1. Introduction

Studies of mate choice reveal a puzzle. When asked to describe traits they value in a relationship partner or asked to choose preferred traits from a provided list, people report that a “good sense of humor” is highly valued (e.g., Buss & Barnes, 1986, Feingold, 1981, Goodwin, 1990, Hendel, 1978). Men and women typically report this preference with about equal frequency (e.g., Daniel et al., 1985, Feingold, 1992). However, two studies that experimentally manipulated humor production yielded a contradictory result: Women preferred relationship partners who produced humor, but men showed no such preference (Bressler & Balshine, 2004, Lundy et al., 1998). Thus, men say they value a partner with a good sense of humor, yet do not rate funny women as more desirable.

Why would ranking the importance of a good sense of humor in a partner produce different results than asking about the desirability of someone who produces humor? One possible resolution is that, by good sense of humor, a man denotes not a partner's humor production, but rather her receptivity to his own humor.

Several studies have investigated the relationship between humor appreciation and interpersonal attraction (e.g., Bippus, 2000, Murstein & Brust, 1985, Priest & Thein, 2003, Rust & Goldstein, 1989), but none have examined sex differences. One exception is the work of Grammer and Eibl-Eibesfeldt (1990) who found that during mixed-sex dyadic conversations, the amount of laughter a woman produced was more predictive of both sexes' interest in dating each other than was the man's laughter. This suggests that a woman's receptivity to humor may signal her sexual interest.

Miller, 2000a, Miller, 2000b, Miller, 2000c, Miller, 2001 suggests that the capacities to produce and appreciate humor have both evolved via sexual selection. He argues that those who carry relatively few deleterious genetic mutations are more competent at the set of cognitive skills—such as intelligence and creativity—required to produce entertaining humor. As a result, sexual selection has favored those who produce humor because it elevates their mating success and those who preferentially mate with funny people, because doing so provides offspring with genetic benefits.

If humor production signals genetic quality and humor receptivity signals sexual interest, the sexes may respond differently to these signals. Signals of genetic quality may have more impact on women's mating decisions, because women's higher minimum parental investment imposes higher costs from suboptimal mating (Trivers, 1972). Conversely, signals of receptivity may be more important in men's mating decisions because male reproductive success is more limited by access to mates (Bateman, 1948, Trivers, 1972). Thus, sexual selection may have more strongly favored women who reacted positively to humor producers and men who attended preferentially to women who appreciated their humor.

In this study, we examined two related questions about men's and women's responses to humor production and receptivity. First, we examined sex differences in the impact of humor production and humor receptivity on men's and women's categorization of others as having a good sense of humor. We measured participant's preferences for a relationship partner's “sense of humor,” humor production, and receptivity to humor. If men's and women's use of the phrase good sense of humor differs, then both sexes may prefer a partner with a sense of humor but have different preferences for humor production and receptivity. Second, we examined men's and women's preference for humor production or receptivity in partners for a variety of sexual relationships. Furthermore, because humor production and appreciation are hypothesized to have evolved for mate attraction, we examined whether preferences would be most apparent in sexual relationships rather than friendships.

2. Method

3. Results

3.1. Are there sex differences in how men and women use the phrase a good sense of humor?

Men and women both valued a good sense of humor in a relationship partner [one-sample t test: women, t(73)=15.4, p<.001; men, t(54)=7.10, p<.001] as well as their partners' receptivity to their own humor [women, t(73)=11.1, p<.001; men, t(54)=6.56, p<.001]. However, only women valued their partner's ability to produce humor [women, t(73)=8.40, p<.001; men, t(54)=0.70, p=1] (see Fig. 1).

Fig. 1. Women's and men's mean ratings of the importance of a relationship partner's sense of humor, humor production, and receptivity to humor. The dashed line at four equals the score expected if participants consistently responded to items with no preference (4=neither agree nor disagree).

There was a significant interaction between the sex of the participant and the scores on the subscales [repeated-measures ANOVA: F(2,252)=9.44, p<.001]. Women valued a partner's sense of humor and production of humor more than men did [independent samples t tests: sense of humor subscale, t(127)=4.20, p<.001; partner's production of humor subscale, t(127)=5.14, p<.001], but the sexes did not differ in the degree of value placed on a partner's receptivity to their own humor [t(127)=0.97, p=1]. See Fig. 1.

Within-sex analyses revealed that women valued a partner's production of humor as much as they valued a partner's receptivity to their own humor [paired t test, t(73)=1.10, p=.10], whereas men valued a partner's receptivity to their own humor more than a partner's production of humor [t(54)=5.07, p<.001] (see Fig. 1).

3.2. Are men and women attracted to different behaviors during humorous interaction?

Participants' choices for humor producers or appreciators depended on their sex and the type of relationship they were considering. Women responded similarly to the different types of relationships [repeated-measures ANOVA, F(4,292)=1.56, p=.17] and showed directional preferences for humor producers for every relationship type. Women's preference for humor producers was significantly different from chance for dates [one-sample t tests: t(73)=3.94, p<.01], long-term relationships [t(73)=2.72, p=.04], and friendships [t(73)=3.45, p<.01], but not one-night stands [t(73)=0.82, p=1] and short-term relationships [t(73)=1.56, p=.60] (see Fig. 2).

Fig. 2. Women's and men's average proportion of choices for the humor producer rather than the humor appreciator for four types of sexual relationships and for friendships. The dashed line represents the chance expectation. Values above chance represent a preference for the humor producer, whereas values below chance represent a preference for the humor appreciator.

Men's preferences were directionally opposite to those of women in most cases; men preferred partners who were receptive to their own humor. However, this preference varied across the different relationship types (repeated-measures ANOVA: F(4,216)=3.22, p=.03). As with women, men's preferences were significantly different from chance for dates (one-sample t tests: t(54)=−4.90, p<.01), and a nonsignificant trend was found for long-term relationships [t(54)=−2.47, p=.09] but not for one-night stands [t(73)=−0.78, p=1], short-term relationships [t(54)=−1.27, p=1], and friendships [t(54)=0.53, p=1]. Between the different relationships, men had a greater preference for receptive partners for dates compared to friendships [t(54)=4.16, p<.001], but there were no differences between any other relationships (see Fig. 2).

Comparing preferences between the sexes for each of the sexual relationships, women showed a stronger preference for humor producers than did men for dates [independent samples t tests: t(127)=6.21, p<.0001] and long-term relationships [t(127)=3.67, p<.001], but not for short-term relationships [t(127)=1.99, p=.20], one-night stands [t(127)=1.12, p=.26], and friendships [t(127)=1.88, p=.24] (see Fig. 2).

4. Discussion

Our two experiments suggest that the sexes differ in the value they place on a partner's humor production vs. receptivity. The results of the categorization questionnaire suggest that women value a partner who can produce humor and who is receptive to their own humor, whereas men value only a partner's receptivity to their own humor. Therefore, the apparent discrepancy between men's valuation of a good sense of humor (e.g., Daniel et al., 1985, Feingold, 1992) and their lack of attraction to funny women (Bressler & Balshine, 2004, Lundy et al., 1998) appears to be resolved: Men's representation of partners who have a good sense of humor does not encompass their ability to produce humor. Furthermore, when forced to choose between humor production and humor appreciation in potential partners, women valued humor production, whereas men valued receptivity to their humor. Only when evaluating friendship partners do men show a directional preference for humor production.

Preferences for humor producers or appreciators varied across the five types of relationships, especially for men. Although men generally chose humor appreciators and women chose humor producers, these preferences were significant only for dates and long-term relationships. We offer two potential explanations for why these preferences did not hold for one-night stands and short-term relationships. First, it is possible that due to our young participants' lack of experience or to ambiguity in the terms used, it may have been more difficult for them to imagine or report preferences for the latter two types of relationships than for the former relationships. Alternately, the lack of humor preferences in these contexts might be genuine. If so, this suggests an imprecision in Miller, 2000a, Miller, 2000b, Miller, 2000c, Miller, 2001 sexual-selection-of-humor model. If humor production signals genetic quality and humor receptivity signals sexual interest, then we should see the most sexually dimorphic preferences in the context of short-term and one-night stand relationships, because such relationships provide minimal benefits (e.g., parental care) beyond genes; but this is not what we observe.

Prior research has shown that certain aspects of humor interaction, such as similarity of sense of humor between partners, predict relationship cohesion and marital satisfaction (Murstein & Brust, 1985, Rust & Goldstein, 1989, Ziv, 1988), suggesting that humor may play more of a role in relationship maintenance than formation. However, we know of no theory to explain why men and women would show opposing preferences for humor production and appreciation for long-term partners. Further research needs to more thoroughly investigate the extent to which humor-related behaviors are valued by men and women for different types of relationships.

More research is necessary on the relationship between humor receptivity and the communication of sexual interest. Our finding, together with that of Grammer and Eibl-Eibesfeldt (1990), suggests that men prefer sexual partners who appreciate their own humor because that response signals sexual interest, but this interpretation is not definitive. A related question is how specific the laughter and smiling of the opposite sex are in communicating sexual interest in comparison to signals of general proceptivity (such as direct eye contact, open posture, and facial expressions).

In summary, our results augment prior studies on the sexual selection of humor. We have provided further evidence that sexual selection may have influenced humor production because it is specifically preferred by women in relationship partners. Furthermore, men's reported preferences for humorous partners may be the result of sexual selection shaping a male preference for partners who signal sexual interest through humor appreciation.