

seems likely that over time, people with very low levels of social contact mayadapt (see Weiss, 1973) and lower their desired level of social relations.

Regardless of the quantity of their social contacts , emerging evidence

suggests that lonely people have a different style of interacting. WarrenJones (in press) videotaped conversations between strangers. Ratings of

these ta pes showed important differences between the social behaviours of

lonely and non-lonely subjects. Lonely subjects made more self-statements,they asked fewer questions of their partners, and they changed the topicmore frequently. Furthermore, lonely subjects responded more slowly to

their partners' statements. Overall, Jones characterized the interaction styleof lonely individuals as"self-focused and non-responsive", and concludedthat this style had detrimental effects for the establishment and maintenanceof relationships.

Similarity. A consistent finding in research on interpersonal attraction is

that, other things being equal, similarity leads to liking (e.g. Dickens andPerlman, 1981). This suggests that the match between an individual and thesocial groups in which he or she participates will affect loneliness. In anygiven social situation, people who are "different" because of their racial or

ethnic background, nationality, religion, age, or interests may be more

likely to be lonely.

Demographic characteristics. Some data indicate that loneliness iscorrelated with gender, marital status, income and age. Although it may

only reflect greate r willingness to reveal their feelings, more women thanmen state that they are lonely (e.g. Weiss, 1973). For the UCLA lonelinessscale, gender effects are small and usually nonsignificant. Loneliness is

lower among married people than unmarried (Weiss, 1973). In one study,

when the unmarried group was further subdivided, loneliness was higheramong widowed and divorced people than among singles, who did not differ

from marrieds (Gubrium, 1974). There is some indication that loneliness is

44 DANIEL PERLMAN AND higher among the poor (Weiss, 1973). Finally, while l

any age, it may be more common at particular poiespecially late adolescence (Rubenstein etal., 1979).

Childhood antecedents. Two findings from surveys 1979) regarding the childhood antecedents of lonelinFirst, people whose parents got divorced experience gearlier the divorce occurred, the greater the sense oflonely respondents remembered their parents as bei

worthy, and disagreeable, whilst other respondentparents as warm, close, and helpful. Similar findings hBrennan and Auslander (1979, p. 200). They sum saying that lonely adolescents come from families maof emotional nurturance, guidance or support. The clundisciplined, and irrational". Among other fin

adolescents reported higher levels of parental rejectio

of rejection as a form of punishment and greater pawith their choice of friends. Finally, lonely offspring fthem very little encouragement to strive for popularity

Cultural and situational factors contributing to lonelines

Both broad cultural values and characteristics of spe

may contribute to loneliness.

Cultural values. Sociologically oriented theorists haresulting from cultural factors that prevent people frofactory relationships. Bowman (1975) identified increasdecreased contacts with primary groups as key sources o

et 01. (1961) characterized Americans as "other-direcabout the evaluation of others to validate self worth: ye

"paradoxically (the other-directed person) remains a lcrowd because he never really comesclose to others or22). Slater (1970) emphasized a basic conflict betweencompetition, uninvolvement and independence on the

needs for community, engagement and dependence on

The conclusion reached by many sociologists is thvalues emphasizing competition,rugged individualismincrease the incidence of loneliness. These values affindividuals, and are reflected in the structuring of social

might expect that in cultures such as China, where co


8/13

SOCIAL PSYCHOLOGY OF LONELINESS 45achievement are stressed, loneliness is less frequent (cf. Zimbardo's dis-cussion of shyness,1977).

Social norms. An individual's own expectations and desires for socialrelations are importantly affected by social norms. According to Gordon:

"It is clear to the teenager that he or she should have a date after school, and it isclear to the average man or woman that he should have a mate, family, a circle offriends."(1976, p. 15)

Cultural expectations for social relationships change with age. For instance,while it is appropriate for young children to have their primary emotional

attachment to their parents, young adults are expected to develop newattachments to dating partners and later to a spouse. When a person's social

relationships do not keep pace with age-related changes in normative stan-dards for relationships, he or she is likely to feel lonely.

One illustration of social norms can be seen in research by Larson el al. (inpress). High school students were asked to wear electronic paging devicesand, whenever they were paged, they indicated whether they were alone orwith others and they indicated how lonely they were feeling. If students werealone on week nights, they reported only moderate feelings of loneliness,but students who were alone on Friday or Saturday nights reported intensefeelings of loneliness. Here the expectation that weekends are for socialactivities appears to be changing students' reactions.

Situational constraints. In any social setting, factors that increase thefrequency of interaction and foster group cohesiveness should affect theincidence of loneliness. This includes values (e.g. the extent to which a workgroup is competitive), but extends to other normative and structured factors

in the situation as well. For example, a well-documented finding (e.g.Dickens and Perlman, 1981) is that physical proximity fosters liking. As aresult, the architecture of housing units affects social interaction and friend-ship formation. The individual who lives or works in a physically isolatedlocation may tend to be socially isolated as well.

Cognitive Processes that Modulate the Loneliness Experience

As indicated earlier, the discrepancy between desired and achieved socialrelations is typically perceived by the individual and labelled as loneliness.But, according to a cognitive perspective, this discrepancy does not leaddirectly and inevitably to loneliness. Several factors may affect the self-labelling process and the intensity of the person's reactions to their situation,

and cognitive processes play a central role in modulating the loneliness

46 DANIEL PERLMAN AN

experience. This section discusses how causa

comparison processes and perceptions of personal cWe will start however, with labelling.

Labelling

It is sometimes difficult to label subjective experiedecide if one is really lonely, or to distinguish lonelilogical states such as anxiety or depression. Cultural of loneliness and when loneliness typically occurs mFor instance, it is considered reasonable for a chillonely on a first trip to camp; and it is appropriateperson has just moved to a new city. To some extent ,own social situation with cultural definitions of lonecues as guides to labelling their personal experience.

Cross-cultural studies suggest that language may self-labelling of loneliness. According to Robert Levof the Tahitians, there exist"no. . . terms for lonelin

depressed or sad because of the lack of friends, com(p. 306). Although Levy notes that the lack of specimean that this state is unexpressible, themes of lonerare in his interviews. In contrast, Jean Briggs' (1970suggests that the Eskimo have several different Hujuujaq is the most general term, meaning "to be absence of other people". Pai refers more specificaleft behind; to miss a person who has gone". Finally

"silent and withdrawn in unhappiness, especially beother people." It is interesting to note that the first

angry loneliness including"hostility"whereas the laand depressed pattern of loneliness. The ways in whand folk understandings affect the experience of lonarea for further investigation.

Causal attributions

The search to understand the causes of loneliness is nand mental health professionals since lonely peopmotivated to explain the reasons for their lonelinunderstanding the causes of loneliness is a first scontrolling and ultimately alleviating loneliness. Tpsychological research on attribution theory indic


9/13

SOCIAL PSYCHOLOGY OF LONELINESS 47explanations for the causes of their behaviour can have important effects ontheir self es teem, expectancies for the future, affective reactions, and coping

behaviour.Of various attributional models, the work of Weiner and his colleagues

(see Weiner et d.,1978) is most relevant to our purposes. He has appliedattribution theory to the achievement domain and this focus is useful forunderstanding loneliness because, in most western societies, one's social

relationships are an indicat ion of success; as Gordon (1976) observed about

Americans, "To be lonely is to have failed". Weiner has demonstrated that

causal attributions can be classified along two primary dimensions: locus of

causality (interna l or personal, versus external or situational) and stability

(stable versus variable over time). For instance, saying"I'm lonely because

I'm unattractive" would represent an internal, stable attribution whereassaying"I'm lonely because I've just moved"would represent an external,unstable attribution. More recently, Weiner has proposed the addition of a

third dimension ofcontrollability, which concerns whether or not people

perceive themselves as having control over the factors that caused theirbehaviour.

Consequences of attributions

According to Weiner's model, the stability dimension is especiallyimportant for the person's future expectancies. Perceiving that loneliness is

due t o stable causes should lead a person to anticipate prolonged loneliness;unstable causes should lead to greater optimism about improving one's

social life. The locus of attributions should have greater impact on the

person's self-

esteem, with self-

blame and lowered self-

worth accompanyinginternal a ttributions. Predictions linking attributions to affective states aresomewhat more complicated (Weiner etal. 1978): internal attributions forloneliness should magnify feelings of shame and inadequacy but stable,

internal attributions should heighten depression-related affects of feelinghopeless, helpless, aimless, or depressed. Finally, Weiner suggests that the ,

dimension of controllab ility is most closely related to other people's evalua-tions of and liking for the lonely individual.

Several studies conducted at UCLA (Michela et al., 1980; Peplau et al.1979) have tested the applicability of Weiner's model to loneliness. One

study (Micheia et al., 1980) examined students' perceptions of commoncauses of loneliness and found that dimensions of internal ity, stability and

control were salient in lay conceptions of loneliness. Other studies ofself-attributions for loneliness (summarized in Peplau et&., 1979) have corro-

borated the proposed link between stable causes for loneliness and

pessimism or low expectancies for the future. Evidence has also been found


that feelings of depression are most likely to accom

self-attributions are stable and internal. In the colle

tively infrequent attributions for loneliness as low p

and fear of rejection were associated with particdepression.

Antecedents of attributions

Given the potential importance of attributions floneliness, it is useful to consider how people m

causality . Kelley (1967) and others have identified aconcerning the attribution process. Their work sugge

should be most likely to make internal or personal attr

feel lonely in many different situations (low distincti

most other people in similar situations are not lonely(3) feel lonely over time (high consistency). In contrasonly a few situations and is felt by others in those external or situational attributions are more likely.

Two important implications of this analysis shoullonged loneliness should foster internal, stable attristudies of college students, Peplau et al. 1979, indica

loneliness is related to internality of attributions);

avoid talking about their loneliness, thereby creatin

listic ignorance. If this is the case, lonely people

uniqueness of their response, and assume that mosatisfactorysocial lives. This should also foster intern

Social comparison and perceived controlIn the process of evaluating a social deficiency, s

attributions may act to modulate one's experience of l

one's social relations, social comparisons with other

may be important (Pettigrew, 1967). The lonely newcompare his or her success in making new friends to and believing that others are doing better at making

feelings of loneliness . Evidence in support of thilongitudinal study of new students at college (Cutron

that loneliness was strongly related t o satisfaction witships, which in turn was related to comparisons with

one's own previous relationships. It appears that so

cesses may affect how large or important a social defic

A final modulator of the loneliness experience is tindividual can exercise personal controlover his o


10/13


11/13

SOCIAL PSYCHOLOGY OF LONELINESS 51

Thus in general the rewards of interacting with a lonely person may belimited but two major exceptions to this generalization should be noted.

First, people may react more positively if the lonely person is a high status orvery attractive newcomer. In this instance, loneliness is attributable to

situational factors, and the possibility of establishing an enduring relation-

ship may have considerable appeal. Second, reactions to loneliness shouldbe more positive if there is a pre-existing relationship with the lonely other.

If the lonely person is a spous e, relative or friend, th ere may be a long historyof shared helping and support, and the expectation of future reciprocity of

nurtur ance. In this instance, being able to help an intimate may be perceivedas rewarding.

Attributional factor s. Reactions to a lonely person are affected by causalattributions about why the person became lonely or has continued to be

lonely. Gordon (1976) suggested that, just as lonely people blame them -

selves for loneliness, so, too, observers may also blame the lonely, and

hence react negatively. There are some data (Peplau et al., 1979) to

document the impact of perceived causal attributions on evaluations oflonely people. It appears that sympathy and liking are greatest for lonelypeople who ar e judged to have had little control over the initiation of their

loneliness, and who have made an effort to overcome their loneliness.

Personality factors.Certa in people may be more likely to sympathize withlonely individuals and this capacity to empathize with the lonely may befacilitated by personal experiences of loneliness, and by perceptions of beingsimilar to t he lonely individual. In a study of psychological androgyny, Bernet al. (1976) found that students whose self-conceptions were androgynousor feminine reacted more effectively in interactions with a confederatedescribing himself as a lonely transfer student than did those with masculineself-conceptions. Individual differences in beliefs about the extent to which

people can control their outcomes as measured by scales of locus of control(e.g. Rotter, 1966) or belief in a just world (e.g. Rubin and Peplau, 1975)may also be relevant.

Regardless of whatever individual differences there may be in responses

to lonely people, as we indicated earlier, we believe these responses mayhave implications for the persistence of loneliness since negative stereotypes

and reactions may aggravate loneliness whilst sympathy and efforts toextend social suppor t may help alleviate the problem of loneliness.

Of course, other people's reactions are not the only factor in how well

people deal with loneliness: lonely people use a number of strategies toalleviate their condition, and helping professionals also have various

therapeutic techniques for intervening (see Rook and Peplau, in press). In

the last section of this chapter, we will review people's own efforts to


overcome loneliness and consider outcome researcsuccess of the rapy for alleviating loneliness.

CopingwithLoneliness

In line with our definition of loneliness, it is conveniestrategies into three broad groups. Coping strategdesired level of social contact, (2) the achieved leve

(3) the importance and/or perceived magnitude of thand achieved levels of contact.

Changing one's desired level of social contact

One general approach to "loneliness managemendesired level of contact, which may be accomplidifferent ways.

Adaptation. Over time, people's expected and drelations tend to converge to their achieved level. F

(1964) found that old people with a long history of sbeen "loners" for some time, were less likely to repold peop le with higher levels of social participatiocommented on the possibility that over time lon"change their standards for appraising their situatio

particula r, tha t standards might shrink to conform mof bleak reality". Weiss does not, however, view this

to loneliness.

Task choice.A second way people can alter thei

contact is to select tasks and situations that they eperson who enjoys reading alone, but only likes tcompanion: this person might avoid arousing fee

spending the evening reading rather than going to t

views that Robert Brown (1979) has conducted loners suggest that people who seek prolonged solitureper toire s of activities they find enjoyable doing alo

Rook and Peplau, in press) have gone so far as to sment in solitary activities as a useful way of alleviwho use this response to loneliness, report ge

(Rubenstein and Shaver, 1980).

DANIEL PERLMAN AND


12/13

54 DANIEL PERLMAN ANDSOCIAL PSYCHOLOGY OF LONELINESS 53

Changed standards. A third technique which people use to reduce theirdesired level of social contact is to change their standards for who is accept-able as a friend. As an example of this phenomenon, consider a professionalwho usually for ms friendships with other high status professionals: if thisperson became lonely, he or she might be willing, even happy, to formfriendships with a much wider set of people. In the UCLA study (Cutron a,in press), i ncreased satisfaction with one's friendships was a strong predictorof recovering from loneliness.

Achieving higher levels of social contact

Perhaps the most obvious way of overcoming loneliness is to establish orimprove social relationships, and in the UCLA study, "finding a boyfriend1girlfriend"was perceived as being the best way to overcome loneliness. On ecan think of many ways of achieving higher social contact: making oneselfmore physically attrac tive, joining clubs, initiating conversations with otherpeople, de epening existing relationships and the like.

In the UCLA study, Cutrona (in press) divided the initially lonelystudents into those who, during the year, did and did not overcome lone-liness. In this study, neither a change in dating statusno r the strategies usedto form relationships had statistically significant effects on overcomingloneliness. Howev er, a s we would expect, a change in number of friendshipswas important: students increasing their friendship networks decreasedtheir loneliness. Similarly, loneliness is less frequent and more transient forpeople wh o react to it by visiting or calling a friend (Rubenstein and Shaver,1980).

Minimizing loneliness

A third major way to cope with loneliness is to alter the importance andlorperceived magnitude of the gap between desired and achieved levels ofsocial interac tion. At least four variations on this theme can be identified:first, lonely pe ople may simply deny that there is a discrepancy between theirdesired and achieved levels of social relations; secondly, lonely people candevalue social contact and rationalize their plight by saying that otherobjectives ar e more important, or by contending that loneliness is a "posi-tive growth experience", thirdly, people can try to reduce loneliness-induced deficits by gratifying their needs in alternative ways (for instance, ifloneliness threatens a person's sense of self-esteem, he or she might engagein non-social means of bolstering self-regard); finally, people can engage in

behaviours designed to alleviate the negative impact of loneliness. One

example of this, consistent with speculation linking loand drug use, would be drinking "to drown one's sorr

Therapeutic interventions

Given the diversity of factors that may precipitateliness, no single cure-all is likely to be found, but museful when appropriately employed (Rook and Panalysis of loneliness suggests a few guidelines for the

First, t o be effective, interventio ns should be taproblems of the lonely individual: a recent widow

social support whereas a college student who has nevneed help with h i she r social skills. The fairly extensivsexual-social anxiety" (reviewed by Cur ran , 1977) approaches t o aiding students who ar e fearful of datiindividual, therapy might emphasize desensitizationthe cor rection of faulty self-evaluations of perfo rmansocial skills training to build a more adequate behcomprehensive analysis of the antecedents of loneliactional styles of lonely people (Jones, in press) will the design of successful therapeutic interventions.

Second, interventions to help the lonely may need individual's own explanations for the causes of his oral. (1979) suggest that people may often underestimsituational causes of loneliness and overestimate th e rOn theoret ical grounds, we would expect this tendenc

in cases whe re loneliness is severe and endur ing. ConsCutrona's (in press) longitudinal study ofUCIA studliness to internal, personal causes in the fall was assopersisting over the academic year. Overestimatinpersonal factors is encouraged by the emphasis in bpsychological thinking on a characterological theory1973). In fact, loneliness typically results from a poindividual's interests, social skills or personal charactsocial environment . Careful consideration should be gof both personal and situational causes for lonelinesalso tend to underestimate the potential changeabilityFor example, they may focus on irrevocable precipitdeath of a spouse), rather than on factors that impedenew, more satisfactory social life. These maintainingsuch as shyness or limited opportunities to meet

SOC S C O OG O O SS56 DANIEL PERLMAN AND


13/13

SOCIAL PSYCHOLOGY OF LONELINESS 55amenable to change. We would advise directing clients' attention to factorsthey can control.

Third, we believe lonely individuals should be encouraged to view their

world more positively. Some degree of negativism may reflect the reality of

their situations, but some of it is undoubtedly due to a negativity bias in theirevaluations. For instance, after interacting with a randomly assignedstranger, lonely subjects rated their partners more negatively than didnon-lonely subjects, an effect that appears t o be in the eye of the beholder

(Jones, in press). We believe that curbing such negative perceptions shouldhelp people overcome their loneliness.

Finally, efforts to reduce loneliness must go beyond the individual to

consider social and cultural factors that foster loneliness. As Gordon (1976,

p. 21)noted,"Mass loneliness is not just a problem that can be coped with bythe particular individuals involved; it is an indication that things are drama-

tically amiss on a societal level". Social institutions might consider ways toassist such at-risk groups as new students, transferred business executives

and their families, or nursing home residents. In addition, social programsfor other groups such as the newly divorced or widowed who are notassociated with a particular institution are useful. Indeed, it seems likely thatinterventions aimed at specific problems-such as retirement or moving toa new community-may be more effective than interventions directed moreglobally at "loneliness".

Outcome research on the success of therapy for alleviating loneliness

Although therapy outcome research on the treatment of loneliness is

limited, it is encouraging. In our opinion, one of the most noteworthy

studies was done by Jones (in press) and his colleagues. Jones' researchgroup had identified three unique character istics of the way lonely people

interact in conversations: (a) they make fewer other-references and askfewer questions of their partner; (b) they change the topic more often; and(c) they delay longer in filling gaps in the conversation. Jones thereforedeveloped a short"social skills training program" to help studentsovercome

these interpersonal deficits, incorporating explanation, modelling, practicewith prompting, and feedback on the students' performance of target

behaviours. Compared with a no treatment and a placebo treatment (con-

versation only) control group, the skills training produced desired changes

in the participants' interactional styles and it reduced their loneliness.Indeed , the magnitude of the reduction in loneliness was appreciable

compared to that repor ted in most psychological research.


Summary and Conclusions

We have now come full cycle in this review. We st

concept and analyzing what leads up to it; then we infactors modera te the intensity of affective reactions tand subsequently indicated how others react to lonhave indicated ways of reversing the process or allev

At this time (circa 1981), loneliness is a topic riphave documented that loneliness is an unpleaexperience. Loneliness is also associated with a vari

such as juvenile delinquency, alcohol abuse and

literature on loneliness is relatively small (and thu

flourishing: and useful theoretical concepts and dat ahave been developed. Although the experimental

liness by researchers may be difficult and raises eth

research strategies have proved fruitful. Initial effo

liness empirically have been rewarded and available psychologists are beginning to learn how to help peop

Yet despite these advances in the field, many impounanswered.

All these factors make loneliness an attractiv

Whatever else happened to the study of loneliness in the most important development was that this rese

closet". Loneliness is now justifiablya social science t

Documents

Toward a Social Psychology of Loneliness (pág. 31...)