Spectatoring and the relationship between body image and sexual experience: self-focus or self-valence?

Following the publication of Duval and Wicklund's (1972) landmark book documenting the pervasive impact of self-focus on behavior, psychological research on self-attention and self-regulation processes escalated rapidly (e.g., Carver & Scheier, 1981). The construct of self-attention or self-focus now permeates a vast range of research topics in social and personality psychology (for a review, see Gibbons, 1990) and psychopathology, most notably alcoholism, anxiety-related disorders, and depression (for a review, see Ingram, 1990). In recent years, the construct of self-focused attention has also come to occupy a central role in theory and research on sexual functioning. Masters and Johnson (1970) were the first to articulate the role of self-attentional factors in sexual functioning by describing self-focus as a cognitive distraction associated with al dysfunction. Through a process they termed spectatoring, Masters and Johnson (1970) suggested that focusing on oneself from a third person perspective during sexual activity, rather than focusing on one's sensations and/or sexual partner, can increase performance fears and cause deleterious effects on sexual performance. Based on clinical observation, Kaplan (1974) outlined various forms of distracting thoughts that seem to interfere with sexual potency. In addition to implicating self-focus in the causation of sexual dysfunction, a number of researchers and clinicians have suggested the use of self-focus mechanisms in the treatment of sexual dysfunction. Some such therapeutic interventions involve redirecting the focus of attention from oneself to one's partner, whereas others, such as sensate focus (Masters & Johnson, 1970), involve focusing on and enjoying one's own sensations of being pleasured.

Barlow (1986) formalized the concept of spectatoring in terms of a causal model of attentional processes in sexual functioning. Briefly, he proposed that deficits in sexual functioning associated with inhibited excitement are primarily due to disruptions in the processing of erotic cues required for arousal. These disruptions occur when sexual performance cues activate performance anxieties that, in turn, cause an attentional shift from reward-motivated focus on arousal cues to threat-motivated focus on the likelihood and consequences of failure. Barlow's model closely resembles cognitive interference models of test anxiety (Sarason, 1975; Wine, 1971), social anxiety (Hartman, 1983), and shyness (Cheek & Melchior, 1990), and each may be viewed as a special case of a general, cognitive interference model of anxiety-related performance deficits (see Barlow, 1988).

The concept of spectatoring described by Masters and Johnson (1970), Kaplan (1974), and Barlow (1986) has also been proposed as an explanation of nonclinical sexual phenomena such as differences in level of sexual experience among college students. On the basis of Barlow's (1986) model, Faith and Schare (1993) hypothesized that individuals chronically, negatively focused on their bodily appearance would tend to be sexually avoidant and would therefore demonstrate lower levels of sexual experience than less self-focused individuals. To allow a preliminary test of this hypothesis, they suggested that negative body image could be considered a dispositional analogue of spectatoring and evaluated the correlation between body image and sexual experience in a sample of college students. As predicted, positive body image was related to higher levels of sexual experience, independently of the effects of sexual liberalism-conservatism, sexual knowledge, and global psychological adjustment. The authors considered these findings supportive of theories implicating spectatoring processes in sexually avoidant behavior (e.g., Barlow, 1986; Masters & Johnson, 1970).

These findings are intriguing in that they document a strong association between body image and sexual experience that cannot be explained by differences in sexual attitudes, sexual knowledge, or psychological distress. A spectatoring interpretation of these findings may, however, be somewhat premature in that it rests on several untested but theoretically important assumptions regarding the construct of spectatoring and its, relation to body image. One of these is the assumption that possession of a negative body image implies chronic self-attentiveness. Another is that the relation between body image and spectatoring is linear. The possibility that chronic self-attention is associated with negative body images does not rule out the possibility that chronic self-attention is also associated with positive body images (e.g., narcissistic tendencies). A third assumption accompanying spectatoring explanations of such effects is that they are mediated by attentional processes (e.g., distraction from arousal cues) and not, for example, by motivational ones (e.g., anticipation of reward versus punishment). For example, in the Faith and Schare (1993) study, attentional mediation is clearly implied by the authors' repeated citation of Barlow's (1986) cognitive interference model of sexual dysfunction. According to Barlow (1986), self-attentional processes disrupt the processing of erotic cues when sexual expectancies are sufficiently negative. That is, given negative expectancies (e.g., negative body image), the central mechanism responsible for precipitating sexual avoidance is cognitive interference in the processing of erotic cues. Reference to this model as an explanation of body-image effects on sexual experience implies they are attributable to dysfunctional attentional processes.

Even if it were the case, however, that chronically negative body images were reliably related to chronically high levels of self-focus, a spectatoring explanation of body image effects would still involve a fundamental ambiguity. By definition, body image is primarily a measure of the valence of self-perceptions (i.e., ranging from highly negative to highly positive). If body image scores are used to infer the frequency of self-perceptions, then attentional causal inferences (e.g., chronic distraction from arousal cues) will be necessarily confounded with motivational ones (e.g., social avoidance or withdrawal). For example, if persons with poor body images tend to be especially pessimistic about their sexual prospects, they will be especially unmotivated to pursue social opportunities that might lead to sexual involvement. That pessimism may have nothing to do with how sexual enjoyable or successful previous sexual encounters have been for that person. In this sense, correlations between body image and sexual experience (e.g., Faith & Schare, 1993) may be due merely to impaired social (or sexual) motivation. They may, or may not, also be a consequence of chronic, debilitating self-focus during sexual interactions.

Although Barlow's (1986) model makes explicit the interactive relation between cognitive and affective components of spectatoring, efforts to extend the notion of spectatoring to chronic, individual differences (e.g., Faith & Schare, 1993) may be prone to a serious confusion. The construct of spectatoring entails both focusing on the self and negative self-evaluations and is therefore inherently "double barreled." In this article we empirically re-examine the Faith and Schare (1993) findings and conclusions in an effort to draw the attention of sex researchers to this potential for confusion in discussions of spectatoring.

A large body of self-attention research attests to the fundamental importance of distinguishing attentional processes from motivational ones when interpreting the effects of self-focus on behavior. Research findings across a number of different behavioral domains suggest that the behavioral consequences of directing attention toward the self depend largely on the expectancies, affects, or self-relevant standards that become salient during self-focus (see Carver & Scheier, 1981, 1986a; Scheier & Carver, 1988). Interactive effects of self-focus and outcome expectancies have been observed in studies of test anxiety (Carver, Peterson, Follansbee, & Scheier, 1983), self-esteem (Brockner, 1979), snake phobia (Carver, Blaney, & Scheier, 1979), and social anxiety (Burgio, Merluzzi, & Pryor, 1986). In general, self-focus serves to activate, facilitate, or potentiate the likely behavioral outcome for a given expectancy. The interactive nature of self-attention and expectancies is explicit in Carver and Scheier's (1981, 1986a) general model of behavioral self-regulation. In this model, the concepts of self-focused attention and outcome expectancies both occupy a central role, but expectancies constitute a "psychological watershed" (Scheier & Carver, 1988, p. 323) that determines whether self-focus potentiates behavioral approach or behavioral withdrawal. Barlow's (1986) cognitive interference model of sexual avoidance, which highlights the interactive nature of self-attention and expectancies on sexual behavior, may be interpreted as a special case of Scheier and Carver's (1988) self-regulation model.

Recently, the dependency of attentional effects on expectancies, feelings, and evaluations has prompted sex researchers to question the importance of self-attention in the construct of spectatoring. Researchers have periodically noted that Masters and Johnson's (1970) concepts of spectatoring and sensate focus each refer to a state of self-focused attention, yet spectatoring is assumed to inhibit sexual arousal and sensate focus to facilitate it (e.g., Abrahamson, Barlow, Beck, Sakheim, & Kelly, 1985; Beck, Barlow, & Sakheim, 1983; Sakheim, Barlow, Beck, & Abrahamson, 1984). Because clinical descriptions generally portray spectatoring as a negatively valenced state and sensate focus as a positively valenced one (e.g., Kaplan, 1974; Masters & Johnson, 1970), it has been suggested that if these descriptions are correct, "the relevant parameter is not the self-focus per se, but rather the valence of the focus that causes it to affect arousal" (Sakheim et al., 1984, p. 151). Similarly, Beck et al. (1983) concluded that

it appears the concept of spectatoring

(Masters & Johnson, 1970) may

involve additional components beyond

intensive self-focus and, by

implication, the recommendation

to focus upon one7s partner may

achieve its therapeutic effects by

blocking or otherwise removing

this negative self-fulfilling expectancy

concerning the ability to

become aroused. Similarly, sensate

focus exercises may also operate to

remove the client from this expectancy

cycle. (p. 7)

This would suggest that, except perhaps in the proximal sense intended by Barlow (1986) in his model of male erectile disorder, "intensive self-focus" is not especially causally significant in sexual avoidance and may merely be an epiphenomenon of negative affect associated with negative self-evaluation. For example, researchers in a recent study of the effects of trait self-focus and sexual performance monitoring (diary keeping) on sexual functioning in men undergoing therapy for erectile dysfunction concluded that "neither dispositional social anxiety, nor the generalized tendency to spectator in social contexts is implicated in erectile disorder the findings on self-monitoring [diary keeping], as those on dispositional self-consciousness, imply that spectatoring, per se, is not a significant contributor to erectile dysfunction" (Fichen, Libman, Takefman, & Brender, 1988, pp. 126-127).

An important implication of the findings and conclusions just reviewed is the possibility that the valence of one's body perceptions explains their effect on sexual approach and avoidance, and although self-focus can be sexually debilitating (e.g., Barlow, 1986), it is not a significant factor in the relation between body image and sexual experience. Because frequency of self-attention was not measured in Faith and Schare's (1993) study, the role of self-attention in their findings remains in question. The purpose of the current study, therefore, was to evaluate the role of chronic self-focused attention in the relation between body image and sexual experience. We examined this issue from two perspectives. First, we measured a range of different self-attentive dispositions by means of self-report questionnaires and evaluated their degree of covariation with body image and sexual experience. If negative body images are indicative of chronic spectatoring (Faith & Schare, 1993), one would expect some level of positive association between body image and measures of chronic self-focused attention.

Second, we evaluated whether a trait directly relevant to sexual approach and avoidance, one we expected to be associated with narcissistic tendencies, flirtatiousness, might explain most of the association between body image and sexual experience. For our purposes, we defined this trait by its ordinary language meaning in English as "to pay amorous attention to someone without serious intentions or emotional commitment" (Webster's New World College Dictionary, 1996), although we make no strong claim about whether respondents interpreted the word flirtatious in exactly this way (e.g., in everyday usage we doubt whether flirtation always signifies an absence of "serious attentions or emotional commitment"). We anticipated that flirtatiousness might offer an interesting perspective on the cognitive/ affective ambiguity of the spectatoring construct for the following reason. We have argued that use of a self-evaluative construct such as body image to measure "spectatoring" tendency necessarily confounds attentional interpretations of body image effects with motivational ones. For example, Faith and Schare's (1993) spectatoring explanation of body-image effects on sexual experience may be alternatively explained in affective-motivational terms (e.g., expectancy-mediated sexual motivation). One way to differentiate attentional from affective interpretations in this instance is to identify a trait that, like poor body image, is theoretically linked to above average levels of self-focus, but that is self-evaluatively positive. If such a trait could be shown to mediate most of the body-image correlation with sexual experience, in an inverse direction from body image, this would argue for an affective interpretation and against an attentional (i.e., spectatoring) interpretation of that correlation.

Persons with narcissistic tendencies tend to possess exceptionally positive body images, a sense of entitlement to life's pleasures and rewards and, as a consequence of their high egocentricity and indifference to others, are prone to use dominant and manipulative tac-tics to obtain their preferred social and material rewards (see Emmons, 1987; Raskin & Hall, 1979). One may assume that narcissistic persons will manifest such tendencies in the realm of sexuality, given that sexuality looms large as a primary incentive in adulthood and, also, therefore constitutes a prime resource for social manipulation. Given these considerations, we hypothesized that traits such as flirtatiousness should be empirically associated with a narcissistic personality profile, and, consequently, with highly positive body images, high levels of sexual experience, and high rather than low levels of self-focused/attention (for evidence that narcissistic dispositions are linked to high levels of self-focused attention, see Emmons, 1987). Because such traits are more specifically relevant to sexual approach and avoidance than general measures of either self-esteem (e.g., Rosenberg, 1965) or dispositional narcissism (e.g., Raskin & Hall, 1979), they are more likely to be correlated with sexual behavior than broadly defined measures. From the 30 trait adjectives contained in the Sex Role subscale of the Derogatis Sexual Functioning Inventory (DSFI; Derogatis, 1978), three adjectives were considered most relevant to body image and, sexually relevant narcissism and corresponded most clearly to the following body image-related content of the Narcissistic Personality Inventory (NPI; Raskin & Hall, 1979): "I like to display my body . . . . I am apt to show-off if I get the chance," and "I like to start new fads and fashions." These three DSFI Sex Role adjectives were flirtatious, seductive, and fashionable. A composite of self-ratings on these adjectives provided a simple index of flirtatiousness, affording us a second method of evaluating the recent proposal that poor body image restricts sexual experience via "spectatoring" processes (e.g., Faith & Schare, 1993).

To provide an empirical test of our assumption that flirtatiousness is especially characteristic of persons with narcissistic personality features, we identified a subset of trait scales from the Revised NEO Personality Inventory (NEO-PI-R or NEO; Costa & McCrae, 1992) corresponding in content to the primary factors of the most widely used measure of dispositional narcissism, the NPI (see Emmons, 1987). The NEO Modesty (versus Arrogance) scale (MOD) inversely corresponds to the NPI factors of Superiority ("Everybody likes to hear my stories") and Self-absorption ("I am an extraordinary person"). The NEO Assertiveness scale (ASR) directly corresponds to the Leadership/Authority factor of the NPI (e.g., "I am assertive"). The NEO Straightforwardness (versus Manipulative) scale (STR) inversely corresponds to the NPI Exploitativeness factor ("I find it easy to manipulate people"). We expected self-reported flirtatiousness to demonstrate the following prototypically narcissistic personality profile: arrogant, dominant, and manipulative.

Finally, we hypothesized that the sexually narcissistic trait of flirtatiousness could be shown to mediate most or all of the association between body image and sexual experience. Such a mediation effect would be difficult to explain in terms of cognitive interference associated with self-focus (e.g., spectatoring) if flirtatiousness is more likely associated with high rather than low self-attentiveness. It would imply that positivity versus negativity of the self (i.e., self-valence) may be a more relevant parameter than self-focus in the sexually inhibitive effects of poor body image.

Methods

Participants

Seven hundred twenty-two (437 female, 285 male) students enrolled in introductory psychology courses at the University of British Columbia voluntarily participated in this study in exchange for partial course credit. Ethnic composition of the sample was 51% East or Southeast Asian ancestry (of which approximately 70% were of Chinese ancestry) and 49% Non-Asian (of which approximately 80% were of European ancestry). Asians were defined as those (a) having listed an East or a Southeast Asian language as their first language or (b) having listed an East or a Southeast Asian country as their country of birth or parents' birth. All other persons, including a small number of South Asians (e.g., East Indian, Sri Lankan) and West Asians (e.g., Middle Eastern, Iranian), were classified as Non-Asian. Median age of the sample was 19 years, with a range of 17 to 55, and with 94% of the sample at or between 18 to 25 years of age. Because of either missing gender or ethnicity information, or missing item responses for some questionnaires, the maximum usable sample size was 709 participants (433 females, 276 males).

Measures

All participants were administered the Derogatis Sexual Functioning Inventory (DSFI; Derogatis, 1978), measures of dispositional self-attentiveness, and additional personality measures as part of a larger study of personality and sexuality (Trapnell & Meston, 1996). Descriptions of these measures, and their reliability estimates in the current sample, are presented next.

Sexuality measures. The measures of sexuality used in the current study were the four DSFI subscales analyzed by Faith and Schare (1993): Body Image, Sexual Information, Sexual Attitude, and Sexual Experience. Comprehensive reliability and validity data for these DSFI scales are reported by Derogatis and Melisarotas (1979). The Body Image scale consists of self-ratings on five gender-specific physical attributes (e.g., "Women/men would find my body attractive") and 10 general body attributes (e.g., "My face is attractive"), for 15 ratings, which are summed to provide a single numerical index of level of dissatisfaction with one's physical appearance or body image. Response format is a five-point scale, ranging from 0 (Not at all true of me) to 4 (Extremely true of me). Coefficient Alphas for the DSFI Body Image scale in the current male and female samples were .81 and .84, respectively.

The Sexual Information Scale is scored as the sum of correct responses to 26 sexual information items administered in a true-false format. The items were selected by Derogatis (1978) so that a majority of them are moderately difficult and a lesser number are minimally or maximally difficult. Twelve items are worded in the true direction (e.g., "Usually men achieve orgasm more quickly than women") and 14 in the false direction (e.g., "The penis must be erect before ejaculation may occur"). Coefficient Alpha for the DSFI Sexual Knowledge scale in the current combined sample was .68.

The Sexual Attitude Scale consists of 30 statements thought to reflect liberal versus conservative sexual beliefs. Fifteen statements are worded in a liberal direction (e.g., "Masturbation is a perfectly normal, healthy sexual behavior") and 15 in a conservative direction (e.g., "Oral-genital sex is not within the range of normal sexuality"). Conservative items, which are all substantially negatively correlated with the liberal items, may be reverse scored, and the 30 items may then be summed to form a single numerical index of sexual liberalism. Response format is a five-point scale, ranging from -2 (Strongly Disagree) to +2 (Strongly Agree). Coefficient Alpha for the DSFI Sexual Attitude Scale in the current combined sample was .90.

The DSFI Sexual Experience subscale consists of 24 specific sexual acts representing a broad spectrum of sexual experiences, ranging from the earliest behaviors occurring in the human sexual behavior sequence (e.g., "Kissing on the lips") to more advanced sexual behaviors (e.g., "Mutual oral stimulation of genitals") and various positions of coitus (e.g., "Intercourse--sitting position"). The Sexual Experience Scale total score is the sum of sexual experience items endorsed as "Yes." Coefficient Alpha for the DSFI Sexual Experience Scale in the current combined sample was .96.

Two additional DSFI subscales, Symptoms and Sex Role, were also used in this study. The Symptom Scale is identical to the Brief Symptom Inventory (Derogatis, 1975) and assesses 53 symptoms of psychological distress (e.g., anxiety, depression, somatic complaints) that may be summed to provide a global numerical index of psychiatric symptomatology. Response format is a five-point scale, ranging from 0 (Not at all bothered by this problem/complaint) to 4 (Extremely bothered by this problem/complaint). Coefficient Alpha for the DSFI Symptom Scale in the current combined sample was .96.

Flirtatiousness. Scores for 3 of the 20 trait adjectives comprising the DSFI Sex Role scale, flirtatious, seductive, and fashionable, were summed to provide a composite index of self-perceived flirtatiousness (Flirt). Instructions for the DSFI Sex Role scale state Below is a list of personality characteristics that are often used to describe people. We would like you to describe yourself in terms of these characteristics. To do this, please indicate the degree to which each trait is typical of you--in other words, how much of each characteristic you have. Use the numbered scale given below, and place the appropriate number alongside each trait. Self-ratings on the 20 Sex Role trait adjectives, including the 3 adjectives scored here for the Flirt scale, were made on a conventional 5-point scale administered with the following scale anchors: Not at all (0), A little Bit (1), Moderately (2), Quite A Bit (3), and Extremely (4). Correlations among the 3 Flirt scale items among men and women, respectively, were as follows: seductive with flirtatious, .56, .64; seductive with fashionable, .44, .46; and flirtatious with fashionable, .28, .36. These values resulted in mean item intercorrelations of .43 and .49, and Alpha reliability estimates for the 3item Flirt scale of .69 and .74, among men and women, respectively. A mean item intercorrelation of .40 would correspond to an alpha coefficient of .91 for a 15-item scale. An estimated reliability of .70 for a 3-item scale may be considered sufficient for the current purposes.

Trait self-focus. A relatively comprehensive set of dispositional self-focus measures, which had been administered to only the first 200 respondents of the current sample for the purposes of another study (Trapnell, 1996) were re-analyzed for the current study. (The mean responses to all other measures, i.e., those administered to the total sample of 709 participants, did not differ significantly between these 200 participants and the remaining 509 participants.) These were the Private Self-Consciousness (PRSC), Public Self-Consciousness (PUSC), and Social Anxiety (SANX) scales of the Fenigstein, Scheier, and Buss (1975) Self-Consciousness Scale and the Reflection and Rumination scales of the Reflection-Rumination Questionnaire (RRQ; Trapnell, 1996). The PRSC is a 10-item questionnaire assessing introspective or self-reflective tendencies (e.g., "I reflect about myself a lot"). The PUSC is a seven-item questionnaire measuring anxious preoccupation with one's outward appearance and one's impression on others ("I'm self-conscious about the way I look"). The SANX is a six-item questionnaire measuring self-consciousness associated with social-evaluative anxiety (e.g., "I have trouble working when someone is watching me"). These three brief scales are among the most widely administered measures of dispositional self-focus in the psychological research literature (for reviews, see Buss, 1980; Carver & Scheier, 1981, 1985; Scheier & Carver, 1983).

The Reflection and Rumination Questionnaire (RRQ; Trapnell, 1996) consists of 24 statements that assess positively or negatively motivated forms of private self-consciousness. The 12-item Reflection scale measures private self-consciousness assumed to be motivated by epistemic curiosity (e.g., "I love exploring my inner self"), is correlated highly with measures of intellectual curiosity and openness to experience, an is unrelated to measures of negative emotionality such as trait anxiety and depression (Trapnell, 1996). The 12-item Rumination scale measures private self-consciousness assumed to be motivated by perceived threats or fears about the self ("My attention is often focused on aspects of myself I wish I'd stop thinking about"), is highly correlated with negative emotionality measures such as trait anxiety and depression, and is unrelated to measures of intellectual curiosity and openness to experience. The two scales tend to correlate less than .25 with one another but are both substantially correlated with the Fenigstein et al. (1975) measure of private self-consciousness, especially the larger of its two subfactors. Trapnell (1996) reported mean correlations for Reflection and Rumination, respectively, of .59, .43 with the PRSC total scale, and .53 and .53 with the PRSC's largest subfactor, Self-Reflection. Alpha reliability estimates in the current combined sample for the Reflection scale and the Rumination scale were .91, and .90, respectively.

Additional personality measures. All participants also completed the Revised NEO Personality Inventory (NEO-PI-R; Costa & McCrae, 1992) as part of a broader set of personality and attitude questionnaires completed on a take-home basis for Session 1 of the study (see Procedure). The NEO-PI-R is a 240-item personality questionnaire designed as a comprehensive measure of 5 broad personality factors: neuroticism (NEUR), extraversion (EXTR), agreeableness (AGRE), conscientiousness (CONSC), and openness to experience (OPNS), commonly referred to in personality literature as "the Big Five" factors of personality (Goldberg, 1993; Wiggins & Trapnell, in press). The NEO-PI-R is one of the most widely used and extensively validated personality inventories in current research. It measures each Big Five factor by means of six brief subscales per factor, each representing a content distinction or "facet" thought to be relatively important or fundamental for a particular Big Five dimension (e.g., Assertiveness facet of EXTR). The NEO-PI-R therefore provides assessment of 30 specific traits (for a review of the reliability and validity of the NEO-PI-R scales, see Costa & McCrae, 1992). Scores for the following NEO-PI-R facets were retained for the current analysis: the extraversion facets of Positive Emotions (POS) and Assertiveness (ASR), the agreeableness facets of Modesty (MOD) and Straightforwardness (STR), and the neuroticism facets of Depression (DEP) and Self-consciousness (SLF). The three NEO-PI-R facets relevant to dispositional narcissism, Assertiveness, Modesty, and Straightforwardness, have already been described. Positive Emotions measures the tendency to experience feelings of lightheartedness, joviality, and enthusiasm. Depression assesses the tendency to experience feelings of worthlessness, loneliness, and sadness. Scores for the neuroticism facet of Depression and the extraversion facet of Positive Emotions were selected to represent dispositional negative and positive affectivity, respectively. Self-Consciousness assesses the tendency to experience feelings of shame, embarrassment, apprehension, and inferiority in social situations. The label Self-Consciousness clearly suggests that high scores on this scale should be associated with heightened self-focus in social situations. Like Depression, the Self-Consciousness facet is also clearly a marker of negative self-evaluation (e.g., "I often feel inferior to others"). Alpha estimates of reliability in the current combined sample ranged from.70 (NEO Self-Consciousness) to .83 (NEO Depression) and were highly similar to those reported by Costa and McCrae (1992).

The following instructions appeared on the first page of the trait questionnaires: Please describe yourself as accurately as possible using each statement below. Indicate how much you agree or disagree with each statement by circling one of the scale categories. Answer quickly and truthfully. All personality measures were administered with a conventional five-point scale using scale anchors of strongly disagree (1), disagree (2), neutral (3), agree (4), and strongly agree (5).

Procedure

Personality measures were distributed at the participants' classes, completed at home, and returned to the Individual Differences Laboratory of the Psychology Department when participants arrived for their second session. During the second session, participants completed the DSFI in groups of 5-10 individuals in a large testing room arranged to provide maximum privacy of responding (e.g., visual barriers between participants). A same-gender researcher was present during all sessions to provide instructions and (quietly) to answer questions. Written instructions stated that no names or other identifying information were to be placed on any answer sheets, all answers were anonymous, responses would be used only for scientific research, and accuracy and honesty of responses were extremely important. Matching of the participants' first and second session measures was done anonymously by assigning a common random number to their first and second session envelopes upon their arrival at Session 2. Upon finishing the questionnaires, participants folded their answer sheets, sealed them in unmarked envelopes, and deposited the envelopes through a slot into a sealed cardboard box prior to leaving the study room. No participants withdrew from Session 2 before they had finished the set of questionnaires.

Results

Because mean differences between men and women and between Asian and Non-Asian participants were anticipated for several DSFI scales (e.g., see Meston, Trapnell, & Gorzalka, 1996), participants' scores were standardized within each of the four gender by ethnic (Asian versus Non-Asian) subsamples. The resulting z-scores, in which mean gender and ethnic differences are eliminated, were used for all subsequent analyses. Preliminary moderated multiple regression analyses were also conducted to test for the presence of gender and ethnic differences in associations relevant to our primary hypotheses. Using the combined sample of 709 participants, Sexual Experience scores were first regressed on the six standardized predictors of interest (Age, Sexual Attitudes, Sexual Knowledge, BSI Symptoms, Body Image, and Flirt), after which the cross-product of each predictor with gender, and with ethnicity, was statistically evaluated for a residual association with Sexual Experience. No cross-products involving ethnicity were significant (for all Fs, p [is greater than].20); therefore, data for Asian participants were combined with those of Non-Asian participants in all subsequent analyses. One cross-product involving gender was significant (gender x Flirt, F = 4.1, p [is less than] .04) and a second nearly significant (gender x Sexual Knowledge, F = 3.5, p [is less than] .06); therefore, data for men and women were analyzed and reported separately. Results of these moderated regressions are available from the first author upon request.

Correlations between the dispositional self-focus measures and body image, sexual experience, and flirtatiousness are presented in Table 1. Neither body image nor sexual experience showed meaningful associations with self-reported tendencies toward inquisitive, self-focus (REF), ruminative self-focus (RUM), preoccupation with one's thoughts and feelings in general (PRSC), nor, most surprisingly, preoccupation with one's physical or behavioral appearance to others (PUSC). Among women, poorer body image was significantly positively associated with greater social anxiety but lower levels of reflective self-focus. In the combined sample, poor body image was weakly, positively associated with social anxiety. The trend in the association between flirtatiousness and reflective self-focus was is in a direction opposite to that for body image. Finally, in the combined sample alone, higher levels of sexual experience were associated with lower levels of social anxiety, and higher levels of flirtatiousness were associated with higher levels of public self-consciousness. The latter correlation provides limited support for our assumption that flirtatiousness is positively associated with self-attentiveness. Interestingly, of the four trait measures of self-focus used here, the trait of public self-consciousness scale may be most directly relevant to the concept of spectatoring. Whereas private self-consciousness, reflection, and rumination center upon preoccupation with one's inner thoughts and feelings, public self-consciousness centers upon preoccupation with how one physically or behaviorally appears to others (e.g., one of the items appearing on the PUSC is "I'm self-conscious about the way I look"). Self-described flirtatious persons do seem to be especially concerned about and focused upon their outward appearance to others.

Table 1

Correlation of Sexual Experience, Body Image,
and Flirtatiousness with Dispositional
Measures of Self-focus and Social Anxiety

                        PUSC    PRSC    REF     RUM     SOANX
Men
 DSFI body image          -07     -10    -01     10      21
 DSFI sexual experience   -01     -02    -07    -10     -17
 Flirtatiousness           24     -04     03     06     -14
Women
 DSFI body image           11     -11    -28(*)  17      40(**)
 DSFI sexual experience    11      06    -03    -04     -23
 Flirtatiousness           18      18     24     10     -20
Combined Sample
 DSFI body image           02     -11    -16     14      31(**)
 DSFI sexual experience    05      02    -05    -07     -20(*)
Flirtatiousness            21(*)   08     15     08     -18

Note: Decimals have been omitted. For men, N = 95-97; for women, N = 107-112. Based on scale scores standardized within ethnic (Asian versus Non-Asian) subsamples. Column labels are Public Self-Consciousness (PUSC), Private Self-Consciousness Scale (PRSC), Ruminative Self-focus (RUM), Reflective Self-focus (REF), and Social Anxiety (SOANX).

(*) p < .003 (p < .05/ 15)

(**) p <.001 (p <.01/15)

Correlations presented in Table 2 indicate that, in both men and women, higher flirtatiousness scores were associated with higher levels of positive affect (POS), feelings of arrogance and superiority (MOD), interpersonal manipulativeness (STR), and social dominance (ASR). Among women, higher flirtatiousness scores were also associated with less depression (DEP) and fewer feelings of social embarrassment and inferiority (SLF). Poor body image showed a similar degree of association with these four NEO-PI-R facet scales, but opposite in sign. More positive perceptions of one's sexual attractiveness were associated, in both men and women, with higher levels of positive affect, feelings of arrogance and superiority, interpersonal manipulativeness, and social dominance. Interestingly, neuroticism facet correlations with both flirtatiousness and sexual experience were significant only among men, and extraversion facet correlations with sexual experience were significant only among men.

Table 2

Correlation of Sexual Experience, Body Image,
and Flirtatiousness with Selected Traits Scales from
the Revised NEO Personality Inventory

                                NEUR                   EXTR
                           SLF         DEP       POS       ASR
Men
 Flirtatiousness         -18(*)      -21(**)    32(**)    38(**)
 DSFI poor body image     40(**)      41(**)   -25(**)   -38(**)
 DSFI sexual experience  -21(**)     -24(**)    20*       26(**)
Women
 Flirtatiousness         -11         -09        31(**)    30(**)
 DSFI poor body image     37(**)      43(**)   -23(**)   -29(**)
 DSFI sexual experience  -07         -04        12        13

                                AGRE
                           MOD         STR
Men
 Flirtatiousness         -31(**)     -37(**)
 DSFI poor body image     24(**)      14
 DSFI sexual experience  -22(**)     -15
Women
 Flirtatiousness         -28(**)     -31(**)
 DSFI poor body image     34(**)      15(*)
 DSFI sexual experience  -13         -17(**)

Note: Decimals have been omitted. For men, N 276; for women, N = 433. Based on scale scores standardized within ethnic (Asian versus Non-Asian) subsamples. Column labels are Neuroticism domain (NEUR), Self-Consciousness facet (SLF), Depression facet (DEP), Extraversion domain (EXTR), Positive Emotions facet (POS), Assertiveness facet (ASR), Agreeableness domain (AGRE), Modesty (vs. conceit) facet (MOD), and Straightforwardness (vs. deceitfulness) facet (STR).

(*) p < .003 (p < .05/18)

(**) p < .001 (p <.01/18)

Table 3 presents intercorrelations among the same DSFI variables reported by Faith and Schare (1993) and the Flirtatiousness scale. Correlations for men (presented below the diagonal) were generally similar to those for women (presented above the diagonal). The correlations showing the largest absolute gender differences were the correlations between sexual attitudes and body image (-.31 versus -.08), sexual attitudes and sexual experience (.49 versus .32), and sexual knowledge and sexual experience (.39 versus .21). Two thirds (11/15) of the correlations were larger among women than among men.

Table 3
Zero-Order Correlations Among DSFI Scales and Flirtatiousness

Variable                1          2       3
1. Experience                     49(**)   39(**)
2. Attitudes            32(**)             52(**)
3. Knowledge            21(**)    41(**)
4. BSI symptoms        -14       -12      -14
5. Poor body image     -26(**)   -08      -06
6. Flirtatiousness      46(**)    18(*)    05

Variable                4          5       6
1. Experience          -03       -31(**)   35(**)
2. Attitudes           -06       -31(**)   26(**)
3. Knowledge           -17(**)   -17(**)   13
4. BSI symptoms                   30(**)  -09
5. Poor body image      31(**)            -37(**)
6. Flirtatiousness     -08       -32(**)

Note: Decimals have been omitted. For men, N = 276; for women, N = 433. Based on scale scores standardized by ethnic group (Asian versus Non-Asian). Values in upper diagonals are for women; values in lower diagonals are for men.

(*)p<.003 (p <.05/ 15)

(**) p < .001 (p <.01/15)

Both among men and women, poor body image was negatively correlated with sexual experience and flirtatiousness, and flirtatiousness was positively correlated with sexual experience. These findings introduce the possibility that flirtatiousness, a sexual signaling trait associated with a narcissistic personality pattern (e.g., Table 2) may account for the association between body image and sexual experience. This possibility was empirically evaluated by means of hierarchical regression. Because, in the current sample, age was significantly correlated with sexual experience, r(708) = .12, p [is less than] .001, and with flirtatiousness, r(708) = -.14, p [is less than] .001, age was entered in Step 1 with the three control variables examined by Faith and Schare (1993), Sexual Attitudes, Sexual Knowledge, and BSI Symptoms. Two separate, parallel analyses were run at Step 2. In the first of these (Step 2a), the independent contribution of flirtatiousness to sexual experience was evaluated controlling for age, sexual attitude, sexual knowledge, and psychological distress. In the second (Step 2b), the independent contribution of body image to sexual experience was evaluated controlling for age, sexual attitude, sexual knowledge, and psychological distress. As may be seen in Table 4, flirtatiousness predicted a significant proportion of the variance in sexual experience beyond the effects of age, sexual attitudes, sexual knowledge, and psychological distress, both in men ([F.sub.R2chg] = 68.7, p [is less than] .001) and women ([F.sub.R2chg] = 43.2, p [is less than] .001). Body image also predicted a significant proportion of the variance in sexual experience beyond the effects of age, sexual attitudes, sexual knowledge, and psychological distress, both in men ([F.sub.R2chg] = 16.0, p [is less than] .001) and women ([F.sub.R2chg] = 17.9, p [is less than] .001).

Table 4
Hierarchical Multiple Regression of DSFI Sexual Experience Scale
on DSFI Sexual Attitudes, DSFI Sexual Knowledge, BSI Symptoms,
DSFI Body Image, and Flirtatiousness

                Age       Symptoms      Knowledge       Attitudes
Men
  Step 1        .020(*)   .009          .006            .063(**)*
  Step 2a
  Step 2b
  Step 3
Women
  Step 1        .013**    .001          .028(***)       .111(***)
  Step 2a
  Step 2b
  Step 3

              Flirtatiousness          Body Image       [R.sup.2]
Men
  Step 1         --                       --            .130(***)
  Step 2a       .176(***)                 --            .306(***)
  Step 2b        --                     .049(***)       .179(***)
  Step 3                                .010            .316(***)
Women
  Step 1         --                       --            .272(***)
                                                              ***)
  Step 2a       .067(***)                 --            .339(***)
  Step 2b        --                     .030(***)       .302(***)
  Step 3                                .006(*)         .345(***)

Note: Values in columns 1 through 6 are squared semi-partial correlations. For men, N=276; for women, N=433. Based on scale scores standardized by ethnic group (Asian versus Non-Asian). Column labels are Brief Symptom Index (Symptoms), DSFI Sexual Knowledge Scale (Knowledge), DSFI Sexual Attitude Scale (Attitudes), DSFI Body Image Scale (Body Image), and three-item flirtatiousness index (Flirtatiousness).

(*) p < .05

(**) p < .01

(***) p < .001

Our principal hypothesis was that flirtatiousness would mediate most association between body image and sexual experience. To evaluate this hypothesis, we first evaluated whether body image effects on sexual experience were significantly different from zero when individual differences in flirtatiousness were controlled as well as age, sexual knowledge and attitudes, and psychological adjustment. With flirtatiousness and the four control variables entered in the equation, the semi-partial correlation of body image with sexual experience was no longer significant among men ([F.sub.R2chg] = 3.8, p [is greater than] .05) but remained significant among women ([F.sub.R2chg] = 4.01, p [is less than] .05). More germane to the mediation hypothesis, however, is not whether body image makes any contribution to the prediction of sexual experience over and above flirtatiousness, but how much of the body image effect on sexual experience (at Step 2) may be explained by individual differences in flirtatiousness (i.e., percentage reduction of the squared semi-partial correlation for body image between Steps 2b and 3; see fifth column of Table 4). This proportion was 80.1% for men and 79.0% for women. The magnitude of this mediation effect was statistically evaluated by means of a dependent sample t-test comparing the semi-partial correlation for body image with sexual experience at Step 2b with the corresponding value at Step 3 (see Steiger & Browne, 1984). The extent to which flirtatiousness mediated the body image effect on sexual experience was statistically significant both for men (t = -4.4, p [is less than] .001) and women (t = -7.6, p [is less than] .001).

Discussion

The results of this study replicate a recently reported finding that body image is predictive of sexual experience independently of the effects of sexual conservatism, sexual knowledge, and level of psychological adjustment (Faith & Schare, 1993). Two lines of evidence were presented, however, which cast doubt on a chronic self-focus or "spectatoring" explanation of this finding: correlations of body image and sexual experience with dispositional self-focus measures and mediation of the body image effect on sexual experience by flirtatiousness, a trait associated with higher rather than lower amounts of self-focused attention.

Chronic (i.e., Dispositional) Self-focused Attention

The pattern of association of body image and sexual experience with a range of scales specifically designed to measure self-focusing tendency (Fenigstein et al., 1975; Trapnell, 1996) were not consistent with a spectatoring interpretation of body image-related sexual inexperience. With one exception, these associations were either nonsignificant or in a direction opposite to that assumed by a spectatoring perspective. The more self-reflective women claimed to be, for example, the more positive were their body images. Although not consistent with Faith and Schare's (1993) spectatoring assumption regarding body image, this result is consistent with previous findings that private self-consciousness correlates positively with both self-ratings (Cash, Cash, & Butters, 1983) and observer ratings (Lipson, Przybyla, & Byrne, 1983) of physical attractiveness and findings that sexually dysfunctional men score lower than sexually functional men in private self-consciousness (Fichen et al., 1988) and public self-consciousness (Beck & Barlow, 1986a,b; Fichen et al., 1988). In the current study, public self-consciousness was positively correlated with flirtatiousness, a result that may provide some insight into these somewhat counterintuitive previous findings on sexuality and public self-consciousness.

The one exception to these dispositional self-focus findings was the trait of social anxiety. Social anxiety was significantly associated with poor body image and lower levels of sexual experience. However, Fenigstein et al. (1975) included the construct of social anxiety within their taxonomy of self-attentive dispositions but considered social anxiety an affective outcome of social self-awareness, rather than a self-attentive disposition (e.g., Fenigstein et al., 1975, p. 523). The tendency for persons with negative appraisals of their physical selves also to report negative appraisals of their social selves may be better understood from the vantage point of recent structural models of self-esteem: Physical self-esteem and social self-esteem define separate but highly correlated facets in most current multidimensional models of the self-concept (e.g., Fleming & Courtney, 1984; Marsh & O'Neil, 1984; Openshaw, Thomas, & Rollins, 1981). The association between body image and social anxiety could easily be due to a shared association with poor self-regard rather than chronic self-regarding. This is not to say that self-attentional processes are not important to the construct of social anxiety (e.g., see Cheek & Melchior, 1990). Rather, the definitional association between social anxiety and negative affect necessarily confounds cognitive-attentional interpretations of social anxiety effects with affective-motivational ones, in exactly the same way as does body image.

Flirtatiousness

An internally consistent composite of three traits linked to dispositional narcissism, flirtatious, seductive, and fashionable, significantly predicted the extent of sexual experience among men and women, beyond the effects of age, sexual conservatism, sexual knowledge, and body image. Most importantly, scores on this composite flirtatiousness index explained almost all of the body-image association with sexual experience, both among men and women. To the extent that flirtatiousness tends to be associated with higher rather than lower characteristic levels of self-focused attention, this finding supports the dispositional self-focus findings in casting doubt on a spectatoring explanation of why body image correlates with sexual experience.

Two lines of evidence suggest that flirtatiousness may, indeed, be associated with higher rather than lower characteristic levels of self-focus. Flirtatiousness was found to be weakly but positively correlated with public self-consciousness, the self-attentive disposition most relevant among those measured in the current study to the construct of spectatoring. Second, among both men and women, flirtatiousness was found to be associated with a personality profile that is prototypical of nonpathological narcissism (e.g., Raskin & Hall, 1979): dominant, arrogant, and manipulative. Unlike body image, flirtatiousness was not highly correlated with negative affect, especially among women. Body image and flirtatiousness did, however, show a remarkably similar pattern and degree of association with interpersonal dispositions, only opposite in sign. Not surprisingly, poor body image was associated with a nonnarcissistic personality profile: unassertiveness, feelings of humility and modesty, and guilelessness or lack of manipulativeness and deceit.

Body Image and Sexual Experience

In the current study, body image was not associated with self-reported frequency of observing the self. By definition, however, body image is associated with the valence of those observations (Anderson & LeGrand, 1991). Self-evaluative valence tends to be systematically biased, ranging from moderate positive bias among most individuals (Taylor & Brown, 1988) to" pronounced negative bias among depressed individuals (Beck, 1976) and pronounced positive bias among narcissistic individuals (Gabriel, Critelli, & Ee, 1994). This suggests that the general relation between self-evaluative bias and self-focus is probably curvilinear: Depressives and narcissists, among the most negatively and most positively biased of individuals, respectively, both exhibit characteristically high levels of self-focused attention (e.g., Emmons, 1987; Ingram, 1990). These facts underscore the current findings and suggest that body-image effects on sexual experience may be mediated by factors more closely associated with self-valence (e.g., motivational disengagement) than with self-focus (e.g., attentional distraction from arousal cues during sexual interactions).

If chronic self-focus may be added to the growing list of variables (e.g., age, sexual knowledge, sexual liberalism, and psychological adjustment) that do not explain why body image is predictive of sexual experience, what alternative psychological mechanisms or factors might explain this effect? Negative body image could inhibit sexual approach in a number of ways that do not necessarily involve dysfunctional self-focus. First, self-appraisals of attractiveness are weakly, but positively, correlated with observer ratings of attractiveness (Feingold, 1992). To the extent that self-perceptions parallel social perceptions of attractiveness, persons with poor body images may, in fact, be less physically attractive than others and therefore have less social opportunity to acquire sexual experience. Stelzer, Desmond, and Price (1987) recently reported, for example, that undergraduate women rated by undergraduate male judges as physically attractive were twice as likely to report having experienced sexual intercourse than those whom the raters considered to be average or unattractive. However, agreement between self and observer ratings of attractiveness is extremely weak (see Feingold, 1992), so weak, in fact, that individual differences in actual, or objectively rated, physical attractiveness cannot account for more than a fraction of the effect of body image on sexual experience.

Although body image is remarkably independent of observer judgments of attractiveness, it is strongly correlated with other self-perceptions, particularly self-perceptions of social presence and social skill (e.g., Feingold, 1992; Marsh & O'Neil, 1984). This suggests that body image is much more closely linked to perceived than to actual social opportunities, which, in turn, suggests a second, more probable route of influence between body image and sexual experience. The current finding that flirtatiousness mediates much of the body-image effect on sexual experience could be interpreted as a sociosexual optimism or self-efficacy effect. Individuals confident of their sexual attractiveness tend also to be confident of their social influencability and likeability, expect they will be socially and sexually successful, and, as a consequence, are more likely to initiate or respond favorably to sociosexual. opportunities.

Cognitive-affective Ambiguity in the Spectatoring Construct

In a discussion of the test anxiety literature, Carver and Scheier (1986b) noted that the ruminative state responsible for impairing test performance among test-anxious individuals was given the label self-focus by leading researchers in this area (e.g., Sarason, 1975; Wine, 1971, 1982). However, given the apparently interactive role of self-focus and expectancies on behavior, Carver and Scheier suggested that this characterization may have been a misleading oversimplification. Masters and Johnson's (1970) choice of the label spectatoring for the dual cognitive and affective processes underlying male erectile disorder may be similarly misleading in that it may encourage tendencies to equate these two processes.

Faith and Schare's (1993) equation of body image valence with spectatoring exemplifies, we believe, this potential conceptual pitfall of the spectatoring construct. For example, conceptual slippage between cognitive and affective meanings of spectatoring is evident in the following statement of their primary hypothesis: "based on Barlow's (1986) theoretical model, subjects who maintain negative conceptualizations about their bodies are expected to be more sexually avoidant than subjects who do not focus on negative aspects of their bodies" (p.350; italics added). This wording implies an equivalency between having a negative opinion about one's appearance and chronically ruminating about it. Certainly, however, it is possible to maintain positive or negative beliefs that one does not routinely dwell on. For example, one may believe one isn't especially physically attractive and prefer not to think too much about it. One may alternatively believe one is exceedingly attractive and frequently like to ponder this reassuring self-perception. By conjoining valence and frequency of self-focus, the concept of spectatoring discourages separate consideration of the effects of self-focusing and the valence of that activity on sexual approach and avoidance. Failure to operationalize attentional and affective components of spectatoring separately naturally precludes empirical evaluation of their relative importance or the manner in which they interact to influence sexual behavior.

Conclusions that may be drawn from this study should be tempered with the usual cautions regarding interpretation of self-reports. Findings reported here for self-reports of chronic self-attention may bear little relation to findings that might be obtained using objective, non-intrusive, behavioral indicators of self-focused attention. The present data merely illustrate a potential problem with the practice of drawing self-attentional inferences from self-appraisal measures like body image (Faith & Schare, 1993). Future researchers on how self-focus affects sexual approach and avoidance should attempt to address two methodological problems in the measurement of self-focused attention: (a) how to quantify individual differences in self-focus independently of motivational confounds such as negative affect or intrinsic interest and (b) how to quantify individual differences in self-focus quickly, objectively, and unobtrusively.

An additional limitation of the current study is its use of the trait flirtatiousness to signify a sexually relevant form of self-consciousness, one with a clearly positive self-valence. Although evidence substantiating this interpretation of flirtatiousness was provided, more direct measures of sexual self-focus and sexual self-valence are clearly called for in future research on this issue. Some narcissistic dispositions may, however, prove useful for examining the interplay of affect and attention in sexual approach and avoidance. Individuals scoring high on dispositional narcissism provide an unusual but theoretically valuable comparison group for such research in that they are markedly self-focused individuals who possess remarkably positive self-valence.

References

Abrahamson, D. J., Barlow, D. H., Beck, J. G., Sakheim, D. K., & Kelly, J. R (1985). The effects of attentional focus and partner responsiveness on sexual responding: Replication and extension. Archives of Sexual Behavior, 14, 361-371.

Anderson, B. L., & LeGrand, J. (1991). Body image for women: Conceptualization, assessment, and a test of its importance to sexual dysfunction and medical illness. The Journal of Sex Research, 28, 457-477.

Barlow, D. H. (1986). Causes of sexual dysfunction: The role of anxiety and cognitive interference. Journal of Consulting and Clinical Psychology, 54, 140-148.

Barlow, D. H. (1988). Anxiety and its disorders. New York: Guilford.

Beck, A. T (1976). Cognitive therapy and the emotional disorders. New York: International Universities Press.

Beck, J. G., & Barlow, D. H. (1986a). The effects of anxiety and attentional focus on sexual responding - I: Physiological patterns in erectile dysfunction. Behaviour Research and Therapy, 24, 9-17.

Beck, J. G., & Barlow, D. H. (1986b). The effects of anxiety and attentional focus on sexual responding - II: Cognitive and affective patterns in erectile dysfunction. Behaviour Research and Therapy, 24, 19-26.

Beck, J. G., Barlow, D. H., & Sakheim, D. K. (1983). The effects of attentional focus and partner arousal on sexual responding in functional and dysfunctional men. Behaviour Research and Therapy, 21, 1-8.

Brockner, J. (1979). The effects of self-esteem, success-failure, and self-consciousness on task performance. Journal of Personality and Social Psychology, 37, 1732-1741.

Burgio, K. L., Merluzzi, T. V, & Pryor, J. B. (1986). Effects of performance expectancy and self-focused attention on social interaction. Journal of Personality and Social Psychology, 50, 1216-1221.

Buss, A. H. (1980). Self-consciousness and social anxiety. San Francisco: Freeman.

Carver, C. S., Blaney, R H., & Scheier, M. F. (1979). Focus of attention, chronic expectancy, and responses to a feared stimulus. Journal of Personality and Social Psychology, 37, 1186-1195.

Carver, C. S., Peterson, L. M., Follansbee, D. J., & Scheier, M. F. (1983). Effects of self-directed attention on performance and persistence among persons high and low in test anxiety. Cognitive Therapy and Research, 7, 333-354.

Carver, C. S., & Scheier, M. F. (1981). Attention and self-regulation: A control-theory approach to human behavior. New York: Springer-Verlag.

Carver, C. S., & Scheier, M. F. (1985). Aspects of self and the control of behavior. In B. R. Schlenker (Ed.), The self and social life (pp. 146-174). New York: McGraw-Hill.

Carver, C. S., & Scheier, M. F. (1986a). Functional and dysfunctional responses to anxiety: The interaction between expectancies and self-focused attention. In R. Schwarzer (Ed.), Self-related cognitions and motivation (pp. 111-141). Hillsdale, NJ: Erlbaum.

Carver, C. S., & Scheier, M. F. (1986b). A control process perspective on anxiety. Anxiety Research, 1, 17-22.

Cash, T. F., Cash, D. W., & Butters, J. W (1983). "Mirror, mirror, on the wall ...?" Contrast effect and self-evaluations of physical attractiveness. Personality and Social Psychology Bulletin, 9, 351-358.

Cheek, J., & Melchior, L. (1990). Shyness, self-esteem, and self-consciousness. In H. Leitenberg (Ed.), Handbook of social and evaluation anxiety (pp. 47-82). New York: Plenum.

Costa, P. T., Jr., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO-PI-R) and NEO Five-Factor Inventory (NEO-FFI) professional manual. Odessa, FL: Psychological Assessment Resources.

Derogatis, L. R. (1975). The brief symptom inventory. Baltimore, MD: Clinical Psychometrics Research.

Derogatis, L. R. (1978). Derogatis sexual functioning inventory. Rev. ed. Baltimore, MD: Clinical Psychometrics Research.

Derogatis, L. R., & Melisaratos, N. (1979). The DSFII A Multidimensional Measure of Sexual Functioning. Journal of Sex and Marital Therapy, 5, 244-281.

Duval, S., & Wicklund, R. A. (1972). A theory of objective self-awareness. New York: Academic Press.

Emmons, R. A. (1987). Narcissism: Theory and measurement. Journal of Personality and Social Psychology, 52, 11-17.

Faith, M. S., & Schare, M. L. (1993). The role of body image in sexually avoidant behavior. Archives of Sexual Behavior, 22, 345-356.

Feingold, A. (1992). Good-looking people are not what we think. Psychological Bulletin, 111, 304-341.

Fenigstein, A., Scheier, M. F., & Buss, A. (1975). Public and private self-consciousness: Assessment and theory. Journal of Consulting and Clinical Psychology, 43, 522-527.

Fichen, C. S., Libman, E., Takefman, J., & Brender, W. (1988). Self-monitoring and self-focus in erectile dysfunction. Journal of Sex and Marital Therapy, 14, 120-128.

Fleming, J. S., & Courtney, B. E. (1984). The dimensionality of self-esteem: Hierarchical facet model for revised measurement scales. Journal of Personality and Social Psychology, 46, 404-421.

Gabriel, M. T., Critelli, J. W., & Ee, J. S. (1994). Narcissistic illusions in self-evaluations of intelligence and attractiveness. Journal of Personality, 62, 143-155.

Gibbons, F. X. (1990). Self-attention and behavior: A review and theoretical update. In M. P. Zanna (Ed.), Advances in experimental social psychology, 23 (pp. 249-295). New York: Academic Press.

Goldberg, L. R. (1993). The structure of phenotypic personality traits. American Psychologist, 48, 26-34.

Hartman, L. M. (1983). A metacognitive model of social anxiety: Implications for treatment. Clinical Psychology Review, 3, 435-456.

Ingram, R. (1990). Self-focused attention in clinical disorders: Review and a conceptual model. Psychological Bulletin, 107, 156-176.

Kaplan, H. S. (1974). The new sex therapy: Active treatment of sexual dysfunctions. New York: Brunner/Mazel.

Lipson, A. L., Przybyla, D. P. J., & Byrne, D. (1983). Physical attractiveness, self-awareness, and mirror-gazing behavior. Bulletin of the Psychonomic Society, 21, 115-116.

Marsh, H. W., & O'Neil, R. (1984). Self-description questionnaire III: The construct validity of multidimensional self-concept ratings by late adolescents. Journal of Educational Measurement, 21, 153-174.

Masters, W. H., & Johnson, V. E. (1970). Human sexual inadequacy. Boston: Little, Brown.

Meston, C. M., Trapnell, P. D., & Gorzalka, B. B. (1996). Ethnic and gender differences in sexuality: Variations in sexual behavior between Asian and Non Asian university students. Archives of Sexual Behavior, 25, 33-72.

Openshaw, D. K, Thomas, D. L., & Rollins, B. C. (1981). Adolescent self-esteem: A multidimensional perspective. Journal of Early Adolescence, 1, 273-282.

Raskin, R. A., & Hall, C. S. (1979). A narcissistic personality inventory. Psychological Reports, 45, 590.

Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.

Sakheim, D. K., Barlow, D. H., Beck, J. G., & Abrahamson, D. J. (1984). The effect of increased awareness of erectile cues on sexual arousal. Behaviour Research and Therapy, 22, 151-158.

Sarason, I. G. (1975). Anxiety and self-preoccupation. In I. G. Sarason & C. D. Spielberger (Eds.), Stress and anxiety (pp. 27-44). New York: Wiley.

Scheier, M. F., & Carver, C. S. (1983). Two sides of the self: One for you and one for me. In J. A. Suls & A. Greenwald (Eds.), Psychological perspectives on the self (Vol. 2) (pp. 123-157). Hillsdale, NJ: Erlbaum.

Scheier, M. F., & Carver, C. S. (1988). A model of behavioral self-regulation: Translating intention into action. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 21) (pp. 303-346). New York: Academic Press.

Steiger, J. H., & Browne, M. W. (1984). The comparison of interdependent correlations between optimal linear composites. Psychometrika, 49, 11-21.

Stelzer, C., Desmond, S. M., & Price, J. H. (1987). Physical attractiveness and sexual activity of college students. Psychological Reports, 60, 567-573.

Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social-psychological perspective on mental health. Journal of Personality and Social Psychology, 103, 193-210.

Trapnell, P. D. (1996). Self-consciousness and the five-factor model of personality: Distinguishing rumination from reflection. Doctoral dissertation, University of British Columbia.

Trapnell, P. D., & Meston, C. M. (1996). Sex and the five factor model of personality. Paper presented at the annual meeting of the American Psychological Association, Toronto, Ontario, Canada.

Webster's New World College Dictionary (3rd ed.). (1996). V. Neufeldt & D. B. Guralnik (Eds.). New York: Simon and Schuster.

Wiggins, J. S., & Trapnell, P. D. (in press). Personality structure: The return of the Big Five. In R. Hogan, J. A. Johnson, & S. R. Briggs (Eds.), Handbook of personality psychology. Orlando, FL: Academic Press.

Wine, J. D. (1971). Test anxiety and direction of attention. Psychological Bulletin, 76, 92104.

Wine, J. D. (1982). Evaluation anxiety: A cognitive-attentional construct. In H. W. Krohne & L. C. Laux (Eds.), Achievement, stress, and anxiety (pp. 207-219). Washington DC: Hemisphere.

Manuscript accepted October 28, 1996

This research was supported in part by a Social Sciences and Humanities Research Council of Canada Postdoctoral Scholarship to Paul D. Trapnell, Ph.D., a Social Sciences and Humanities Research Council of Canada Postdoctoral Scholarship to Cindy M. Meston, Ph.D., and a British Columbia Health Research Foundation Studentship Grant to Cindy M. Meston, Ph.D.

We gratefully acknowledge Lisa Billett and Diane Fredrickson for their assistance in the administration of the study.

Address reprint requests to Paul D. Trapnell, Ph.D., Department of Psychology, University of British Columbia, Vancouver, British Columbia, V6T 1Z4, Canada.