Abstract

Section 1

Section 2

Section 3

Section 4

Section 5

Discussion
Board

Shyness fulfills the two key criteria in the definition of a temperament: "inherited personality traits present in early childhood" (Buss & Plomin, 1984, p. 84). Therefore it may seem puzzling that Kagan (1994) has concluded that "most adults who say they are shy do not belong to the temperamental category favoring this quality" (p. 42). The solution to this puzzle may depend upon the timing of when shyness first becomes a salient characteristic for a particular individual. Baldwin (1894) described a developmental distinction between "primary" or "organic" bashfulness, the shyness seen in infants, young children, and animals, and "true" bashfulness, the kind of shyness seen in humans only after age three, "which shows reflection in its simpler form, upon self and the actions of self [and] represents the child's direct application of what he knows of persons to his own inner life" (p. 439). McDougall (1963) extended Baldwin's analysis to explain the development of individual differences in the trait shyness, and he suggested a third stage of development in which the intensification of self-consciousness at the onset of puberty interacts with the development of the self-regarding sentiment to shape shyness and modesty as qualities of adult character and conduct (see Table 1).

More recently, Buss (1980, 1986) has proposed a distinction between early- developing, fearful shyness and later-developing, self-conscious shyness that is framed in the language of contemporary research on temperament and personality development. The fearful type of shyness typically emerges during the first year of life and is influenced by temperamental qualities of wariness and emotionality that include a substantial genetic component (Kagan & Reznick, 1986; Plomin & Rowe, 1979). Because the effects of these temperamental factors precede the development of a cognitive self-concept, Buss specifically excluded low self-esteem as a potential cause of early-developing shyness. In light of attachment theory (Bowlby, 1988), however, it might be better to conceptualize a concurrent transactional development of temperament and the "working model" of emotional self-esteem during early childhood.

Buss's self-conscious type of shyness first appears around age 4 or 5 when the cognitive self has already begun to develop, becomes more intense around age 8 as social comparison processes become more salient in self-evaluation (Harter, 1986), and peaks between 14 and 17 as adolescents cope with cognitive egocentrism (the "imaginary audience" phenomenon) and identity issues (Cheek, Carpentieri, Smith, Rierdan, & Koff, 1986; Cheek & Melchior, 1990). The peak of adolescent self-consciousness is significantly higher for females than for males, at least in American society (Elkind & Bowen, 1979; Simmons & Rosenberg, 1975; Zimbardo, 1977). In contrast to the fearfulness and somatic anxiety that characterizes early-developing shyness, later-developing shyness involves cognitive symptoms of psychic anxiety such as painful self-consciousness and anxious self-preoccupation (e.g., Crozier, 1979; Ishiyama, 1984).

Because adolescent self-consciousness declines significantly after age 14 or 15, whereas the influence of inherited temperament should be more stable, it seems reasonable that shyness would be more of an enduring characteristic for people who were first shy in early childhood than for those who were first shy in later childhood or early adolescence. As part of our research on subtypes of shyness, we asked 590 students whether they considered themselves currently shy, and, if not, whether there had been some previous period in their life during which they had considered themselves to be a shy person. The 48% of respondents who were currently shy and the 38% who were previously shy identified the age range in which they first remember being shy, and their answers are presented in Table 2 (which replicates in a larger sample the results in Table 1 of Cheek et al., 1986).

Our new data are quite consistent with previous surveys employing retrospective reports of college students that have revealed four findings relevant to Buss's conceptualization: 1) about 40% of currently shy respondents indicated that they had been shy since early childhood; 2) early-developing shyness is more enduring, with about 75% of those who said they were shy in early childhood reporting still being shy currently, but only about 50% of those who were first shy during late childhood or early adolescence saying that they are currently shy; 3) the early-developing shy respondents also had developed cognitive symptoms of shyness upon entering adolescence, so that they differed from those with later-developing shyness by having more somatic anxiety symptoms but did not have fewer cognitive symptoms; and 4) early-developing shyness appears to be more of an adjustment problem, with males in that group reporting the most behavioral symptoms of shyness (Bruch, Giordano, & Pearl, 1986; Cheek, et al., 1986; Shedlack, 1987).

According to recent estimates, approximately 40 to 50% of American college students consider themselves to be shy (Carducci & Zimbardo, 1997). Given the large number of shy people, it is reasonable to expect that there is substantial heterogeneity among them, not only in terms of age of onset and shyness symptoms, but also in terms of social motivation. Indeed, previous research has demonstrated that there is a wide range of individual differences among shy college students in sociability (Cheek & Buss, 1981), dependency (Das, 1991), and social avoidance (Pilkonis, 1977). It is possible that there might be several distinct subtypes of shyness that could be based on individual differences in preferred or habitual interpersonal styles. An identification of these subtypes would help to resolve a long-standing controversy in the shyness literature: the relative emphasis on withdrawal versus dependency in the characterization of shy people's interpersonal styles.

The withdrawal model equates shyness with inhibition, reticence, and social avoidance. For example, Caspi, Elder, and Bem (1988) characterized life-course patterns of shy children in terms of Horney's (1945) interpersonal style "moving away from the world" and used the terms "shy" and "withdrawn" interchangeably (p. 824). We see at least two problems with this emphasis. First, not all socially withdrawn children are shy; some may be simply unsociable, and others may have been actively rejected by their peers (Harrist, Zaia, Bates, Dodge, & Pettit, 1997). Secondly, not all shy children are socially withdrawn; in fact, Conn (1941), another child psychologist who was influenced by Horney, emphasized dependency and conformity to the standards of others in his report of play-interviews with shy children.

In the literature on adolescent and adult shyness, Lewinsky (1941) emphasized a dependent interpersonal style of "going along to get along" that many shy people adopt in social situations, and some research on the social behavior of shy people has supported the interpretation that a protective self-presentational style is a predominant characteristic of shyness (Leary & Kowalski, 1995; Meleshko & Alden, 1993). In laboratory experiments, shy people have been shown to change their views to please an authority figure and to conform with a majority opinion (e.g., Santee & Maslach, 1982; Turner, 1977). Such an overly compliant interpersonal style resembles Horney's (1945) "moving toward people" type, rather than her "moving away" type which was invoked by Caspi et al. (1988). According to Horney, the tendency to move toward people represents a self-effacing, "compulsive compliance" solution of basic insecurity, while the tendency to move away from people represents a self-sufficient, "compulsive detachment" solution of insecurity.

It is plausible that both of these conceptualizations of shyness are partially true, in that each of them describes a subtype of shyness. Following Horney (1945), shy people have in common feelings of insecurity and neurotic conflict, but they might choose different solutions in trying to adapt to or ameliorate their negative feelings. Some shy people withdraw from social interactions, choosing the compulsive detachment solution. Others become excessively dependent on other people, choosing the compulsive compliance solution. Yet other shy people oscillate between excessive withdrawal and excessive dependency.

In a new set of studies, Cheek and Krasnoperova (in press) used the Personal Styles Inventory II (Robins et al., 1994) to explore withdrawn and dependent subtypes of shyness. The shy group was divided into four subgroups based on their scores on the 12-item Defensive Separation subscale of the PSI Autonomy scale (which was used as an index of Horney's "moving away from people") and a weighted sum of the 7-item Dependency and the 10-item Pleasing Others subscales of the PSI Sociotropy scale (which was used as an index of "moving toward people"). Shy participants were divided using the shy group's median for each of these two measures: "Withdrawn" shy group -- high on "away" index and low on "toward" index; "Dependent" shy group -- low on "away" index and high on "toward" index; "Conflicted" shy group -- high on both; and "Secure" shy group -- low on both.

The four-group subtype of shyness approach appears to be superior on both empirical and theoretical grounds to Cheek and Buss's (1981) two-group classification of shy individuals into shy-sociable and shy-unsociable. First, it permits identification of a relatively better-adjusted, more psychologically secure shy group (cf. Aron & Aron, 1997; Gough & Thorne, 1986). Second, using measures of both dependency and withdrawal allows a more precise classification of what used to be called the shy-sociable and the shy- unsociable groups. Third, the four-group approach separates out the conflicted shy group, whose members are high on both dependency and withdrawal. Cheek and Krasnoperova (in press) suggested that this group may present more difficult and complex treatment issues, because conflicted shy people need help in two areas simultaneously: how not to be too passively dependent on others, and how not to be too disconnected from others. We conclude that moving beyond the simple dichotomy of shy versus not shy to examine individual differences among shy people is likely to become increasingly important for both research and treatment.

Back to the top.

Back to the top.

Table 1

Ages and Stages in the Development of Shyness

											
 
Age                    Stage

0       temperamental "organic" bashfulness (Baldwin, 1894)

        early developing shyness (Buss, 1980; Kagan & Reznick, 1986)

4       "true" bashfulness with simpler form of self-reflection
        (Baldwin, 1894)

        first appearance of self-conscious shyness (Buss, 1986)

8       pre-puberty intensification of self-consciousness 
        (McDougall, 1963)

        increased use of social comparison in self-evaluation 
        (Harter, 1986)

14      peak of adolescent self-consciousness, higher for females 
        than males
        (Elkind & Bowen, 1979; Simmons & Rosenberg, 1975; 
        Zimbardo, 1977)

________________________________________________________________________ 

Back to the top.

Table 2

Frequency Distribution for Self-Reports of the Time of First Shyness

												

                                    Currently shy	Previously shy
                                    respondents		respondents		

            Years of late developing self-conscious shyness

                                                                                
On entering college                      16               5

On entering high school                  30              35

On entering junior high school           52              51

During later elementary school years     31              37

On entering elementary school            29              45

     Subtotal                           158             173
___________________________________________________________________
												

		Years of early developing fearful shyness

Before starting elementary school        16              27

As long as I can remember               112              21

    Subtotal                            128              48		
__________________________________________________________________

Totals                                  286             221

__________________________________________________________________
												

	Chi Square = 29.19, p < .01.

Back to the top.

Back to the top.