Temperamental Contributions to Social Behavior

About 15% of Caucasian children in the second year of life are consistently shy and emotionally subdued in unfamiliar situations, whereas another 15% are consistently sociable and affectively spontaneous. A majority of the children in these two groups retain these profiles through their eighth year. In addition, the two groups differ in physiological qualities that imply differential thresholds in limbic sites, especially the amygdala and the hypothalamus, suggesting that the two temperamental groups are analogous to closely related strains of mammals. However, the behavioral profiles of the children are influenced in a major way by environmental conditions existing during the early years of life. The word temperament is used by most, but not all, behavioral scientists to refer to those psychological qualities that display considerable variation among infants and, in addition, have a relatively, but not indefinitely, stable biological basis in the organism's gcnotype, even though the inherited physiological processes mediate different phenotypic displays as the child grows. It is reasonable to suggest that some of the temperamental differences among children are analogous to the biobehavioral differences among closely related strains of dogs, cats, or monkeys (Adamec & Stark-Adamec, 1986; Clarke, Mason, & Moberg, 1988). The temperamental qualities that are most obvious to contemporary American parents, and that are investigated most often by psychologists, include irritability, smiling, motor activity, and adaptability to new situations. These qualities are popular, in part, because they have implications for the ease with which parents can socialize their infant. It is not clear at the moment how many temperamental qualities will be discovered; it certainly will be more than 6, but hopefully less than 60. We will have to wait for history's answer. Inhibited and Uninhibited Chi ld ren Steven Reznick, Nancy Snidman, and I, together with Cynthia Garcia-Coll, Wendy Coster, Michcle Gersten, and many others in our laboratory, have been studying two categories from the larger set of temperamental qualities (Garcia-Coll, Kagan, & Reznick, 1984; Kagan, Reznick, Clarke, Snidman, & Garcia-Coll, 1984; Kagan, Reznick & Snidman, 1987, 1988; Kagan, Reznick, Snidman, Gibbons, & Johnson, 1988; Reznick et al., 1986). The original behavioral referent for each of the qualities was the response profile of 20to 30-month-old children when they were in unfamiliar situations. Some children consistently become quiet, vigilant, and restrained while they assess the situation and their resources before acting. Others act with spontaneity, as though the distinctions between familiar and novel situations were of minimal psychological consequence. The situations that best reveal these two qualities in young children are encounters with unfamiliar children or adults, perhaps because other people are the most frequent basis for categorizing most settings as unfamiliar. Of course, it is rare to find a large number of children who are consistently shy and affcctively restrained or outgoing and spontaneous regardless of the social context. There is, however, a small group of children (my colleagues and I estimate it to be about 10% to 15%) who usually bring one or the other of these behavioral styles to new situations. We call the shy children inhibited and the sociable children uninhibited. Our current studies of inhibited and uninhibited children trace their beginnings to an early collaboration with Howard Moss, which was summarized in 1962 in the book entitled Birth to Maturity (Kagan & Moss, 1962). A large group of families was participating in the Fels Institute's longitudinal project, which began in the early 1930s. The children in these families were observed from birth to adolescence in their homes, the Institute's nursery school, and their own school settings, and they were tested and interviewed regularly. Moss rated each child on a set of variables for consecutive, chronological epochs, using as evidence the extensive corpus of information available on each subject. I was in another room interviewing these same subjects, who were then in their 20s, and administering a relevant battery of tests, but I was unaware of the early information Moss was reading. It is of interest that the most important discovery of the Fels study was that the only psychological quality preserved from the first three years of life through adulthood was the characteristic we now call behavioral inhibition, although we called it passivity in 1962. Although Moss and I suggested that this predisposition might be a partial function of biological variables, the Zeitgeist during the early 1960s was not prepared to award much formative power to temperamental factors. Unfortunately, our faith in a temperamental interpretation of these data was not suf668 April 1989 • American Psychologist Copyright 1989 by the American Psychological Association, Inc. 0003-066X/89/$00.75 Vol. 44, No. 4, 668-674 ficiently strong, and neither of us pursued this phenomenon. In a later collaboration, Richard Kearsley, Philip Zelazo, and I enrolled Chinese-American and Caucasian infants from similar social class backgrounds in a longitudinal study of the effect of day care across the period from 3 to 29 months of age. We noted in our 1978 monograph, called In fancy (Kagan, Kearsley, & Zeiazo, 1978), that although the effect of day care on the children was minimal, the Chinese infants, whether attending our day care center or raised only at home, were, relative to the Caucasians, more subdued, shy, and fearful when they met unfamiliar adults or children, and they cried more intensely when their mothers left them for a brief separation. In addition, the Chinese children consistently showed more stable heart rates than the Caucasians during the laboratory episodes. This association implied a biological basis for the inhibition among the Chinese children. The unexpected association between shy, timid behavior and a minimally variable heart rate provoked me to pursue this phenomenon more directly. Cynthia Garcia-Coll and Nancy Snidman, in independent dissertation research, selected from large samples of young Caucasian children (aged 21 months for Cohort 1 and 31 months for Cohort 2) those who were either consistently shy and fearful (behaviorally inhibited) or sociable and fearless (uninhibited) when they encountered unfamiliar people or objects in unfamiliar laboratory rooms. They had to screen over 400 children in order to find 54 consistently inhibited and 53 consistently uninhibited children, about 15% of the children screened, with equal numbers of boys and girls in each group. These children have been seen on three additional occasions; at the last assessment at 71/2 years of age, there were 41 children in each of the two cohorts---a loss of about 20% of the original sample. In each of the assessments, the children were observed in different situations. Usually the assessments inc ludeda testing session with a female examiner and, on a different day, a play situation with an unfamiliar child of the same age and sex. At 51/2 years of age the aggregate index included observations of the child's behavior in his or her school setting (Gersten, 1986). Details of the procedures can be found in previously published articles (see Garcia-Coll et al., 1984; Kagan et al., 1988, Reznick et Editor's note. This article was originally presented as a Distinguished Scientific Contributions award address at the meeting of the American Psychological Association in Atlanta in August 1988. Award-based manuscripts appearing in the American Psychologist are scholarly articles based in part on earlier award addresses presented at the APA convention. In keeping with the policy of recognizing these distinguished contributors to the field, these submi.~sioas are given special consideration in the editorial selection process. /a~thor's note. The research for this article was SUPl~rted by the John D. and Catherine 1". MacArthur Foundation. I thank J. Steven Reznick, Nancy Suidman, Jane Gibbons, and Maureen O. Johnson for their contributions. Correspondence concerning this article should be addressed to Jerome g~a~ , Delgtrtment of Psychology, Harvard University, William James Hall, 33 Kirkland St., Cambridge, MA 02138. al., 1986; and Snidman, 1984). We computed aggregate indexes of inhibition at each age, based on the child's tendency to be quiet, shy, and emotionally subdued in each of the different contexts. The indexes of inhibition at 71/2 years were based on behavior in two laboratory situations. The first was a play situation involving 7 to 10 unfamiliar children of the same age and sex. The two critical variables were number of spontaneous comments to the other children or supervising adults and proportion of time spent standing or playing apart from any other child in the room during the free-play intervals. The second assessment context was an individual testing session with an unfamiliar female examiner who did not know the child's prior status. The two critical variables were latency to the sixth spontaneous comment to the examiner and the total number of spontaneous comments over the 90-minute session. The aggregate index of inhibition represented the average standard scores for the indexes from the two assessment situations. The intercoder reliabilities for these behavioral variables coded from videotapes were above 0.90. Preservation of Behavior There was moderate but significant preservation of the inhibited and uninhibited behavioral styles from the first assessments, at either 21 or 31 months, through 71/2 years of age. The correlation between the original index of inhibition (21 months for Cohort 1 and 31 months for Cohort 2) and the aggregate index at 71/2 years was .67 (p < .001) for Cohort 1 and .39 (p < .01) for Cohort 2. About three fourths of the children in each cohort retained their expected classification, based on whether their standard score on the aggregate index at 71/2 years was positive or negative. Furthermore, the children who exhibited most extreme behavior initially were most likely to maintain their behavioral style. Although one half of the original group of inhibited children from both cohorts no longer displayed an extreme degree of shyness at 71/2 years, most (80%) had still not acquired the unusually spontaneous demeanor characteristic of the typical uninhibited child. A smaller number of uninhibited children became shy; 10% of the original group of uninhibited children had become very timid at 71/2 years of age (see Figures 1 and 2). Additionally, about three fourths of the inhibited 71/2 year olds, compared with only one fourth of the uninhibited children, had one or more unusual fears, such as speaking voluntarily in the classroom, attending summer camp, remaining alone in the home, taking out the rubbish at night, or going to their bedroom alone in the evening. Furthermore, one third of the siblings of the inhibited children, but not one sibling of an uninhibited child, had one or more of these unusual fears. The inhibited children also displayed characteristics indicative of caution and motor tension. For example, when they were 51/2 years old an examiner asked them, as part of a play episode, to fall backward onto a mattress. Significantly more inhibited than uninhibited children April 1989 • American Psychologist 669 F i g u r e 1 Relation Between Original Index of Inhibition at 21 Months and Aggregate Index of Inhibition at 7V2 Years for Cohort 1