Ann C. Gerson and Daniel Perlman. University of Manitoba, Winnipeg, Canada. This experiment examined the communication skills of 66 female undergrad-.
Journal of Abnormal Psychology 1979, Vol. 88, No. 3, 258-261
Loneliness and Expressive Communication Ann C. Gerson and Daniel Perlman University of Manitoba, Winnipeg, Canada This experiment examined the communication skills of 66 female undergraduates who were either chronically lonely, situationally lonely, or not lonely. "Sender" subjects were videotaped while they watched and rated the pleasantness of 25 slides. These videotapes were then shown to "receiver" subjects who made judgments about the sender subjects' reactions to each slide. The results indicated that the situationally lonely subjects were more successful as communication senders than were the chronically lonely or nonlonely subjects. This finding was interpreted within a cognitive framework as reflecting the increased motivation created by attributing one's situational loneliness to unstable causes. Beck depression scores were also obtained. In general, highly depressed subjects were less successful as communication senders, but this relation did not hold for the situationally lonely group.
Loneliness reflects a deficiency in one's social relationships that is almost always accompanied by a gnawing sense of discomfort. As a phenomenon, it is widespread. We believe that loneliness experiences can profitably be divided into two classes: those which are chronic versus those which are temporary or situationally induced. In writing on this topic, psychologists have advanced widely discrepant views of the motivational consequences of loneliness. On the one hand, Sullivan (1953) considered loneliness to be arousing, a "driving force." On the other hand, Fromm-Reichmann (1959) and others have argued that lonely people are apathetic, passive, and depressed. Peplau and Perlman (1979) offered an attributionally based view of loneliness that suggests a resolution of this controversy. Assume that situationally lonely people commonly attribute their feelings to unstable causes, whereas chronically lonely people commonly attribute their condition to interAnn Gerson is now at the Family Health Program, 323 South 6th Street East, Salt Lake City, Utah 84102. The authors wish to thank L. Anne Peplau for her contributions to their work. Requests for reprints should be sent to Daniel Perlman, Department of Psychology, University of Manitoba, Winnipeg, Canada R3T 2N2.
nal, stable causes; then, according to attribution theory, the situationally lonely people should have higher expectations that their loneliness will change. In turn, Peplau and Perlman maintained that situationally lonely people should be highly aroused, whereas chronically lonely people should be apathetic. Motivational arousal has often been linked with enhanced expressiveness. Thus situationally lonely individuals, being in an aroused state, should be especially successful in emotional communication. The primary purpose of the present experiment was to test this prediction. If situationally lonely persons, being aroused, are especially effective in sending emotional messages, then one might expect chronically lonely persons, being despondent, to be poor communicators. Two previous studies provided encouragement for this line of reasoning. First, Miller, Caul, and Mirsky (1967) reported deficits in the communication skills of monkeys following prolonged social isolation. Second, Prkachin, Craig, Papageorgis, and Reith (1977) found that chronically depressed patients were poor communication senders. (These patients were, however, equal to normal subjects as communication receivers, suggesting that the effects of loneliness may hold only for expressiveness.) Given the impetus of these data,
Copyright 1979 by the American Psychological Association, Inc. 002I-843X/79/8803-0258$00.75
258
LONELINESS AND COMMUNICATION
a second purpose of the present study was to explore further the relation of communication skills to chronic loneliness and depression. Method Subjects Over 300 female undergraduate students taking introductory psychology courses were screened for participation in this study. By serving in this study, the students partially fulfilled an introductory psychology research-participation requirement. In agreeing to serve in the study, the subjects knew they would be videotaped but were not told when the taping would occur. The decision to consider only females was based both on practical considerations and on the fact that Prkachin et al. (1977) had used females. Naturally, this restriction limits the generality of the findings, but we had no a priori reasons for believing the results should be gender specific. A total of 66 subjects were selected on the basis of their responses to the UCLA Loneliness Scale (Russell, Peplau, & Ferguson, 1978). The students completed two versions of this scale: one indicating how they had felt during the past two weeks or so, and a second indicating how they usually felt in their life. Most subjects (« = 56) also completed the Beck Depression Scale (Beck, 1967). The distributions of scores for both versions of the loneliness measure were used in constituting the three groups of subjects. The members of the nonlonely group (n = 24) had scores in the lower third of the distributions for both recent (M — 28) and general (M = 29) loneliness. The members of the situationally lonely group (n = 19) had scores in the top third of the distribution for recent loneliness (M — 52), but in the lower third for general loneliness (Af = 34). The members of the chronically lonely group (n — 23) had scores in the top third for both recent (M = 55) and general (M — 60) loneliness.
Procedure The communications paradigm used in the present study was developed by Buck (1978). The procedure involved first videotaping the subjects' expressive behavior during exposure to a set of emotionally loaded slides (Session 1) and later having four other subjects decode this behavior (Session 2). During the first session, the subjects were run individually and the videotaping was done in an unobtrusive manner through a one-way mirror. Each subject was shown a set of 25 slides, 5 from each of the following content categories: sexual, scenic, novel, pleasant people, and unpleasant people. As each slide was shown, the subject described her feelings about the slide and then rated it on a 7-point pleasantness scale.
259
During Session 2, each subject viewed four videotapes which had been recorded during Session 1. Each observer viewed the expressive behavior of at least one person from each of the three subject groups. The observer subjects saw—but did not hear—the sender subjects' reactions to each of the 25 slides. The observer's task was to guess the category of the slide the sender was seeing, and to estimate the sender subject's ratings of the slide's pleasantness. During Session 2, four observer subjects were run simultaneously.
Dependent Measures Four measures were derived from this paradigm, two reflecting the subject's abilities in expressing themselves and two reflecting the subjects' accuracy as receivers. Each subject's category expressiveness score consisted of the number of times her observers correctly guessed the type of slide the sender was viewing. Since four observers each watched the sender react to 25 slides, the maximum possible score was 100. Each subject's pleasantness expressiveness score reflected the correlations between the sender subject's pleasantness ratings and the four observer subjects' guesses of her ratings. Four separate correlations were obtained and then averaged. The receiver accuracy scores were calculated from the same data base in an analogous but slightly different manner. Whereas each sender expressiveness score reflected one sender's ability to communicate to four receivers, each receiver accuracy score reflected one receiver's ability to interpret the expressive behavior emitted by jour senders. Thus, each receiver category accuracy score reflected the number of times one subject as a receiver correctly guessed the type of slide being viewed by four sender subjects. Each receiver pleasantness accuracy score reflected the correlations between one receiver subject's guesses and the actual pleasantness ratings made by four different sender subjects.
Results Results were analyzed using a separate oneway analysis of variance for each of the four dependent measures. There was a significant main effect of loneliness on sender expressiveness, both for category transmission, ^(2, 63) = 4.24, p < .02, and for pleasantness transmission, F(2, 63) =3.3 7, p < .04. Pairwise comparisons among the means, controlled for hypothesis-wise error rate, were performed via orthogonal Scheffd S tests. As predicted, the situationally lonely subjects were more expressive by both measures; the chronically lonely and the nonlonely groups did not differ significantly from one another. The mean cate-
260
ANN C. GERSON AND DANIEL PERLMAN
gory expressiveness scores for the three groups, respectively, were 30.24, 26.20, 24.64; the mean pleasantness expressiveness scores (correlations) were .27, .17, and .15. Loneliness did not have any significant effect on either receiver accuracy measure. The Beck depression scores of the subjects, where they were available, were also analyzed. As expected, the situationally lonely individuals (M = 42.S) and the chronically lonely individuals (M = 42.9) were significantly more depressed than the nonlonely individuals (M = 32.8), F(2, 53) = 12.35, p < .0001. The two lonely groups did not differ significantly from each other in terms of their depression scores. Pearson product-moment correlations were also calculated between depression scores and sender expressiveness. Across all conditions, the depression-pleasantness correlations approached statistical significance, r(54) = -.21, p < .06. As Prkachin et al. (1977) had found, the more depressed subjects were less successful as communication senders. Further analyses revealed a clear inverse relation between depression and expressive success both for the chronically lonely group [for category, r(20) = -.21, p < .09; for pleasantness, r(20) = -.62, p < .001] and for the nonlonely group [for category, r(16) = —.30, p < .06; for pleasantness, r(16) = -.29, p < .06]. However, among the situationally lonely subjects, the depression-communication sending correlations were near zero [for category, /•(14) = —.09, p < .38; for pleasantness, r(14) = -.03, p < .45]. One paradox in the overall pattern of data should be noted. Given the correlational evidence, one might expect the most depressed subject group (the chronically lonely) to be the poorest communication senders, and the least depressed subject group (the nonlonely) to be the best communication senders. This expectation was not confirmed. Instead, as indicated by previously cited results, there was no significant difference in the sending ability of these two groups. Discussion Perhaps the most important finding of this study was the greater success of situationally
lonely subjects as communication senders. This result was as predicted from the view that situationally lonely people are motivationally aroused. However, the findings do not justify extension of this viewpoint to reach the conclusion that chronically lonely people are poor communicators. Naturally one may wonder: Why weren't situationally lonely subjects also superior as communication receivers? This is especially perplexing since some psychologists (Weiss, 1973) see loneliness as arousing vigilance about interpersonal relationships and producing oversensitivity to minimal cues. The receiver accuracy data are incompatible with this view. Instead, the receiver accuracy results are more consistent with recent evidence (Jones, Note 1) that lonely people are more selffocused in their interaction patterns. In particular, lonely people talk more about themselves and ask others fewer questions. Perhaps the onset of loneliness experienced by members of the situationally lonely group generates an egocentric orientation detrimental to receiver accuracy. The relation between depression and expressiveness also merits comment. Overall and within two of the three subject groups, depressed individuals tended to be less successful as communication senders. These results are consistent with the earlier findings reported by Prkachin et al. (1977). However, this correlation did not hold in the situationally lonely group. Prkachin and his associates interpreted the poorer sending skills of depressed individuals in terms of Lewinsohn's (1974) framework. According to this view, poor social skills predispose people to depression because such deficiencies result in a low rate of responsecontingent positive reinforcement. Inasmuch as social skills are relatively enduring, this process would seem to be most important in explaining people's "typical" level of depression. When people temporarily encounter unpleasant situations, their sense of depression may increase because of factors other than their lack of social skills. No relation between depression and expressiveness would be ex-
LONELINESS AND COMMUNICATION
pected. As the situationally lonely subjects may well have been temporarily facing adverse circumstances, the failure to find significant depression-expressiveness correlations within this subsample is not surprising. To summarize, this study has demonstrated the success of situationally lonely subjects as communication senders. As in the Prkachin et al. study, greater depression was generally asssociated with poorer expressive communication, but this relation did not hold for the situationally lonely group. In conclusion, it is worth noting that many types of loneliness have been postulated but few have been empirically demonstrated. The results of this study testify to the importance of the chronic versus temporary distinction in trying to reach a better understanding of loneliness experiences. Reference Note 1. Jones, W. H. The persistence of loneliness. In D. Perlman (Chair), Toward a psychology of loneliness. Symposium presented at the meeting of the American Psychological Association, Toronto, 1978.
References Beck, A. T. Depression: Causes and treatment. Philadelphia: University of Pennsylvania Press, 1967.
261
Buck, R. The slide-viewing technique for measuring nonverbal sending accuracy: A guide for replication. JSAS Catalog of Selected Documents in Psychology, 1978, 8, 63. (Ms. No. 1723) Fromm-Reichmann, F. Loneliness. Psychiatry, 1959, 22, 1-15. Lewinsohn, P. M. Clinical and theoretical aspects of depression. In K. S. Calhoun, H. E. Adams, & K. M. Mitchell (Eds.), Innovative treatment methods in psychopathology. New York: Wiley, 1974. Miller, R. E., Caul, W. F., & Mirsky, I. A. The communication of affects between feral and social isolated monkeys. Journal of Personality and Social Psychology, 1967, 7, 231-239. Peplau, L. A., & Perlman, D. Toward a social psychological theory of loneliness. In M. Cook & G. Wilson (Eds.), Love and attraction: Proceedings of an international conference. Oxford, England: Pergamon, 1979. Prkachin, K. M., Craig, K. D., Papageorgis, D., & Reith, G. Nonverbal communication deficits and response to performance feedback in depression. Journal of Abnormal Psychology, 1977, 86, 224234. Russell, D., Peplau, L. A., & Ferguson, M. L. Developing a measure of loneliness. Journal of Personality Assessment, 1978, 42, 290-294. Sullivan, H. S. The interpersonal theory of psychiatry. New York: Norton, 1953. Weiss, R. S. (Ed.), Loneliness: The experience of emotional and social isolation. Cambridge, Mass.: MIT Press, 1973. Received January 22, 1979 •
