Role of sexual self-disclosure in the sexual satisfaction of long-term heterosexual couples.

Partners can and do differ from one another in their sexual preferences (Davies, Katz, & Jackson, 1999; McCarthy & Bodner, 2005; Miller & Byers, 2004). Of interest to sex therapists and researchers alike is how couples, despite these differences, maximize and maintain their sexual satisfaction within a long-term relationship. It is unlikely that individuals develop and maintain a highly sexually satisfying relationship by choosing a partner who happens to share exactly the same sexual likes and dislikes. Nor is it likely that an individual's ideal or preferred sexual script remains constant over the course of a long-term relationship (Laumann & Gagnon, 1995). A frequently hypothesized, but less frequently studied, explanation is that couples develop and maintain sexual satisfaction in long-term relationships through communication in general and SSD in particular (Cupach & Metts, 1991; MacNeil & Byers, 1997, 2005; Purnine & Carey, 1997).

Studies have shown that greater SSD is associated with higher sexual satisfaction in long-term relationships (Cupach & Comstock, 1990; MacNeil & Byers, 1997, 2005). There has been little research, however, investigating the mechanisms by which SSD results in higher sexual satisfaction. Cupach and Metts (1991) have theorized that sexual communication impacts sexual satisfaction in two distinct ways. First, disclosure of specific sexual likes and dislikes is a means of informing and obtaining more of what one wants and less of what one does not want sexually from one's partner, thereby resulting in greater sexual satisfaction. Second, self-disclosure about sexual preferences enhances intimacy and relationship satisfaction. In turn, greater relationship satisfaction contributes to greater sexual satisfaction. The present study labeled these two pathways the instrumental pathway and the expressive pathway and they are presented in Figures 1 and 2, respectively. The proposed pathways integrate a dyadic approach to studying self-disclosure and sexual satisfaction with script theory and the empirically supported interpersonal model of sexual satisfaction (IEMSS; Byers & MacNeil, 2006; Lawrance & Byers, 1995). Although theoretically these pathways apply equally to same-sex and opposite-sex couples, because few same-sex couples volunteered, this study was limited to heterosexual couples.

[FIGURE 1 OMITTED]

The Instrumental Pathway

Sexual script theory proposes that sexual interactions are scripted creating a common sequence of behaviors based on social influence and culture, but also varying between individuals (Gagnon & Simon, 1973; Laumann & Gagnon, 1995). Gagnon, Rosen, and Leiblum (1982) distinguish between individuals' cognitive scripts (e.g., plans and fantasies for sexual interactions) and their performative scripts (that is, actual sequences of behavior). The present study makes three assumptions based on sexual script theory: (1) an individual's preferred cognitive script can differ from the performative script he or she enacts in sexual interactions with a partner; (2) differences between and within partners' cognitive and performative scripts are likely to affect their individual sexual satisfaction; and (3) sexual exchanges (that is, sexual rewards and sexual costs as described by the IEMSS) can be thought of as discrete elements of a sexual script about which partners can communicate.

[FIGURE 2 OMITTED]

Cupach and Metts (1991) suggested that disclosure of specific sexual likes and dislikes is a means of informing and obtaining more of what one wants and less of what one does not want sexually from one's partner (i.e., minimizing differences between the cognitive and the performative scripts), thereby resulting in greater sexual satisfaction. By disclosing sexual preferences, couples are able to reconcile their disparate preferences, maximizing sexual rewards and minimizing sexual costs for both partners, resulting in a more favorable balance of sexual rewards to sexual costs; however, this will only occur if partners understand these disclosed preferences and respond to them (Laing, Phillipson, & Lee, 1966; Rosen & Leiblum, 1988). According to the instrumental pathway, SSD leads to greater partner understanding of sexual rewards and costs. Greater partner understanding of rewards and costs, in turn, leads to more rewarding and less costly sexual exchanges and to greater sexual satisfaction.

The instrumental pathway has never been fully tested among individuals or couples in long-term relationships, although there is empirical support for some of the proposed paths. In keeping with social exchange theory, research based on the IEMSS has found individual sexual satisfaction is greater to the extent that both individuals and their partners experience high sexual rewards and low sexual costs (Byers & MacNeil, 2006; Lawrance & Byers, 1995). Researchers have shown that there is a positive association between sexual satisfaction and both SSD and the balance of sexual rewards to costs (Byers & MacNeil, 2006; Lawrance & Byers, 1995). Further, MacNeil and Byers (1997) found that greater disclosure of sexual likes and dislikes to a sexual partner was related to greater sexual satisfaction even after controlling for the extent of NSD, suggesting that sexual satisfaction may be enhanced by sharing of sexual likes and dislikes specifically, and not just self-disclosure in general. Supporting this view, couples seeking sex therapy are significantly poorer than are sexually satisfied couples in the extent and quality of their self-disclosure in general and their self-disclosure about sexual preferences in particular (Chesney, Blakeney, Cole, & Chan, 1981; Hoch, Safir, Peres, & Shepher, 1981). Although the IEMSS proposes that there is a dyadic component to sexual satisfaction, these studies did not assess partner sexual understanding--that is, a process by which sexual communication could come to be associated with sexual satisfaction.

By surveying heterosexual couples, Purnine and Carey (1997) found that partner understanding of sexual likes was associated with sexual satisfaction for men (although not for women). They interpreted their findings as indicating that partner understanding was a result of effective sexual communication. However, they did not measure SSD directly. It may be that partner understanding is a result of increased empathy or sensitivity over time rather than communication per se. If this is the case, empathy by a partner may contribute directly to a more positive affective evaluation of the overall sexual relationship (i.e., sexual satisfaction) rather than affecting sexual satisfaction indirectly by creating a more favorable balance of sexual rewards to costs as proposed by the instrumental pathway. Further, although Purnine and Carey examined understanding of sexual likes, they did not assess understanding of sexual dislikes.

MacNeil and Byers (2005) tested the instrumental pathways with dating couples. They found support for the instrumental pathway for both men and women. That is, greater sexual self-disclosure was associated with greater partner understanding of sexual rewards (but not sexual costs), a more favorable balance of sexual rewards to sexual costs, and greater sexual satisfaction for both couple members. Further, partner understanding of sexual rewards and the balance of sexual rewards to costs were found to fully mediate the relationship between SSD and sexual satisfaction for men and to partially mediate this relationship for women.

There are some reasons to think that understanding of sexual costs is more strongly associated with sexual satisfaction in long-term rather than in short-term relationships. Whereas in short-term relationships sexual satisfaction is strongly linked to global appraisals (Byers, 1999), it is likely that it becomes more difficult for couples to maintain sexual satisfaction over the long term without taking steps to minimize their sexual costs. Indeed, research has shown that sexual costs have a stronger association with sexual satisfaction in long-term relationships than in dating relationships (r= -.77 and r= -.37, respectively; Lawrance & Byers, 1995; Byers & Demmons, 1999). Thus, it is likely that both understanding of a partner's sexual rewards and understanding of their sexual costs will contribute to the instrumental pathway in long-term relationships.

The Expressive Pathway

The second proposed pathway offered by Cupach and Metts (1991) linking sexual self-disclosure to sexual satisfaction emphasizes the role of affect. According to the expressive pathway, greater sexual and NSD are associated with greater sexual and nonsexual self-disclosure by one's partner. Mutual sexual and NSD are proposed to contribute to greater intimacy, which in turn contributes to greater sexual satisfaction.

The primary difficulty in testing Cupach and Metts's (1991) suggestion that intimacy mediates the association between self-disclosure and sexual satisfaction is that researchers have used several correlated constructs to operationally define intimacy. For example, mutual self-disclosure, mutual partner understanding, emotional closeness, and relationship satisfaction all have been described as defining features of intimacy (Clark & Reis, 1988; Perlman & Fehr, 1987). We adopted the view that sexual self-disclosure is an intentional behavior that enhances intimacy, whereas relationship satisfaction can be thought of as an affective response or evaluation of intimacy. In order not to confound the subjective evaluation of the global relationship with elements of self-disclosure or partner empathy or understanding, we used a measure of global relationship satisfaction to assess intimacy rather than using a measure of dyadic adjustment (Lawrance & Byers, 1998).

There is some support for the relationships proposed in the expressive pathway for couples in long-term relationships. Previous research has found significant relationships between relationship satisfaction and self-disclosure in general (Hendrick, 1981; Litzinger & Gordon, 2005), sexual self-disclosure specifically (Banmen & Vogel, 1985), and sexual satisfaction (Byers, 2005; Cupach & Comstock, 1990; Haavio-Mannila & Kontula, 1997). In addition, there is some evidence that mutual NSD contributes to both relationship satisfaction and sexual satisfaction (Larson, Anderson, Holman, & Niemann, 1998; Markman, Renick, Floyd, Stanley, & Clements, 1993). Further, MacNeil and Byers (1997) found that both SSD and NSD contributed uniquely and significantly to the prediction of sexual satisfaction among individuals in long-term relationships.

MacNeil and Byers (2005) tested the expressive pathway with couples in dating relationships. They found relationship satisfaction mediated the association between own SSD and sexual satisfaction among the women and the association between own NSD and sexual satisfaction among the men. They did not find evidence, however, for the importance of mutual SSD or NSD to the sexual satisfaction of the men or women.

There is reason to believe that there would be more support for the expressive pathway involving mutual sexual and NSD for couples in long-term relationships than was found for daters. According to social penetration theory, intimacy increases as a relationship develops (Altman & Taylor, 1973). Several studies have found self-disclosure to be positively associated with longer or more stable relationships and greater partner self-disclosure or involvement (Hendrick, 1981; Meeks, Hendrick, & Hendrick, 1998; Rubin, Hill, Peplau, & Dunkel-Schetter, 1980). A greater extent, depth, and personal nature of information disclosed are considered indicative of greater intimacy (Clark & Reis, 1988). If so, it is likely that the expressive or relational role of SSD is more important for men in long-term relationships than for men in dating relationships and, as with women, affects their sexual satisfaction through the expressive route as well as through the instrumental route. Thus, in keeping with their roles in maintaining intimacy, we expected that mutual sexual and NSD would be important to the sexual satisfaction of both men and women in long-term relationships.

Hypotheses

We used path analysis to test the instrumental and affective pathways. Path analysis is a statistical procedure that can be used to draw inferences about the causal order of a set of variables based on correlational data, in this case a possible causal order of key variables linking self-disclosure to sexual satisfaction (Klem, 1995). The unidirectional arrows in Figures 1 and 2 represent the causal direction or "influence" of one variable on another.

Based on the instrumental pathway and the criteria for establishing mediation (Baron & Kenny, 1986), we made the following predictions: (1) greater SSD would be associated with greater partner understanding of sexual rewards and costs, a more favorable balance of sexual exchanges (rewards and costs), and greater sexual satisfaction; (2) greater understanding of sexual rewards and costs would be associated with a more favorable balance of sexual exchanges and greater sexual satisfaction; (3) a more favorable balance of sexual exchanges would be associated with greater sexual satisfaction; (4) partner understanding would mediate the relationship between SSD and the balance of rewards and costs and (5) the balance of rewards and costs would mediate the relationship between partner understanding and sexual satisfaction.

Based on the expressive pathway, we made the following predictions: (1) greater SSD would be associated with greater NSD and greater SSD and NSD by the partner; (2) greater mutual self-disclosure would be associated with greater sexual satisfaction and greater relationship satisfaction; (3) greater relationship satisfaction would be associated with greater sexual satisfaction; and, (4) relationship satisfaction would mediate the relationship between mutual self-disclosure and sexual satisfaction.

Method

Participants and Procedure

Alumni, faculty, staff, and students at a midsize Canadian university as well as community members in a small Canadian city and their partners were recruited to participate in a study of sexual behavior in long-term romantic relationships. In order to receive a questionnaire package, interested individuals were asked to send their name and address in confidence to an e-mail address or to a messaging system via a toll-free number. One hundred nineteen (47%) of the 253 couples who requested a questionnaire package returned completed questionnaires. Twelve of these couples were dropped from the sample in order to increase its homogeneity (six couples in same-sex relationships, four couples living outside of Canada, and two couples in which one partner had a serious medical illness); three couples were removed due to excessive missing data. The final sample consisted of 104 couples in opposite-sex relationships. On average, the couples had been in their relationship for 14.5 years (SD = 10.1). The male partners ranged in age from 24 to 68 (M=40.1, SD = 10.5); the female partners ranged in age from 22 to 63 (M=38.4, SD = 9.9). Most (84%) were married.

Each questionnaire package included a description of the study, two questionnaire booklets (one for each partner), and two smaller "privacy" envelopes (one for each partner's completed booklet) to be returned together in a larger preaddressed stamped envelope. The cover letter stressed anonymity and confidentiality and instructed participants not to discuss the questionnaire or their responses with their partner.

Measures

The measures used to test the instrumental and expressive pathways are described below. The Sexual Self-Disclosure Questionnaire, Nonsexual Self-Disclosure Questionnaire, Sexual Rewards and Costs Checklist, Sexual Exchanges Questionnaire, and Global Measure of Sexual Satisfaction were used to test the instrumental pathway. The Sexual Self-Disclosure Questionnaire, Nonsexual Self-Disclosure Questionnaire, Global Measure of Relationship Satisfaction, and Global Measure of Sexual Satisfaction were used to test the expressive pathway.

A background questionnaire was used to collect demographic information such as age, sexual orientation, length and type of romantic relationship, and types of sexual activities.

The Nonsexual Self-Disclosure Questionnaire was adapted from the self-disclosure questionnaire developed by Rubin et al. (1980) and was used to assess NSD between partners. This measure consists of 17 potential topics of self-disclosure, including relationships with parents, self-concept, attitudes, and day-today activities. Three items related to SSD were omitted to eliminate overlap with the measure of SSD. Respondents indicated the extent of their self-disclosure for each item on a 7-point scale ranging from nothing at all (1) to everything (7). Scores range from 14 to 98, with higher scores indicating greater self-disclosure. Using a 4-point response format, Byers and Demmons (1999) found this scale to have high internal consistency and to be correlated with other measures associated with disclosure and intimacy including SSD, relationship satisfaction, and length of relationship. Internal consistency was high in the present study ([alpha] = .94 for men and .89 for women).

The Sexual Self-Disclosure Questionnaire (Byers & Demmons, 1999) was used to assess the extent of SSD between partners. Six of 12 items assess how much participants have told their partners about what they like about a range of sexual activities they have engaged in as a couple (e.g., kissing, touching, intercourse); the other six items assess how much they have told their partner about what they do not like about these sexual activities. Participants indicated the extent of their self-disclosure on the same 7-point scale as for the Nonsexual Self-Disclosure Questionnaire such that scores range from 12 to 84. For six couples who reported not engaging in oral sex on the demographic questionnaire, the items relating to oral sex were dropped and their scale score was prorated based on the remaining eight items. Byers and Demmons reported high internal consistency for daters for this measure using a 4-point response scale. Internal consistency for participants who completed all 12 items was also high in the present study ([alpha] = .94 for men and .92 for women).

The Global Measure of Relationship Satisfaction (GMREL; Lawrance & Byers, 1998) was used to assess satisfaction with the overall relationship. In response to the question, "In general, how would you describe your overall relationship with your partner?" respondents rate their overall relationship on five 7-point bipolar scales: good-bad, pleasant-unpleasant, positive-negative, satisfying-unsatisfying, and valuable-worthless. Possible scores range from 5 to 35, with high scores indicating greater relationship satisfaction. Byers and her colleagues (Byers & MacNeil, 2006; Lawrance & Byers, 1998) provided evidence that the GMREL is significantly correlated with other measures of relationship satisfaction and has high internal consistency and test-retest reliability. Internal consistency was high in the current study ([alpha] = .94 for men and .94 for women.)

The Global Measure of Sexual Satisfaction (GMSEX; Lawrance & Byers, 1998) assesses satisfaction with the sexual relationship in response to the question, "In general, how would you describe your sexual relationship with your partner?" using the same scales as for the GMREL. The GMSEX has been shown to be significantly correlated with other measures of sexual satisfaction, and have high internal consistency and test-retest reliability (Lawrance & Byers, 1998; Byers & MacNeil, 2006). Internal consistency in the present sample was high ([alpha] = .95 for the men and .95 for the women).

The 46-item Sexual Rewards and Costs Checklist (Lawrance & Byers, 1998) was used to assess partner understanding of sexual rewards and costs. Each item represents a sexual exchange (e.g., level of affection expressed during sexual activity, oral sex) that respondents could rate as a reward, a cost, both a reward and a cost, or neither a reward nor a cost in their sexual relationship. This measure has been found to be correlated with other measures of sexual satisfaction and to have acceptable test-retest reliability given that sexual rewards and costs are expected to vary over time (Lawrance & Byers, 1998). In the present study, partners completed the checklist both from their own perspective and in terms of their perception of whether each item represented a reward or a cost for their partner. Two items were omitted due to difficulty or redundancy of rewording these items from one participant's perspective to the perception of the participant's partner. To calculate men's understanding of rewards ([UND.sub.REW]), a man's ratings of rewards for his partner was compared with the female partner's ratings of rewards for herself. The number of matching responses was divided by the total number of responses made by both partners. The same procedure was used to calculate [UND.sub.REW] for women and understanding of costs ([UND.sub.CST]) for men and for women.

Participants completed the two items from the Sexual Exchanges Questionnaire (Lawrance & Byers, 1998). They indicated how rewarding (REW) and costly (CST) they considered their sexual relationship with their partner to be on separate 9-point scales ranging from Not at all rewarding [costly] (1) to Extremely rewarding [costly] (9). The difference between these two ratings (REW-CST) was used as a measure of the balance of sexual exchanges in the relationship. Possible REW-CST scores ranged from 8 to-8, with higher scores indicating a more favorable balance of sexual exchanges. REW-CST has been found to predict sexual satisfaction in long-term relationships (Byers & MacNeil, 2006; Lawrance & Byers, 1995).

Data Analysis

The instrumental and expressive pathways were tested separately for women and men. We used the procedures proposed by Baron and Kenny (1986) to test the path models and establish mediation. A mediator variable is a variable that acts as a mechanism by which one variable affects or contributes to another. Baron and Kenny describe three conditions that are required to establish mediation: (1) the independent variable accounts for a significant proportion of the variance in the mediator; (2) the independent variable accounts for a significant proportion of the variance in the dependent variable; and (3) the mediator accounts for a significant proportion of the variance in the dependent variable when the independent variable is controlled. If these conditions are met, there is evidence for mediation if the relationship between the independent variable and the dependent variable is reduced (partial mediation) or becomes zero (full mediation) after controlling for the mediator. The Sobel test assesses whether the indirect effect (i.e., through the mediator) of the independent variable on the dependent variable was significantly different from zero. This test was used to confirm whether the mediation effects were statistically significant.

Results

On average, male and female participants reported high relationship (M=30.8, SD=4.1, and M=30.8, SD=4.9, respectively) and high sexual satisfaction (M=29.1, SD=5.4, and M=29.9, SD= 5.2, respectively). The zero order correlations between all the variables in the instrumental and expressive pathways are presented in Table 1. As predicted, greater SSD was associated with greater sexual satisfaction for both men and women. Further, with the exception of the correlation between women's SSD and men's sexual satisfaction, all of the variables in the two pathways were significantly correlated with sexual satisfaction as predicted. Table 1 also presents the correspondence or similarity between the ratings made by the male and female partners. Individuals high in understanding of sexual rewards and costs, balance of rewards and costs, relationship satisfaction, and sexual satisfaction tended to have partners who were also high in these attributes. The extent of the men's and the women's self-disclosure, however, were not closely related: the men's and women's SSD were not significantly correlated; and the men's and women's NSD, although significantly correlated, shared only 6% of their variance.

We examined whether understanding of sexual rewards and sexual costs differed for the men/or the women using a 2 (gender) x 2 (likes/dislikes) ANOVA. Neither the main effect for gender nor the interaction were significant, F(1,103)= 0.54 and 0.77, p > .05. Both the men and the women, however, had greater understanding of their partner's rewards than of their partner's costs, F(1, 103)=259.03, p=000, [[eta].sup.2] = .72. On average, participants understood 62% of their partner's sexual rewards but only 26% of their partner's sexual costs.

In order to determine whether differential understanding of sexual rewards and costs reflected differences in self-disclosure, we compared self-disclosure about the six items on the SSD questionnaire assessing sexual likes with self-disclosure about the six items assessing sexual dislikes. The main effects for likes/dislikes and for gender were significant, F(1, 103) = 5.03, p=.027, [[eta].sup.2]=.054 and F(1, 103)=86.80, p=.000, [[eta].sup.]=.46, respectively. Both the men and the women reported self-disclosing more about their sexual likes than about their sexual dislikes. The main effect for gender, however, was qualified by a significant interaction, F(1,103) = 19.32, p = .000, n = .16. Although the women and men did not differ in the amount they reported disclosing about their sexual likes (M's=61.8 and 61.2, respectively), the women reported disclosing more than the men did about their sexual dislikes (M's = 57.1 and 47.6, respectively).

Testing the Instrumental Pathway

Women. The significant pathways for women are depicted in Figure 3a. Examination of the zero-order correlations (see Table 1) indicated that, as predicted, women's SSD was significantly associated with their balance of sexual rewards to costs (REW-CST) and sexual satisfaction as well as with their partner's understanding of their sexual rewards ([UND.sub.REW]) but not with their partner's understanding of their sexual costs ([UND.sub.csT]). The associations between [UND.sub.REW], REW-CST, and sexual satisfaction also were significant. Contrary to predictions, however, women who had partners with higher [UND.sub.CST] reported lower REW-CST and sexual satisfaction. (1) Therefore, [UND.sub.CST] was dropped from the planned analyses. Finally, the relationship between men's [UND.sub.REW] and women's REW-CST remained significant when SSD was controlled ([beta] =.36, p < .001). Similarly, the relationship between women's REWCST and GMSEX remained significant when men's [UND.sub.REW] ([beta] = .77, p < .001) was controlled.

[FIGURE 3 OMITTED]

These results satisfy the conditions necessary to test whether partner understanding of the woman's rewards mediated the relationship between women's SSD and their balance of rewards and costs, as well as whether women's REW-CST mediated the relationship between the male partner's [UND.sub.REW] and women's sexual satisfaction. First, after controlling for the male partner's [UND.sub.REW], the relationship between women's SSD and their REW-CST was reduced from [beta] =.18 to [beta] =.12 and was no longer significant (Sobel test = l.78, p =.08). This suggests that partner understanding of rewards mediates the relationship between women's SSD and their balance of rewards and costs. Second, after controlling for women's SSD (the first or exogenous variable in the instrumental pathway) and REW-CST, the association between men's [UND.sub.REW] and women's sexual satisfaction was reduced from [beta].=.38 to [beta].=.08 and was no longer significant (Sobel test=3.79, p< .001). This suggests that women's REW-CST mediates the relationship between the male partner's [UND.sub.REW] and women's sexual satisfaction. In addition, after controlling for men's [UND.sub.REW] and women's REW-CST, the contribution of women's SSD to their sexual satisfaction was reduced from [beta] =.18 to [beta]=.03 and was no longer significant. Together, these results provide support for the instrumental pathway for women (with the exception of [UND.sub.CST]).

Men. The same steps were used to test the instrumental pathway for the men; the significant paths are depicted in Figure 3b. Examination of the zero-order correlations (see Table 1) indicated that, as predicted, men's SSD was significantly associated with their REW-CST and sexual satisfaction as well as with their partner's [UND.sub.REW] but not with their partner's [UND.sub.CST]. The associations between [UND.sub.REW] and REW-CST and sexual satisfaction were also significant. Contrary to predictions and similar to the findings for the women, however, men who had partners with higher [UND.sub.CST] reported lower REW-CST and sexual satisfaction. (2) Finally, the relationship between women's [UND.sub.REW] and men's REW-CST remained significant when SSD was controlled ([beta] = .48, p < .001). Similarly, the relationship between men's REW-CST and their GMSEX remained significant when women's [UND.sub.REW] ([beta] = .71, p < .001) was controlled.

These results satisfy the conditions necessary to test whether partner understanding of the man's rewards mediated the relationship between men's SSD and their balance of rewards and costs, as well as whether men's REW-CST mediated the relationship between the female partner's [UND.sub.REW] and men's sexual satisfaction. First, after controlling for the female partner's [UND.sub.REW], the relationship between men's SSD and their REW-CST was reduced from [beta] =.33 to [beta] =.16 and was no longer significant (Sobel test=3.19, p < .001). This suggests that partner understanding of men's sexual rewards mediated the relationship between men's SSD and their balance of rewards and costs. Second, after controlling for men's SSD and REW-CST, we found that the association between women's [UND.sub.REW] and men's sexual satisfaction was reduced substantially from [beta] = .57 to [beta]. = .14, p=.024 (Sobel test = 6.62, p < .001). This suggests that men's REW-CST partially mediated the relationship between the female partner's [UND.sub.REW] and men's sexual satisfaction. In addition, after controlling for women's [UND.sub.REW] and men's REW-CST, we noted that the contribution of men's SSD to their sexual satisfaction was further reduced from [beta] = .42 to [beta] = .14, p= .020. These results provide support for the instrumental pathway for men. Together, however, partner [UND.sub.REW] and men's REW-CST only partially accounted for the association between men's SSD and their sexual satisfaction.

Testing the Expressive Pathway

Women. In keeping with predictions based on the expressive pathway, women who reported higher relationship satisfaction reported higher sexual satisfaction (see Table 1). To test the other conditions for establishing mediation, we first regressed relationship satisfaction on the mutual self-disclosure set (men's NSD and SSD and women's NSD and SSD). The self-disclosure set was significantly related to women's relationship satisfaction, [R.sup.2]=.20, F(4, 99)=6.30, p = .000. Although all four self-disclosure variables were associated with women's relationship satisfaction on a bivariate basis, only women's NSD made a unique contribution to the equation. Therefore, men's and women's SSD and men's NSD were dropped from further analysis of the expressive pathway for women. Women's NSD was significantly associated with their sexual satisfaction (see Table 1), and women's GMREL was significantly associated with their GMSEX after controlling for NSD ([beta] =.73, p < .001), establishing the final conditions for testing mediation.

The significant paths for women for the expressive pathway are depicted in Figure 4a. After controlling for relationship satisfaction, the association between women's NSD and their sexual satisfaction was reduced from [beta] = .22 to [beta] = -.06 and was no longer significant (Sobel test = 3.71, p < .001). This indicates that relationship satisfaction mediated the relationship between women's NSD and their sexual satisfaction.

[FIGURE 4 OMITTED]

Men. The same steps were used to test the expressive pathway for men. In keeping with predictions based on the expressive pathway, men who reported higher relationship satisfaction reported higher sexual satisfaction (see Table 1). Regression of relationship satisfaction on the mutual self-disclosure set was significant, [R.sup.2]=.20, F(4, 99)=6.30, p=.000. Although men's NSD was associated with men's relationship satisfaction on a bivariate basis, only women's NSD and men's SSD made unique contributions to the equation ([beta] = .26, p = .009 and [beta] = .28, p = .009, respectively). Regression of women's NSD and men's SSD on men's sexual satisfaction indicated that they each made unique contributions to the equation ([beta]=.19, p=.038 and [beta]=.41, p = 000). In addition, men's GMREL was significantly associated with their GMSEX after controlling for men's SSD and women's NSD ([beta] = .66, p < .001), establishing the final conditions for testing mediation.

The significant paths for men are depicted in Figure 4b. After controlling for relationship satisfaction, the association between men's SSD and their sexual satisfaction, although still significant, was reduced substantially from [beta]=.41 to [beta]=.19, p=.009 (Sobel test=3.38, p < .001). The relationship between women's NSD and men's sexual satisfaction was reduced from [beta]= .19 to .03 and was no longer significant (Sobel test= 1.85, p=.06). This indicates that relationship satisfaction partially mediated the relationship between men's SSD and their sexual satisfaction and fully mediated the relationship between women's NSD and men's sexual satisfaction.

Discussion

Consistent with previous research with dating and long-term couples (Byers & Demmons, 1999; Herold & Way, 1988; MacNeil & Byers, 1997, 2005), both male and female partners who reported self-disclosing more about their sexual likes and dislikes also reported greater sexual satisfaction. We tested two possible mechanisms to account for this association: an instrumental pathway and an expressive pathway. Our results suggest that both pathways are important for the sexual satisfaction of men. We also found some evidence for the instrumental pathway between SSD and sexual satisfaction for women. For women, however, the expressive pathway was between NSD and sexual satisfaction. Further, SSD and sexual satisfaction shared 18% of their variance for men but only 3% of their variance for women. Similarly, men's SSD accounted for 14% of the variance in women's understanding, whereas women's SSD only accounted for 3% of the variance in men's understanding of their sexual rewards. Thus, SSD does not appear to be as effective a strategy for enhancing partner understanding and sexual satisfaction for women as it is for men. This may be because our measure of SSD assessed only verbal SSD. Although we found that the women and the men did not differ in the extent to which they had shared their sexual likes and dislikes with their partner, it may be that men are more likely than are women to adopt verbal SSD as an intentional or instrumental behavior in sexual situations, be it to achieve more rewarding and less costly sexual exchanges or to generate greater intimacy with their partners (Clark & Reis, 1988). It is also possible that, for women, nonverbal communication plays an important role in enhancing their partner's understanding and their own sexual satisfaction. Alternately, it may be that women are better than men at learning from their partner's self-disclosure and thus partner verbal self-disclosure enhances women's understanding more than it does men's understanding. Future research needs to take a more comprehensive approach to assessing how partners share their sexual likes and dislikes as well as the extent to which partners benefit from this type of communication.

The Instrumental Pathway

Both sex therapists and sexuality researchers have argued that better communication accounts for why some couples maintain their satisfaction with their sexual relationship despite changes in sexual desires, differences between partner's ideal sexual scripts, or experiencing sexual difficulties (Byers, 1999; Chesney et al., 1981; Meston & Trapnell, 2005). That is, couples' SSD helps them to negotiate and renegotiate mutually satisfying sexual scripts. The instrumental pathway provides a model of this mechanism. As predicted, we found support for the contention that, for couples in long-term relationships, greater SSD contributes to greater sexual satisfaction by enhancing partner understanding of sexual rewards (although not sexual costs), which, in turn, contributes to a more favorable balance of sexual rewards to costs and ultimately to greater sexual satisfaction. Thus, SSD may be an intentional or instrumental strategy to enhance one's sexual relationship by informing one's partner about one's sexual preferences in established relationships just as it is in new relationships. This extends previous research that investigated the relationship between partner understanding of sexual preferences and sexual satisfaction in long-term relationships but failed to assess the extent of SSD (Purnine & Carey, 1997). Whereas, however, Purnine and Carey found that understanding of the partner's rewards was associated with men's sexual satisfaction, we found that it also was associated with women's sexual satisfaction. Further, our results challenge their inference that partner understanding is the result of effective verbal communication alone, especially for women. These findings also demonstrate the importance of using a dyadic approach (i.e., assessing both partners) in studying the development and maintenance of sexual satisfaction.

Consistent with previous research on daters (Byers & Demmons, 1999; MacNeil & Byers, 2005), the results suggest that it is self-disclosure of sexual rewards that is important for couples seeking to maintain or enhance their sexual relationship over the long term. In contrast, greater SSD did not appear to enhance partner understanding of sexual costs. Further, contrary to predictions, partner understanding of sexual costs was associated with a less favorable balance of sexual rewards to sexual costs and lower sexual and relationship satisfaction. As the couples in the present study reported high sexual satisfaction, many sexual rewards, and few sexual costs, it may be that they experienced too few sexual costs for self-disclosure to influence partner understanding of costs. In fact, both the men and the women had disclosed significantly more about their sexual likes than about their sexual dislikes and, in turn, partners understood significantly more of their partner's sexual rewards than of their sexual costs. Focusing more on sexual rewards than on sexual costs may be adaptive or characteristic of satisfying relationships. Alternately, communication about and understanding of negative aspects of the sexual relationship may follow a different path from that of positive aspects of the relationship. Indeed, women who understood more of their partner's sexual rewards understood significantly fewer of their partner's sexual costs and vice versa. Therefore, it is possible that partners become aware of sexual costs only in the context of low sexual rewards, high sexual costs, and low sexual satisfaction. This may be because, for the most part, partners do not disclose sexual dislikes unless their sexual costs are high relative to their sexual rewards. Negotiating negative aspects of the sexual relationship to achieve greater satisfaction may be more difficult or more complicated than negotiating for sexual likes or rewards. Research with less satisfied or sexually distressed couples may help to clarify the role of SSD and partner understanding of sexual costs in sexual satisfaction.

The Expressive Pathway

Less support was found for the expressive pathway that proposes that greater mutual self-disclosure (sexual and nonsexual) contributes to greater relationship satisfaction, which in turn contributes to greater sexual satisfaction. First, we did not find evidence for mutual SSD. That is, the extent of women's and men's SSD was not significantly related, suggesting that SSD is not reciprocated. In contrast, there was evidence that partners reciprocate NSD to at least some degree. Further, it appears that individuals' self-disclosure style generalizes across sexual and nonsexual issues in that both men and women who self-disclosed more about sexual preferences also self-disclosed more about nonsexual issues. We did find evidence for reciprocity in sexual understanding, particularly of sexual rewards, however. That is, individuals who understood their partner's sexual rewards tended to have partners who also understood their sexual rewards; similarly, individuals who understood their partner's sexual costs tended to have partners who understood their sexual costs. This suggests that there may be other mechanisms by which partners come to understand each other's sexual rewards and costs in addition to verbal self-disclosure assessed in the current study, such as nonverbal communication and body language, or how the actual sexual script gets enacted over time and across situations. It may be, for example, that couples discover some of their partner's sexual preferences as a result of spontaneous changes, situational demands, or experimentation in their sexual interactions rather than through SSD. This may be particularly true with respect to men's understanding since women's SSD accounted for less than 4% of the variance in men's understanding of their sexual rewards.

Second, we found evidence for an expressive pathway between SSD and sexual satisfaction for men but not for women. That is, men's relationship satisfaction partially mediated the association between their SSD and their sexual satisfaction. In contrast, MacNeil and Byers (2005) found support for an expressive pathway between SSD and sexual satisfaction for women but not for men in dating relationships. This suggests that, in addition to its instrumental role, in long-term relationships SSD plays an expressive role in contributing to men's sexual satisfaction by enhancing their relationship satisfaction. Further, the men and women were equally accurate in their understanding of their partner's sexual rewards and costs, suggesting that men are not primarily focused on achieving their own optimal balance of sexual rewards to costs. These results are in keeping with research that has shown that young men ideally want to be more expressive in sexual situations than they currently are (Lawrance, Taylor, & Byers, 1996). Our results suggest that in long-term relationships it is their own and not their partners' sharing of sexual preferences that contributes to men's feelings of intimacy and relationship satisfaction and hence to greater sexual satisfaction.

We found an expressive pathway from NSD to sexual satisfaction for both the men and the women. That is, we found that relationship satisfaction fully mediated the relationship between the women's NSD and both men's and women's sexual satisfaction. It may be that the extent to which women self-disclose sets the context for mutual relationship satisfaction and sexual satisfaction. The men and women tended to reciprocate NSD, however, and both men's and women's NSD were associated with sexual satisfaction for both partners on a bivariate basis. Thus, it is more likely that it is being in a generally disclosing relationship rather than women's NSD per se that contributes to sexual satisfaction in both partners by enhancing relationship satisfaction. This interpretation is consistent with past research that has shown that mutual NSD is associated with sexual satisfaction (Larson et al., 1998; Litzinger & Gordon, 2005; Markman et al., 1993). It appears that for women in long-term relationships, verbal SSD does not have an expressive role in enhancing their sexual satisfaction over and above being in a disclosing and intimate relationship in general.

Limitations and Conclusions

The results must be interpreted in light of some of the limitations of the study. First, the sample consisted of highly educated, relationally and sexually satisfied couples. It may be that self-disclosure has a more important function in enhancing both the balance of sexual rewards and costs, intimacy, and sexual satisfaction among couples experiencing lower sexual satisfaction than those without such difficulties (Chesney et al., 1981; Hoch et al., 1981). Second, although a strength of the study is that we recruited a community sample, participants may have differed in key ways from nonresponders. Research has shown that volunteers for sex surveys tend to have more permissive sexual attitudes and engage in more frequent and a broader range of sexual activity (Wiederman, 1999). Although we sent reminder letters to all those who requested questionnaire packages in order to maximize response rate and increase the representativeness of the sample, only 47% completed and returned the questionnaires. Third, although path analysis is an accepted way to infer or suggest causal relationships (Baron & Kenny, 1986), a longitudinal or quasi-experimental design would provide a stronger test of the instrumental and expressive pathways by allowing researchers to see the effects of self-disclosure of sexual rewards and costs in action: when they occur, how they are used, how self-disclosure changes over the length of a relationship, and its effects on the sexual relationship. Further, path analysis cannot rule out other possible causal models. For example, given the strong association between relationship satisfaction and sexual satisfaction, it is possible that sexual self-disclosure results in increased sexual satisfaction, which in turn enhances relationship satisfaction or that there is a reciprocal relationship between relationship satisfaction and sexual satisfaction, rather than through the expressive pathway proposed by Cupach and Metts (1991). That is, the direction of the relationship between relationship satisfaction and sexual satisfaction is not known. Indeed, in a longitudinal study, Byers (2005) found that although relationship satisfaction and sexual satisfaction changed concurrently, there was only limited evidence to support a causal relationship between the two constructs in either direction. However, (nonsexual) communication accounted for some of the concurrent change in both types of satisfaction. Fourth, although we instructed partners to complete the questionnaire separately and provided individual privacy envelopes, we cannot be sure that they did complete the questionnaires separately.

Nonetheless, the results of the present study suggest that partner understanding of rewards is a significant contributor to sexual satisfaction for both men and women because it helps couples establish and maintain a mutually pleasurable sexual script. For men, sexual self-disclosure appears to play a role in both increasing partner understanding of their sexual likes and in enhancing intimacy and relationship satisfaction, leading to greater sexual satisfaction. Our results suggest, however, that (verbal) SSD is not as effective a strategy for enhancing partner understanding, relationship satisfaction, or sexual satisfaction for women as it is for men. Future research will need to investigate further what factors other than SSD enhance partner understanding of sexual likes and dislikes and sexual satisfaction, particularly for women. Certainly there is room to enhance partner understanding of sexual preferences in that these couples, who had been together for an average of 14.5 years, understood less than two-thirds of their partner's sexual rewards and only about one-quarter of their partner's sexual costs.

References

Altman, I. & Taylor, D. A. (1973). Social penetration: The development of interpersonal relationships. New York: Holt, Rinehart, & Winston.

Banmen, J. & Vogel, N. A. (1985). The relationship between marital quality and interpersonal sexual self-disclosure. Family Therapy, 12, 45-58.

Baron, R. M. & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychology research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173 1182.

Byers, E. S. (1999). The interpersonal exchange model of sexual satisfaction: Implications for sex therapy with couples. Canadian Journal of Counselling, 33, 95-111.

Byers, E. S. (2005). Relationship satisfaction and sexual satisfaction: A longitudinal study of individuals in long-term relationships. Journal of Sex Research, 42, 113-118.

Byers, E. S. & Demmons, S. (1999). Sexual satisfaction and sexual self-disclosure within dating relationships. Journal of Sex Research, 36, 1-10.

Byers, E. S. & MacNeil, S. (2006). Further validation of the interpersonal exchange model of sexual satisfaction. Journal of Sex and Marital Therapy, 32, 53 69.

Chesney, A. P., Blakeney, P. E., Cole, C. M., & Chan, F. A. (1981). A comparison of couples who have sought sex therapy with couples who have not. Journal of Sex and Marital Therapy, 7, 131 140.

Clark, M. S. & Reis, H. T. (1988). Interpersonal processes in close relationships. Annual Review of Psychology, 39, 609-672.

Cupach, W. R. & Comstock, J. (1990). Satisfaction with sexual communication in marriage: Links to sexual satisfaction and dyadic adjustment. Journal of Social and Personal Relationships, 7, 179 186.

Cupach, W. R. & Metts, S. (1991). Sexuality and communication in close relationships. In K. McKinney & S. Sprecher (Eds.), Sexuality in close relationships (pp. 93 110). Hillsdale, NJ: Lawrence Erlbaum.

Davies, S., Katz, J., & Jackson, J. L. (1999). Sexual desire discrepancies: Effects on sexual and relationship satisfaction in heterosexual dating couples. Archives of Sexual Behavior, 28, 553 567.

Gagnon, J. H., & Simon, W. (1973). Sexual conduct. The social sources of human sexuality. Chicago: Aldine.

Gagnon, J. J. Rosen, R. C., & Leiblum, S. R. (1982). Cognitive and social aspects of sexual dysfunction: Sexual scripts in sex therapy. Journal of Sex and Marital Therapy, 8, 44-56.

Haavio-Mannila, E. & Kontula, O. (1997). Correlates of increased sexual satisfaction. Archives of Sexual Behavior, 26, 399419.

Hendrick, S. (1981). Self-disclosure and marital satisfaction. Journal of Personality and Social Psychology, 40, 1150-1159.

Herold, E. S. & Way, L. (1988). Sexual self-disclosure among university women. Journal of Sex Research, 24, 1-14.

Hoch, Z., Safir, M. P., Peres, Y., & Shepher, J. (1981). An evaluation of sexual performance: Comparison between sexually dysfunctional and functional couples. Journal of Sex and Marital Therapy, 7, 195-206.

Klem, L. (1995). Path analysis. In L. G. Grimm & P. R. Yarnold Eds.), Reading and understanding multivariate statistics pp. 65-97). Washington, DC: American Psychological Association.

Laing, R. D., Phillipson, H., & Lee, A. R. (1966). Interpersonal perception: A theory and a method of research. New York: Springer.

Larson, J. H., Anderson, S. M., Holman, T. B., & Niemann, B. K. (1998). A longitudinal study of the effects of premarital self-disclosure, relationship stability, and self-esteem on sexual satisfaction in the first year of marriage. Journal of Sex and Marital Therapy, 24, 193-206.

Laumann, E. O. & Gagnon, J. H. (1995). A sociological perspective on sexual action. In R. G. Parker & J. H. Gagnon (Eds.), Conceiving sexuality: Approaches to sex research in a postmodern worm (pp. 183 213). New York: Routledge.

Lawrance, K. & Byers, E. S. (1995). Sexual satisfaction in long-term heterosexual relationships: The interpersonal exchange model of sexual satisfaction. Personal Relationships, 2, 267 285.

Lawrance, K. & Byers, E. S. (1998). Interpersonal Exchange Model of Sexual Satisfaction Questionnaire. In C. M. Davis, W. L. Yarber, R. Bauserman, G. Shreer, & S. L. Davis (Eds.), Sexuality-related measures." A compendium (2nd ed., pp. 514-519). Thousand Oaks, CA: Sage.

Lawrance, K., Taylor, D., & Byers, E. S. (1996). Differences in men's and women's global, sexual, and ideal-sexual expressiveness and instrumentality. Sex Roles, 34, 337 357.

Litzinger, S. & Gordon, K. C. (2005). Exploring relationships among communication, sexual satisfaction, and marital satisfaction. Journal of Sex and Marital Therapy, 31,409~,24.

MacNeil, S. & Byers, E. S. (1997). The relationships between sexual problems, communication and sexual satisfaction. Canadian Journal of Human Sexuality, 6, 277-283.

MacNeil, S. & Byers, E. S. (2005). Dyadic assessment of sexual self-disclosure and sexual satisfaction in heterosexual dating couples. Journal of Social and Personal Relationships, 22, 169-181.

Markman, H. J., Renick, M. J., Floyd, F. J., Stanley, S. M., & Clements, M. (1993). Preventing marital distress through communication and conflict management training: A 4- and 5-year follow up. Journal of Consulting and Clinical Psychology, 61, 70-77.

McCarthy, B. W. & Bodner, L. E. (2005). The equity model of sexuality: Navigating and negotiating the similarities and differences between men and women in sexual behaviour, roles and values. Sexuality and Relationship Therapy, 20, 225-235.

Meeks, B. S., Hendrick, S. S., & Hendrick, C. (1998). Self-disclosure, love, and relationship satisfaction. Journal of Social and Personal Relationships, 15, 755-773.

Meston, C. & Trapnell, P. (2005). Development and validation of a five-factor sexual satisfaction and distress scale for women: The Sexual Satisfaction Scale for Women (SSS-W). Journal of Sexual Medicine, 2, 66-81.

Miller, S. A. & Byers, E. S. (2004). Actual and desired duration of foreplay and intercourse: Discordance and misperceptions within heterosexual couples. The Journal of Sex Research, 41, 301-309.

Perlman, D. & Fehr, B. (1987). The development of intimate relationships. In D. Perlman & S. Duck (Eds.), Intimate relationships." Development, dynamics, and deterioration (pp. 1342). Newbury Park, CA: Sage.

Purnine, D. M. & Carey, M. P. (1997). Interpersonal communication and sexual adjustment: The roles of understanding and agreement. Journal of Consulting and Clinical Psychology, 65, 1017-1025.

Rosen, R. C. & Leiblum, S. R. (1988). A sexual scripting approach to problems of desire. In S. R. Leiblum & R. C. Rosen (Eds.), Sexual desire disorders (pp. 168-191). New York: Guilford.

Rubin, Z., Hill, C. T., Peplau, L. A., & Dunkel-Schetter, C. (1980). Self-disclosure in dating couples: Sex roles and the ethic of openness. Journal of Marriage and Family, 42, 305 317.

Wiederman, M. W. (1999). Volunteer bias in sexuality research using college student participants. Journal of Sex Research, 36, 59-66.

Sheila MacNeil

Student Counselling, Dalhousie University

E. Sandra Byers

Department of Psychology, University of New Brunswick

This study was conducted by the first author in partial fulfillment of the doctoral degree in clinical psychology under the supervision of the second author. This research was supported by a graduate fellowship from the Social Sciences and Humanities Research Council. The authors would like to thank Barry Spinner, Enrico DiTommaso, Carmen Poulin, and the members of the Human Sexuality Research Group at the University of New Brunswick who provided invaluable feedback in the development of this study.

Correspondence should be addressed to E. Sandra Byers, Department of Psychology, University of New Brunswick, Fredericton, New Brunswick, E3B 6E4 Canada. E-mail: [e-mail]

(1) After controlling for men's [UND.sub.REW], women's sexual satisfaction remained negatively correlated with men's [UND.sub.CST].

(2) After controlling for women's [UND.sub.REW], men's sexual satisfaction remained negatively correlated with women's [UND.sub.CST].

Table 1. Intercorrelations Among Variables Included in the
Instrumental and Expressive Pathyways

                                1            2            3

1. Women's SSD                            .36 ***      .14
2. Women's NSD               .36 ***                   .05
3. Men's SSD                 .14          .05
4. Men's NSD                 .13          .25 **       .53 ***
5. Partner [UND.sub.REW]     .18 *        .14          .37 ***
6. Partner [UND.sub.CST]    -.11         -.11         -.01
7. Own REW-CST               .18 *        .30 ***      .22 **
8. Own GMREL                 .00          .39 ***      .15
9. Own GMSEX                 .18 *        .22 **       .28 **

                                4            5            6

1. Women's SSD               .13          .12          -.10
2. Women's NSD               .25 **       .13          -.01
3. Men's SSD                 .53 ***      .37 ***      -.04
4. Men's NSD                              .32 ***      -.04
5. Partner [UND.sub.REW]     .41 ***      .70 ***      -.24 **
6. Partner [UND.sub.CST]     .06         -.14           .51 ***
7. Own REW-CST               .23 **       .38 ***      -.36 ***
8. Own GMREL                 .24 **       .24 **       -.19 *
9. Own GMSEX                 .18 *        .38 ***      -.33 ***

                                7            8            9

1. Women's SSD               .08          .00          .15
2. Women's NSD               .11          .26 **       .21 *
3. Men's SSD                 .33 ***      .35 ***      .42 ***
4. Men's NSD                 .35 ***      .34 ***      .43 ***
5. Partner [UND.sub.REW]     .53 ***      .45 ***      .57 ***
6. Partner [UND.sub.CST]    -.25 **      -.16 *       -.28 **
7. Own REW-CST               .46 ***      .65 ***      .83 ***
8. Own GMREL                 .61 ***      .61 ***      .73 ***
9. Own GMSEX                 .81 ***      .71 ***      .64 ***

Note. N= 104. Correlations for the pathways for men are above the
diagonal; correlations for the pathways for women are below the
diagonal. Numbers on the diagonal represent similarity between
partners. SSD = Sexual Self-Disclosure; NSD = Nonsexual
Self-Disclosure; Partner [UND.sub.REW] = Partner Understanding of
Sexual Rewards; Partner [UND.sub.CST] = Partner Understanding of
Sexual Costs; REW-CST= Balance of Sexual Rewards to Sexual Costs;
GMREL=Relationship Satisfaction; GMSEX=Sexual Satisfaction.

* p < .05; ** p < .01; *** p < .001