This item is only available as the following downloads:
THE EXTENT TO WHICH BIOLOGICAL AND SOCIOCULTURAL FA CTORS ACCOUNT FOR DIFFERENCES BETWEEN THE SEXES IN HUMOR TYPE PREFERENCE BY EMILY ACLE A Thesis Submitted to the Division of Social Sciences New College of Florida in partial fulfillment of the requirements for the degree Bachelor of Arts Under the sponsorship of Steven Graham Sarasota, Florida May, 2012
Sex and Biological and Social Factors in Humor ii Acknowledgements I would like to thank Professor Graham, Professor B arton, Professor Harley, everyone at USF who assisted me and/or volunteered for my study and Professor Cooper. Without them this thesis would not be possible. I would als o like to thank my friends, especially Katya Ariano, Cassie Wood, Natasha Burnett, and Lis a Spear, my loving boyfriend Scott Smith, and my family, for their unconditional suppo rt on this thesis, and beyond.
Sex and Biological and Social Factors in Humor iii TABLE OF CONTENTS Page Number Acknowledgements ii Table of Contents iii Abstract iv Introduction 1 Background on Humor 3 Sex Differences in Humor 7 Social Arguments for Gender Differences in Humor U se 13 Biological Arguments for Sex Differences in Humor Use 19 The Relationship between Gender Role and 2D:4D Rat io 37 The Current Study 37 Method 41 Results 44 Discussion 50 References 64 Footnotes 71 Table 1 72 Table 2 73 Table 3 74 Appendix A 75
Sex and Biological and Social Factors in Humor iv THE EXTENT TO WHICH BIOLOGICAL AND SOCIOCULTURAL FA CTORS ACCOUNT FOR DIFFERENCES BETWEEN THE SEXES IN HUMOR TYPE PREFERENCE Emily Acle New College of Florida, 2012 ABSTRACT Differences between men and women in their use of f our humor types were investigated, with an emphasis on the extent to whi ch one biological factor 2D:4D ratio and one sociocultural factor gender role score accounted for these differences. It was hypothesized that men would demonstrate greater use of aggressive humor than women, and no hypotheses were made regarding differ ences between mens and womens use of self-deprecating, affiliative, or se lf-enhancing humor. Furthermore, no hypotheses were made regarding the relationship bet ween 2D:4D ratio and use of any of the humor types. Lastly, it was hypothesized that m ore male-typical gender role scores would predict greater use of aggressive humor, grea ter use of self-deprecating humor, and greater use of self-enhancing humor, and no hypothe sis was made regarding the relationship between gender role scores and affilia tive humor. To test this, 35 participants first completed a pac ket of questionnaires, then had a photocopy of both their hands taken to determine 2D :4D ratio for each hand. The packet included the Humor Styles Questionnaire to determin e how frequently each of the four
Sex and Biological and Social Factors in Humor v humor types was used, the short-form Bem Sex Role I nventory Revised to determine their gender role score, and a demographic question naire. Findings were mixed and difficult to interpret. T-t ests demonstrated greater use of aggressive humor in males than females but multiple regression analyses did not find this effect; furthermore, no differences arose between m en and women in their use of selfdeprecating, affiliative, or self-enhancing humor. Correlations demonstrated relationships between 2D:4D ratio and aggressive humor use but mu ltiple regression analyses did not find this effect; furthermore, no relationships wer e found between 2D:4D ratio and any of the other humor types. Lastly, no relationships wer e found between gender role score and humor type use. Thus, it is unclear whether the hypotheses were sup ported, especially considering the problematic composition of this particular samp le. Future studies should therefore use a larger and more representative sample. Dr. Steven Graham Division of Social Sciences
1 The Extent to Which Biological and Sociocultural Fa ctors Account for Differences Between the Sexes in Humor Type Prefere nce Humor is a phenomenon that permeates the world, mos t notably playing a role in social interaction (though it can also occur in iso lation), and has a variety of important functions, such as relieving tension, bonding, and asserting social dominance or status. Considering the multiple functions of humor, then, it would be reasonable to expect large variation in peoples uses and preferences for humo r. One area in which these differences manifest themselves is gender. Take the following j oke into consideration: On Christmas morning a cop on horseback is sitting at a traffic light, and next to him is a kid on his brand new bike. The cop says to the kid, Nice bike you got there. Did Santa bring that to you? T he kid says, Yeah. The cop says, Well, next year tell Santa to put a tail-light on that bike. The cop then proceeds to issue the kid a $20.00 bic ycle safety violation ticket. The kid takes the ticket and before the cop rides off says, By the way, that's a nice horse you got there. Did Santa b ring that to you? Humoring the kid, the cop says, Yeah, he sure did. The kid says, Well, next year tell Santa to put the dick underneath the horse, instead of on top. (Funny Web Jokes, 2012). Although the majority of people would find this jok e humorous, studies suggest that males would view it as more humorous than fema les would, due to its aggressive and condescending nature. Indeed, several differences b etween the sexes in their humor style preferences have been reported. From here, it is on ly natural to wonder why these differences exist; are the reasons biological in na ture, sociocultural, or both? How much
2 control do we have over them? Such an understanding could prove especially useful for people who use humor to attack others, or even them selves, by promoting more healthy or peaceful relations, as only after understanding the root of certain behaviors can one make changes to them. Thus, this topic will be addressed by reviewing bas ic background information on humor. From here, the focus will become more specif ic, demonstrating the extent to which sex differences in humor exist. Emphasis is p laced on the type of humor used most often, as it carries the heaviest social implicatio ns and will be the focus of the study. This will set the stage for the next section of the thesis the social arguments for gender differences in humor use as part of the ov erall analysis of exactly why these gender differences exist. Again, these gender diffe rences can most likely be attributed to either sociocultural and/or biological factors; the refore, this section will be addressing the possibility of sociocultural factors contributing t o gender differences in humor use. The next major section of the thesis will address t he possibility of biological factors contributing to sex differences in humor us e. Lack of research on the relationship between biological gender measures and humor type u se preference drives the biological analysis in the current study. This analysis will first review studies that demons trate a biological specifically, genetic influence on humor type preferences, for the sake of showing that humor type preference is affected by at least one biological f actor. (The biological factor the current study focuses on is 2D:4D ratio, not genetics.) How ever, these studies do not take gender or hormones into account. This section will the n evolve into a brief discussion of studies that demonstrate humor-related sex differen ces though not in humor type use
3 preference, as no such studies exist that manifes t themselves biologically. However, while it is possible that these biological manifest ations cause the aforementioned sex differences in humor type use preferences, it is al so possible that they do not. Due to this limitation of non-experimental studies (i.e., that direction of causality cannot be inferred), the next section within the biological segment of t he study proposes using a biological measure known as 2D:4D ratio as a way of possibly i nferring causality in sex differences in humor. Following this proposal, because 2D:4D ratio is onl y partially established in the literature, several studies on its reliability will be reviewed. From here, studies reviewing the relationship between traits parallel to the pre viously mentioned four types of humor (i.e., looking at aggression for aggressive humor) and 2D:4D ratio will be discussed. These will be used as parallel arguments to support the idea that biological factors specifically 2D:4D ratio may potentially explain sex differences in humor type use preferences, although they very well may not. Finally, studies reviewing a possible relationship between 2D:4D ratio and gender role will be explored, in order to address the poss ibility that gender role may be influenced by the persons underlying biology, crea ting the need for a particularly careful analysis. Background on Humor Theories of Humor Many theorists consider humor its own mod e of discourse among social interactions. Mulkay (1988) categorized discourse i nto either serious or non-serious modes. In the serious mode of discourse, the aim is to create a single, objective reality
4 that all (or at least most) speakers can agree on. Self-contradiction, as well as ambiguity and disagreements of interpretation, are avoided in the interest of smooth social functioning. However, in the non-serious mode of di scourse (i.e. humor), nonsense is used with certain limitations. Ambiguity, paradox, multiple interpretations of reality, and partially resolved incongruity are acceptable (and even defining of) humorous discourse. In humor mode, socially unspeakable topics can be d iscussed without fear of repercussion, due to the ambiguity of humor. Classifications of Humor Although the majority of people have a ge neral notion of what humor is and can both produce and recognize it, achieving an actual understanding of humor has proven surprisingly difficult. This is likely due to the f act that sense of humor is not a single unitary concept, but rather, a variety of loosely r elated concepts, such as humor creation abilities, enjoyment of particular types of humorou s stimuli, the tendency to tell jokes and amuse others, and use of humor as a coping mechanis m. It is important to note that each of these requires different methods of measurement, and that, while they may be loosely related, they are generally not strongly inter-corr elated (Martin et al., 2003). As a result, a variety of humor classification systems have come i nto existence. One popular classification system was created with a particular emphasis on how humor arises in social interactions (Graeven & Morr is, 1975). The four categories of humor found to occur socially were: (1) humor arisi ng from a mass media source; (2) humor emerging spontaneously within a situation; (3 ) humor arising from a memorized joke; and (4) humor arising from the telling or rec alling of a previous event. They found that humor emerging spontaneously within a situatio n occurred the most frequently,
5 whereas humor arising from memorized jokes, mass me dia sources, or the telling or recalling of a previous event were the least freque nt. This is suggestive of the fact that, while studying the humor behind jokes and the like (as many humor studies have done) is somewhat important, it is even more important that studies focus on spontaneous and natural humor that arises in social situations sinc e it represents the humor most frequently used. Indeed, several studies have found that the aspect of humor most relevant to emotional and psychosocial well-being is spontaneou s humor used in interactions with others. Martin and colleagues (2003) set out to ide ntify the motives or intentions behind social uses of humor, resulting in a new classifica tion system of humor that emphasized the psychological healthiness of each type of humor This led to a final categorization of humor use, with four possible subcategories: aggres sive, self-deprecating, affiliative, and self-enhancing. Aggressive humor is defined as humo r meant to manipulate or attack other people; this includes sarcasm, disparagement, teasing, and offensive humor. Selfdeprecating humor includes attacking or disparaging ones self, usually for the purpose of amusing other people at ones own expense, as well as defensive denial. Affiliative humor involves using humor to bond, to be friendly or amusing, or to release tension, and often occurs in the form of spontaneous witty bante r. Self-enhancing humor involves the tendency to maintain a humorous outlook on life, to be amused by the incongruities in life, and to cope with stress. These humor types intercorrelate to a small degree, with a positive relationship between affiliative and self-enhancing humor, a wea k positive relationship between affiliative and aggressive humor, and a positive co rrelation between aggressive and self-
6 deprecating humor. However, these relationships are small enough to conclude that each subscale measures dimensions relatively distinct fr om the others. Both aggressive humor and self-deprecating humor have been shown in vario us studies to relate to generally negative or harmful personality traits and relation ships, whereas affiliative humor and self-enhancing humor have been shown to relate to g enerally positive personality traits and relationships. For example, Martin and colleagues (2003) administe red the HSQ and numerous other questionnaires including the Center for Epi demiological Studies Depression Scale (CESD), the Life Orientation Test (LOT), the CookMedley Hostility Scale (Ho), the BussPerry Aggression Questionnaire (BPAQ), the Symptom Checklist-90-R (SCL90-R), the State-Trait Anxiety Inventory (STAI), th e Index of Self-Esteem (ISE), the Rosenberg Self-Esteem Inventory (RSEI), the Social Support Questionnaire (SSQ), and the Miller Social Intimacy Scale (MSIS) to sample s ranging from 94 to 485 participants, and found a variety of significant co rrelations. Aggressive humor was found to relate positively to the BPAQ and the Ho, indica ting greater hostile attitudes and behaviors toward others. Self-deprecating humor was found to relate positively to the CESD, the STAI, the BPAQ and Ho, and the SCL-90-R, indicating greater depression, anxiety, aggression, and interpersonal sensitivity. Self-deprecating humor also related negatively to the RSEI, the ISE, the MSIS, and the SSQ, indicating lower self-esteem, lower social support and lower social intimacy. Aff iliative humor related positively to the RSEI, ISE, and the MSIS, indicating greater self-es teem and social intimacy, and related negatively to the CESD, and the STAI, indicating le ss depression and less anxiety. Finally, self-enhancing humor related positively to the RSEI, the ISE, the LOT, the
7 MSIS, and the SSQ, indicating greater self-esteem, optimism, social intimacy, and social support, and related negatively to the CESD and STA I, indicating less depression and less anxiety. These findings have further been supporte d in studies conducted by Veselka, Schermer, Martin, and Vernon (2010), and Yip and M artin (2006). In the first, 201 twin pairs completed the HSQ and the MT48, which measure s global toughness and its eight factors (Commitment, Control, Emotional Control, Co ntrol over Life, Confidence, Confidence in Abilities, Interpersonal Confidence, Challenge). Results showed that affiliative and self-enhancing humor correlated pos itively with all eight factors, whereas aggressive and self-deprecating humor correlated ne gatively with all the factors except commitment, control over life, and challenge. In th e second study, 111 participants completed the HSQ as well as measures of trait chee rfulness, social competence, and emotional intelligence. Self-enhancing humor was po sitively correlated with social competence, whereas aggressive and self-deprecating humor were negatively related to social competence and emotional intelligence. Sex Differences in Humor Studies have shown differences between th e sexes in various aspects of humor, including type of humor considered funniest, freque ncy of public laughter, attractive qualities in the opposite sex, and type of humor us ed most often. Type of Humor Considered Funniest Men and women generally consider different types of humor to be funniest. Johnson (1992) had participants rate how funny they found jokes presented on a computer screen. He found that women generally pref erred nonsense to aggressive
8 humor, silly to sick humor, role-reversal to word-p lay humor, and disease disparagement (defined as joking about the ill) to irreverent (de fined as joking about those in positions of power) humor. Conversely, men preferred aggressi ve to nonsense humor, sick to silly humor, word-play to role-reversal humor, and irreve rent to disease disparagement humor. Furthermore, Jackson and Jackson (1997) found that, upon rating jokes with either a female initiator/male target or vice versa, males c onsidered both types equally funny, whereas women considered jokes with female initiato rs and male targets funnier than the opposite. Frequency of Public Laughter Several studies have indicated that men a nd women differ in their frequency of public laughter, though some studies have not found this effect. One study coded 10, 419 people for laughter in malls, schools, restaurants, comedy clubs, playgrounds, and stores, and found that females laugh more than males on any given day (Chapell, Batten, Brown, Gonzalez, Herquet, Massar, & Pedroche, 2002) Anot her study, on the other hand, found no difference between men and women in their freque ncy of laughter, unless personality type was taken into account (Martin & Kuiper, 1999) They had 80 participants complete a daily laughter record for three days, as well as a self-report measure of Type A personality characteristics. The amount of type A t raits in men positively correlated with daily laughter, whereas the opposite trend occurred in women; the amount of type A traits in women negatively correlated with daily laughter. Thus, there appear to be gender differences in frequency of public laughter as well Attractive Qualities in the Opposite Sex Another difference regarding humor involv es what each sex finds attractive in
9 members of the opposite sex. The specific qualities that were looked at were production versus receptivity of humor, and interactions betwe en status and humor type. One study that looked at production versu s receptivity of humor as attractive qualities in potential sexual partners was conducte d by Bressler, Martin, and Balshine (2006). Their study consisted of two parts, the fir st being an analysis of whether men and women defined sense of humor in others differentl y depending on whether the others produced humor or were receptive to it, and the sec ond being whether men and women were more attracted to a member of the opposite sex who produced humor versus one who was receptive to humor. In the first part of th eir study, the authors administered a categorization questionnaire they had created to un iversity students that consisted of three subscales: the importance of a relationship partner s sense of humor, the importance of a partners receptivity to the participants own humo r, and the importance of a relationship partners production of humor. Significance was det ermined by comparing men and womens mean scores on a particular subscale to the score expected if participants had consistently answered each item without a preferenc e, i.e. a 4 neither agree nor disagree. Results showed that while both men and women rated the importance of a partners sense of humor highly enough to reach sig nificance, women rated importance of both receptivity and productivity highly enough to reach significance, whereas men only rated importance of receptivity not of productivi ty highly enough to reach significance. In other words, men and women differ in how they define sense of humor in terms of the opposite sex while women consider both receptivity and productivity to constitute sense of humor, men only seem to consi der receptivity sense of humor. Thus, Bressler and colleagues (2006) have brought t o light a discrepancy between mens
10 and womens definition of sense of humor, which n ot only explains inconsistent findings in previous literature on this topic, but which also shows the importance of taking this discrepancy into account in future stud ies. In the second part of their study, Bressl er and colleagues (2006) administered two forced-choice questionnaires to the same university students to determine whether they found receptivity or productivity of humor more att ractive in the opposite sex. Each questionnaire consisted of two opposing hypothetica l situations. In the first situation, participants were instructed to imagine that they h ad interacted with an attractive member of the opposite sex (Member 1), and that the partic ipant had produced humor which Member 1 enjoyed, but that Member 1 had produced hu mor the participant did not find funny. In the second situation, participants were i nstructed to imagine that they had interacted with another attractive member of the op posite sex (Member 2) who was identical to Member 1 in every way, except for the following: when the participant produced humor, Member 2 did not enjoy it, though w hen Member 2 produced humor, the participant enjoyed it. In each questionnaire, participants were forced to choose which person Member 1 or Member 2 they would choose f or a one-night stand, a date, a short-term relationship, a long-term relationship, and a friendship. Results indicated that women significantly preferred humor producers overa ll, but only significantly for dates, long-term relationships, and friendships. Men, on t he other hand, significantly preferred partners receptive to their own humor overall, but only significantly for dates, with a trend approaching significance for long-term relati onships. Thus, the data suggest that women may generally prefer partners who produce hum or over partners receptive to their own humor, whereas men may generally prefer partner s to be receptive to their own
11 humor over partners who produce humor. Type of Humor Used Most Often Generally speaking, men tend to use more aggressive or more negative forms of humor than women do. One study that demonstrated th is effect was conducted by Martin and colleagues (2003), the creators of the Humor St yles Questionnaire (HSQ). This selfreport measure tests for personal use of the previo usly listed four types of humor: affiliative humor, self-enhancing humor, aggressive humor, and self-deprecating humor. After finalizing the HSQ, they collected responses to the HSQ as well as demographic information from 1195 participants (725 women and 4 70 men), ranging in age from 14 to 87 years. T-tests revealed that males reported grea ter use of self-deprecating humor and especially aggressive humor than females, indicatin g that not only do males attack themselves using self-disparaging humor and allow o thers to laugh at their expense more than females do, but that they also attack others w ith sarcasm, teasing, and hurtful comments more than females do. Using this same humor scale, a subsequent study by Greengross and Miller (2008) partially replicated the findings of Martin and col leagues (2003). They had 64 female and 32 male university students complete a basic demogr aphic inventory, as well as the HSQ. Statistical analysis revealed that men reported usi ng aggressive humor more than women did. However, they found no significant difference between men and women on use of self-deprecating humor, in contrast with Martin and colleagues (2003) findings. A possible explanation may reside in the fact that Ma rtin and colleagues study (2003) used a sample with a much larger age range than the one used by Greengross and Miller
12 (2008), who only used university students, and that some studies have demonstrated interactions between age and sex on humor use style Somewhat related to the idea that men use more aggr essive humor than women is the idea that men use a higher proportion of differ entiating humor than women, whereas women may use a higher proportion of cohesion-build ing humor than men. According to Robinson and Smith-Lovin (2001), this difference ca n be explained by social status. They argue that people of lower social status (i.e., wom en) are less likely to be socially aggressive, due to the limitations of their status, and are therefore forced to resort to cohesion-building humor, whereas people of higher s ocial status (i.e., men) are more likely to be socially aggressive because they can g et away with it, as well as also using differentiating humor to maintain their high status This idea was supported in Robinson and Smith-Lovins (2001) study, in which they revie wed conversation transcripts, containing group and individual laughter, of 29 six -person university-age discussion groups originally recorded in the early 1980s. Of 5,640 speaker turns, there were 375 total humor attempts, defined as either remarks th at lead to group laughter, remarks that are followed by speaker laughter, and remarks that all four coders agreed were humor attempts but which did not lead to laughter (Robin son & Smith-Lovin, 2001, p. 10). Results showed that over 35% of mens humor attempt s were differentiating, whereas only 26% of womens humor attempts were differentia ting. However, results regarding cohesion-building humor attempts merely approached statistical significance, with 56% of womens humor attempts being cohesion-building v ersus 50% of mens attempts. Either way, these results are somewhat in line with other studies showing that men use more aggressive humor than women, as there is a fin e line between differentiating and
13 aggressive humor. This can be said because differen tiating humor is arguably a mild form of aggression, as its purpose is to maintain high s tatus. Perhaps the only limitation of this study is that the data were collected almost 30 yea rs ago, a time span during which the gap between mens and womens behaviors may have sh runk. More current studies would have to be conducted to determine whether thi s particular sex difference still applies. Social Arguments for Gender Differences in Humor Us e Gender Theory Crawford (2003) discusses various theories of gende r, including the essentialist viewpoint, the social interactionist viewpoint, and the social constructionist viewpoint. From the essentialist viewpoint, gender is conceive d as a fundamental attribute of individuals. Social interactionists and social cons tructionists, on the other hand, view gender as a social construct, or a system of meanin gs that governs access to power and resources and that organizes interactions. According to Crawford (2003), the social interactionist perspective views gender as a system of meanings that operates at individual interactional, and social structural levels (p. 1416). At the individual level, gender is the internalization of a masculine or feminine schema; it is part of the self-concept. At the interactional level, gender is both portrayed and reinforced in interactions with other s. At the social structural level, gender is used to categorize people into groups, which inf luence access to power, status, and material resources. The social constructionist perspective, o n the other hand, takes the interactionist perspective a step further and considers gender as nothing but a way of making sense of
14 transactions. It is not an attribute of individuals but rather, considered a verb (doing gender). As Crawford states, Doing gender reflec ts the social constructionist view that gender is a salient social and cognitive category t hrough which information is filtered, selectively processed, and differentially acted upo n to produce self-fulfilling prophecies about women and men (p. 1417). These viewpoints are important when study ing gender and humor, because both perspectives, especially the social constructionist perspective, emphasize language or speech style used in social interaction as a way of creating and defining gender; and, it is undeniable that humor is used as a mode of discours e during social interactions. Crawford holds that humor is used both as a means of constructing and deconstructing gender in social interactions. Speci fically, as she phrases it, The power of conversational humor in constructing and presenting a self is related to its flexibility, indirectness, and ambiguity (p. 1421). By acting i n gender-typical ways through humor use, a person can reinforce their gender to others. However, humor also can be used as a safe medium to act in gender-atypical ways, due to its flexibility, indirectness, and ambiguity. Generally speaking, men tend to use humor to construct their gender in ways that emphasize power. Mulkay (1988) found that mens hum or often objectified women, emphasized their sexual availability, and generally categorized them as inferior. Furthermore, Cameron (1997) found that men often jo ked and gossiped about other men, who they called gay, as a way of distinguishing t hemselves as superior. One could say that mens more frequent use of aggressive humor ma y be used as a means of displaying their power, and as a way of constructing and reinf orcing their male gender to others.
15 On the other hand, women tend to use humo r to build cohesion, at least amongst other women. Crawford argues that this may be a for m of emancipation from the power men hold over them. She states that because women a re socially subordinated, they create a sense of collective identity in their humor with each other, using their humor as a form of gender construction. Gender Role Identification and Humor In line with these theories, several studies have d emonstrated connections between gender role identification and humor. One s uch study was conducted by Mio (2009), who compared types of humor found funny bet ween individuals classified by the Bem Sex Role Inventory Revised (BSRI-R; Bem, 1977 ) as sex-typed females (defined as females who score high on the feminine scale and low on the masculine scale), sextyped males (defined as males who score high on the masculine scale and low on the feminine scale), and androgynous females and males (defined as individuals who score high on both the masculine and feminine scales). Th ey did not include participants categorized as undifferentiated, defined as individ uals who score low on both the masculinity and femininity scales, citing no partic ular reason for this. This study consisted of two parts. In the first part, which co nsisted of 136 participants and was conducted in the early 1990s, they administered th e BSRI-R, a questionnaire consisting of traditional masculine, traditional feminine, and neutral items, for a total of 30 items, each being rated by the participant regarding how m uch the item described his or her self. Following this questionnaire, participants were giv en 32 mirror metaphor pairs, defined as a metaphor comparing a high-status item to a low -status item and vice versa (such as My surgeon is a butcher among doctors and My but cher is a surgeon among meat-
16 cutters). Participants were instructed to place a check mark next to the metaphor they found funnier in each pair either the disparaging metaphor, or the uplifting metaphor. Results were calculated by subtracting the total nu mber of uplifting metaphors chosen as funnier from the total number of disparaging metaph ors chosen as funnier for each participant, meaning a positive value showed the pa rticipant found disparaging metaphors to be more humorous and a negative value showed the participant found uplifting metaphors to be more humorous. It was found that 59 % of the sex-typed females felt that disparaging metaphors were funnier, 71% of the andr ogynous participants felt that disparaging metaphors were funnier, and 78% of the sex-typed males felt that disparaging metaphors were funnier. Thus, although all particip ants, regardless of gender association, found disparaging metaphors funnier than uplifting metaphors, the more sex-typed male the participant was, the funnier they found the dis paraging metaphors. This suggests that sex-typed (masculine) males may appreciate aggressi ve humor more than androgynous (equally masculine and feminine) males, just as sex -typed (feminine) females may appreciate aggressive humor less than androgynous ( equally masculine and feminine) females do. It is worth noting, however, that no in ferential statistics were run on these percentages, making it unknown whether these differ ences were significant. The second part of Mios (2009) study, which occurr ed around the late 1990s years after the first part of his study tested mo re for differences in type of humor used/created between sex-typed females, sex-typed m ales, and androgynous females and males, rather than the type of humor they found fun niest. Again, the BSRI-R was administered to a large pool of participants, but t he authors stated that they had considerable difficulty finding sex-typed men, thou gh not sex-typed women, for this
17 study. Participants were given metaphor stems, each of which included either a highstatus or low-status item in the topic position, an d which also included a blank that was to be filled in by one of three choices a high, medi um, or low-status choice. (For example: a metaphor stem with a high-status item in the topi c position would include My surgeon is a _______ among doctors with choices including star person and butcher). Participants were to circle the choice they felt wo uld make the metaphor the funniest. Disparagement scores were calculated by subtracting the number of high-status choices chosen from the number of low-status choices chosen for only the high-status topic metaphors. As in the first part of Mios (2009) stu dy, a positive score indicated that the participant found disparaging metaphors to be funni er, whereas a negative score indicated that they found uplifting metaphors funnier. Unexpectedly, results showed that while sex-typed m en had the highest disparagement scores, followed by androgynous women sex-typed women, and lastly, by androgynous men, these differences were not signifi cant. The only significant effect was that both men and women found disparaging metaphors funnier than uplifting metaphors. Thus, the second part of Mios (2009) study did not find any differences in the type of humor created between sex-typed males, sex-typed fe males, and androgynous males and females, in contrast to the first part of his study Mio speculated that this may be due to the difficulty he encountered in finding sex-typed males for the second part of his study, conducted about a decade after the first part, whic h in turn led him to consider the possibility that the BSRI-R has become an outdated measure. He also noted that gender attitudes may have shifted over the years, possibly accounting for the lack of significant results.
18 In support of Mios first study and in co ntrast to his second study, Martin and colleagues (2003) did find evidence of gender ident ification predicting the type of humor personally used. They used a different measure of g ender identification than Mio, however. Undergraduates were asked to complete the Extended Personal Attributes Questionnaire (EPAQ; Spence, Helmreich, & Holahan, 1979), which provides four different types of scores positively valued mascu linity (e.g., Independent), negatively valued masculinity (e.g., Hostile), positively valu ed femininity (e.g., Warm), and negatively valued femininity (e.g., Servile). Parti cipants also completed the HSQ. Results showed that positively valued masculinity was posit ively correlated with self-enhancing humor, but that negatively valued masculinity was p ositively correlated with aggressive humor. On the other hand, positively valued feminin ity was positively related to affiliative humor, and negatively related to aggres sive and self-deprecating humor; furthermore, negatively valued femininity was negat ively related to self-enhancing humor and to affiliative humor. Put simply, higher mascul inity scores predicted greater use of self-enhancing humor and aggressive humor; higher f emininity scores predicted both greater and lesser use of affiliative humor (depend ing on whether it was positively or negatively valued femininity), and lesser use of ag gressive, self-deprecating, and selfenhancing humor. Thus, Martin and colleagues (2003) demonstrated that gender role orientation can predict aggressive humor use, selfdeprecating humor use, and selfenhancing humor use. Thus, differences between the sexes in th e type of humor they use may be accounted for, at least in part, by social factors. As humor is its own mode of discourse, it may be used as a means of gender construction in a manner consistent with societys
19 expectations for members of each sex. This theory i s supported by studies demonstrating relationships consistent with cultural norms betwee n gender role identity and the type of humor used by both men and women. However, the poss ibility that humor differences between the sexes may be at least somewhat biologic al particularly hormonal still remains. Biological Arguments for Sex Differences in Humor U se Not Taking Sex into Account: Biological Factors in Humor Type Preferences Several studies have found that genetics contribute at least partially to preferences for different humor types. One such study had 49 pa irs of monozygotic (identical) twins and 52 pairs of dizygotic (fraternal) twins rate co mics of various humor styles, including nonsense, satire, aggression, and sex (Wilson, Rust & Kasriel, 1977). Goodness of fit of genetic and between-family environment models for t hese humor styles were determined using chi-square tests, as well as intraclass corre lations. This analysis revealed a better fit of the environmental model than the genetic model f or nonsense, satire, and sex humor styles, and revealed a better fit of the genetic mo del than the environmental model, as well as a higher intraclass correlation for monozyg otic twins than dizygotic twins, for aggressive humor styles. In other words, this study found more of a genetic contribution to aggressive humor styles, but more of an environm ental (rather than genetic) contribution to nonsense, satire, and sex humor sty les. A larger study conducted by Vernon, Marti n, Schermer, Cherkas, and Spector (2008) had 1073 pairs of monozygotic twins and 895 same-sex pairs of dizygotic twins complete the Humor Styles Questionnaire. All four h umor styles aggressive, affiliative, self-enhancing, and self-deprecating were found t o have a substantial genetic
20 component to them, evidenced by monozygotic correla tions that were almost twice as high as dizygotic correlations. Heritability coeffi cients ranged from .34 to .49. However, neither of the above-mentioned studies analyzed sex differences. Taking Sex into Account: Humor-Related Biological D ifferences Kohn, Kellermann, Gur, Schneider, and Habel (2011) performed a functional neuroimaging study on humor, with an emphasis on di fferences between male and female brain activation patterns. They performed fMRIs on 29 healthy, right-handed individuals while showing them humorous cartoons for 7 seconds each. Results showed that in women, the ventral system implicated in detection and appraisal of emotion was activated, including the amygdala, insula, and Ante rior Cingulate Cortex (ACC). However, men showed activation in both the ventral and dorsal processing systems. These results indicate that women process humor tho ugh limbic reactivity, involving appraisal of its emotional features, while men appl y more evaluative, executive resources in humor processing. Though this may initially lead one to believe that gender differences in humor are biological, the fact is, it is unclear whether the thought causes the brain activity (the brain activity being the only visible thing in an fMRI) or if the brain activity causes the thought. In other words, it is possible that after learning and conforming to gendered social expectations, each gender thinks a certain way about humor, which happens to create their corresponding brain activit y. This issue can possibly be resolved by testing for 2d:4d ratio, defined as the ratio be tween the index and ring finger, which may be a stronger causal indicator of biological se x-typed tendencies. This is because 2D:4D ratio is determined before an individual is b orn and therefore, determined before this individual can even be influenced by sociocult ural factors.
21 2D:4D Ratio as a Reliable Measure of Prenatal Hormo ne Exposure Although not yet widely accepted, there is a growin g body of evidence that 2D:4D ratio is affected by (and therefore represent ative of) exposure to sex hormones in the womb, during which sexual differentiation of th e brain occurs. Specifically, males tend to exhibit lower 2D:4D ratios (with the ring f inger being proportionally longer than the index finger) than females. Because of this, 2D :4D ratio can, to some extent, be used as a causal indicator of how biologically male-typi cal or female-typical someone is. Several studies support the idea that sexual differ entiation of the brain is determined by exposure to sex hormones in both anim als and humans, especially during a critical period in the womb. For example, mammalian animal studies have shown that sex differences in neural structure and neuroanatomy, r eproductive and social behaviors, and cognitive function are affected by exposure to pren atal sex hormones (Hines, 2002). One study that showed this was conducted by de Vries, R issman, Simerly, Yang, Scordalakes, Auger, and Arnold (2002). They modified some geneti cally female (XX chromosome) mice to develop testes in place of ovaries, and mod ified some genetically male (XY) mice to develop ovaries in place of testes. They al so had control groups where the genetically female mice had ovaries, and the geneti cally male mice had testes. They found that copulatory behavior, social exploration behavior, and sexually dimorphic neuroanatomical structures were affected by the pre sence of testes or ovaries, rather than by the genetic sex of the mouse, indicating a stron g hormonal effect on these sexually dimorphic traits. Humans are likely similar to animals, particularly mammals, in this regard, considering similarities in development (Hines, 200 2.) Although experimental data in
22 humans do not exist due to the unethical nature of manipulating fetal hormone levels, several experiments of nature that demonstrate th e importance of sex hormones on sexual differentiation do exist. For one, females w ith disorders that expose them to greater amounts of androgens in the womb, such as C ongenital Adrenal Hyperplasia (CAH), are more likely to identify as male or expre ss dissatisfaction at being female, more likely to display male-typical childhood play patterns, as well as more likely to be sexually attracted to females. Likewise, males with disorders that prevent exposure to androgens in the womb, such as Complete Androgen In sensitivity Syndrome (CAIS), are more likely to identify as female, more likely to d isplay female-typical childhood play, and more likely to be sexually attracted to males ( Hines, 2002). Thus, it appears as though exposure to fetal sex hormones affects sexua l differentiation in humans in several regards. From here, it is only natural to wonder how prenata l sex hormones would affect 2D:4D ratio. One theory is that fetal cartilaginous tissue contains androgen receptors, making the digits of the hands sensitive to androge ns in the womb (Ben-Hur et al., 1997). Another slightly more complicated theory was propos ed by Manning and colleagues (1998). The differentiation of the fetal gonads, wh ich are responsible for the fetuss exposure to sex hormones, is controlled by Hox gene s. However, Hox genes also control the growth and differentiation of the fingers and t oes. Therefore, Manning and colleagues (1998) speculated that this sharing of causal facto rs in digit and gonad differentiation makes it possible that patterns of digit formation can reflect gonad function and, therefore, concentration of prenatal sex hormones.
23 Empirical evidence suggesting that 2D:4D ratio refl ects prenatal androgen exposure takes many forms. One is in the form of an imal experiments, which are important considering the lack of human experiments due to ethical limitations of manipulating human fetal hormone levels. One such a nimal study was conducted by Romano, Rubolini, Martinelli, Alquati, and Saino (2 005). They increased testosterone concentrations in the eggs of ring-necked pheasants and found that the digit ratios of adult pheasants who had received the injections wer e affected. Another similar study was conducted by Talaroviov, Krkov, and Blaekov (2009). They increased levels of maternal testosterone in rats during pregnancy, and found that both the male and female offspring of these rats had lower 2D:4D ratios and an elongated 4D on the left and right forepaws compared to controls. Taken together, thes e studies indicate that, at least in animals, digit ratios can be affected by exposure t o testosterone during prenatal development. Assuming Manning and colleagues (1998) theory abou t Hox genes mediating the relationship between digit ratio and prenatal sex h ormone levels is correct, it is not unlikely that the aforementioned animal studies can be extended to humans as well. This is because Hox genes are conserved in vertebrates ( Romano et al., 2005), making similar patterns of variation in digit ratios across verteb rate classes likely. In fact, these animal studies, particularly Talaroviov and colleagues study (2009), are even more compelling because the specific change that occurre d in 2D:4D ratio after testosterone exposure was in a direction representative of human male 2D:4D ratios. As previously stated, study after study h as demonstrated that 2D:4D ratio is sexually dimorphic in humans, with males having low er 2D:4D ratios than females,
24 which is suggestive of the fact that 2D:4D ratio is affected by prenatal testosterone levels since, according to a review paper by Breedlove, Co oke, and Jordan (1998), most somatic sex differences involve sex hormones. Further evide nce that 2D:4D ratio is likely affected by sex hormones in the womb is the fact th at this sex difference in 2D:4D ratio is generally well established early in utero (Galis, Ten Broek, Van Dongen, & Wijnaendts, 2010; Malas, Dogan, Evcil, & Desdicioglu, 2006) and remains relatively stable across development (Hnekopp, Bartholdt, Beier, & Liebert, 2007; Trivers, Manning, & Jacobson, 2006). However, some of the biggest forms of evi dence supporting the idea that 2D:4D ratio is affected by prenatal hormones are studies evaluating digit ratio in individuals with sex hormone disorders. One such disorder is Congeni tal Adrenal Hyperplasia (CAH), where the body produces abnormally high amounts of androgens. Generally, CAH is diagnosed shortly after birth, and is treated to no rmalize these androgens postnatally. Since these hormone levels are normalized either at or shortly after birth, with diagnosis resulting from a salt losing crisis, any difference s between individuals with and without CAH can presumably be attributed to this excess and rogen exposure while still in the womb (rather than hormone exposure after birth). On e of the differences found between females with and without CAH is their 2D:4D ratio: females with CAH tend to have lower, more male-typical, digit ratios than related females without CAH (Brown, Hines, Fane, & Breedlove, 2002; Ciumas, Hirschberg, & Savi c, 2009; kten, Kalyoncu, & Yari, 2002). In other words, females with CAH have more male-typical digit ratios; but it is unlikely that genetics have created this diff erence, because even related females have higher digit ratios than the CAH females. This stro ngly suggests that the lower 2D:4D
25 ratio is due to the excess androgen exposure in the womb. A similar experiment of nature includes complete androgen insensitivity syndrome in males. As the name of the disorder impl ies, androgen insensitivity syndrome is a disorder characterized by a lack of sensitivit y to androgens in the body, resulting in various feminized traits. Berenbaum, Bryk, Nowak, Q uigley, and Moffat (2009) found that males with androgen insensitivity syndrome had higher, more female-typical 2D:4D ratios than controls. This, taken in conjunction wi th more male-typical 2D:4D ratios in females with CAH, is suggestive of the fact that 2D :4D ratio is influenced by prenatal androgen exposure. Another form of evidence suggestive of th is are studies that look at maternal habits and testosterone levels during pregnancy. For examp le, Rizwan, Manning, and Brabin (2006) found that children of mothers who smoked du ring pregnancy which has been linked with higher fetal testosterone levels have lower 2D:4D ratios than children of mothers who did not smoke during pregnancy. Further more, mothers who have higher testosterone levels tend to produce children with l ower 2D:4D. Specifically, the higher the waist-to-hip ratio of the mother (an indicator of higher testosterone levels), the lower the 2D:4D ratio of her children (Manning, Trivers, Singh, & Thornhill, 1999). Furthermore, a study in which routine amniocentesis was performed, found that high levels of fetal testosterone relative to fetal estr ogen predicted lower 2D:4D ratios in the offspring (Lutchmaya, Baron-Cohen, Raggatt, Knickme yer, & Manning, 2004). Thus, the above studies indicate that higher testosterone lev els during pregnancy raise the likelihood of a lower 2D:4D ratio. Lastly, a multitude of studies have found associations between 2D:4D ratio and an
26 assortment of outcomes, many of which are thought t o be primarily male or femaletypical. Examples include lower digit ratio predict ing success in sports (Hnekopp & Schuster, 2010), sensation seeking (Fink, Neave, La ughton, & Manning, 2006), success in financial short-term trading (Coates, Gurnell, & Rustichini, 2009), and spatial ability (Kempel, Gohlke, Klempau, Zinsberger, Reuter, & Hen nig, 2005; Loehlin, Medland, & Martin, 2009), to name a few. Thus, taken as a whole, all of the above is strongl y suggestive of the notion that 2D:4D ratio is representative of prenatal sex hormo ne exposure, particularly testosterone. Assuming this is the case, a correlation of 2D:4D r atio with humor type preference would indicate that at least one innate, biological force namely, prenatal sex hormone exposure is at least partly responsible for the d ifference between men and women in their humor type preferences. Relationships Between Digit Ratio and the 4 Humor T ypes (Aggressive, SelfDeprecating, Affiliative, and Self-Enhancing): Because no studies have been conducted measuring th e relationship between digit ratio and humor type use (and, therefore, cannot be reviewed), this section cites studies that looked at the relationship between digit ratio and traits similar to the humor types (such as looking at aggression for aggressive humor ). However, because studies of this sort only exist for traits parallel to aggressive h umor, the other three humor types are not included in this section. Any analysis of the other three types of humor is completely exploratory. Aggressive humor. In men. Hampson, Ellis and Tenk (2008) studied the relationship between 2D:4D ratio and aggression in 87 undergraduate men. They
27 measured participants 2D:4D ratios, for both the r ight and left hands, then administered the Buss and Perry Aggression Questionnaire (1992). This measures four components of aggression physical aggression, verbal aggression anger, and hostility with a separate subscale for each. A total aggression score can be calculated by summing the scores of all four components, though the scores of individual su bscales can be (and are) used as well. Results showed that verbal aggression was negativel y predicted by both right and left hand 2D:4D ratios, but that physical aggression, an ger, hostility, and total aggression were not predicted by either right or left hand 2D: 4D ratio. In other words, the lower (or more male-typical) the 2D:4D ratio, the greater the verbal aggression score. A similar online study conducted by Hnekopp (2011) looked at the relationship between 2D:4D ratio and aggression levels in 1,254 males. They had participants measure and report their own 2D:4D ratios, and then complete the scales for verbal and physical aggression from the Buss and Perry (1992) Aggression Questionnaire. Results showed a weak negative correlation between left-han d 2D:4D ratio and verbal aggression, but no correlation between 2D:4D ratio and physical aggression, in line with Hampsons and colleagues (2008) study (except for the lack of significance for the right hand 2D:4D ratio). In other words, both studies fou nd that a lower, more male-typical 2D:4D ratio predicted verbal aggression. Thus, base d on these studies, it seems likely that 2D:4D ratio would negatively correlate with aggress ive humor use, at least in men, as humor is a verbal phenomenon. Somewhat in contrast to Hampson and colleagues (20 08) and Hnekopps (2011) findings, however, Bailey and Hurd (2005) did not f ind a correlation between 2D:4D ratio and verbal aggression in men. Rather, after a dministering the Buss and Perry (1992)
28 Aggression Questionnaire to 149 males, as well as t he Paulhus Deception Scale (PDS) to determine who was likely to have answered in a soci ally desirable manner and eliminating them, they found that right hand digit ratio predicted physical aggression, but not verbal aggression, hostility, or anger. They di d not present results for the left hand digit ratio, the reason for this being cited as hav ing found greater sexual dimorphism for the right hand than the left. Their lack of analysi s of the left hand 2D:4D ratio may partially explain the inconsistent results between these studies, specifically results on verbal aggression, considering that Hnekopp (2011) only found a significant correlation between digit ratio and verbal aggression for the l eft hand. However, it is more unclear as to why Bailey and Hurd (2005) found significance fo r physical aggression at all, while Hampson and colleagues (2008) and Hnekopp (2011) d id not. It is possible that eliminating participants who answered in a socially desirable manner contributed to these inconsistencies, though the exact mechanism by whic h they could have done so is also unclear. While it may be possible that the tendency to respond in a socially desirable manner correlates with digit ratio, Bailey and Hurd did not run this particular analysis, and no literature exists on the subject. Kuepper and Hennig (2007) used a somewhat different approach to studying the relationship between 2D:4D ratio and aggression. Sp ecifically, they considered the possibility that self-report measures could yield d ifferent results than behavioral measures. This could be the case for several reason s. For one, it is possible that people are less self-aware than they think, and may thus a ttribute certain characteristics to themselves that, in reality, do not accurately desc ribe them. It is also possible that certain people will alter their responses to appear more so cially desirable, as was potentially the
29 case in Bailey and Hurds study and which led them to eliminate certain participants. Therefore, in their study, in addition to measuring right and left hand digit ratios, Kuepper and Hennig (2007) had 98 male participants complete the aggression scale of the Freiburger Persnlichkeitsinventar (FPI-R; Fahr enberg, Hampel & Selg, 1984), a selfreport measure, as well as a modified version of th e Taylor aggression paradigm (Bond & Lader, 1986; Taylor, 1967), a behavioral measure of aggression. In this modified version, participants were told they were competing in a rea ction-time task against an opponent (who, in reality, did not exist), in which the aim of the game was to press a button as quickly as possible after a red square appeared on the computer screen. The loser would receive a blast of noise preset by their opp onent. There were 25 trials total, each consisting of four steps. The first step was to set the noise level of the blast that would be received by the opponent should the opponent lose; this could range from Level 0 (no noise) to Level 10 (the loudest.) The second step w as to set the duration of the blast that would be received by the opponent. The third step i nvolved the actual task, followed by a blast of noise if the participant lost. The fourth step involved disclosure of the noise level (but not the duration) set by the participants opp onent, regardless of whether they had won or lost that round. Thus, the noise level chose n by the participant was used as a measure of overt aggression, since the participant chose this level knowing it would be disclosed to their opponent, whereas the noise dura tion chosen by the participant was used as a measure of covert aggression, since the p articipant chose the duration knowing it would not be disclosed to their opponent. Interestingly enough, only the noise leve l chosen by the participant was correlated with aggression scores on the FPI-R; this was not t he case with the noise duration chosen
30 by the participant, indicating that both types of m easures may very well represent different types of aggression (and possibly account for inconsistencies in the literature). With respect to both right hand and left hand 2D:4D ratio, results showed no correlation with the FPI-R aggression subscale. However, result s did show a correlation between various aspects of the behavioral aggression measur e and both right hand and left hand 2D:4D ratio. Mean noise level and duration, first n oise duration, second noise level and duration, and the amount of Level 10 noise levels a ll correlated negatively with 2D:4D ratio. Thus, while no relationship was found betwee n digit ratio and aggression when the self-report scale was used, a multitude of signific ant relationships were found between digit ratio and aggression (both overt and covert) when the behavioral measure was used. This finding carries a great amount of weight, as i t supports the notion that self-report measures may not always yield entirely accurate dat a, possibly skewing results of other studies (Hampson, Ellis, & Tenk, 2008; Hnekopp, 20 11; Bailey & Hurd, 2005) that did rely solely on self-report data. Furthermore, this study provides greater reason to believe that 2D:4D ratio may correlate with an aggressive h umor style, at least in men. Another different, but enlightening, appr oach taken toward determining the relationship between 2D:4D ratio and aggression was used by Millet and Dewitte (2007). They speculated that context cues (specifically agg ressive ones), defined as cues dependent on the environment and situation, might m ediate the relationship between 2D:4D ratio and aggression, possibly accounting for inconsistencies in the literature. They exposed 47 males to either an aggressive or no n-aggressive music video, and then administered both the Aggression Provocation Questi onnaire (OConnor, Archer, & Wu, 2001) and the physical aggression scale of Buss and Perrys (1992) Aggression
31 Questionnaire and summed these scores to obtain one physical aggression score. Following this, they measured the right hand 2D:4D ratio. Interestingly enough, results showed that 2D:4D ratio was negatively correlated w ith aggression scores, but only in the aggressive music video condition; in the non-aggres sive music video condition, no trend emerged. In other words, the relationship between d igit ratio and aggression only became apparent after participants had been exposed to an aggressive context cue. This is something that should be kept in mind when looking at other studies on the relationship between 2D:4D ratio and aggression, particularly wh en inconsistencies are encountered. Although many studies found a negative re lationship between male 2D:4D ratio and aggression levels, a few studies found no relat ionship between the variables. One such study was conducted by Benderlioglu and Nelson (2004). They measured digit ratio and directional asymmetry (DA), defined as the diff erence between the left and right digit ratios, in 51 male university students, and measure d (as control variables) testosterone levels, age, handedness, height, and weight. Two se parate measures of aggression were used. In the first, participants were asked to call two male confederates in an attempt to persuade them to donate money to a charity, and par ticipants were told they would receive two free movie tickets if they managed to r aise any money. However, both confederates refused to donate any money; the first showed interest in the charity but stated that he had no money, and the second insulte d the charity and refused to donate out of hostility. This first call was considered the l ow provocation condition, while the second call was considered the high provocation c ondition; all participants experienced both conditions. The force the participant used to hang up the phone in each condition, used as a marker of aggression, was measured using a balance plate built into the desk.
32 The second measure of aggression involved having th e participant anonymously choose one of three pre-written follow up letters to send to the confederates, with the first letter being the least aggressive and the third letter bei ng the most aggressive. Results showed that, regardless of condition, no re lationship was found between male 2D:4D ratio or directional asymmetry and eithe r measure of aggression. The authors cite one possible reason for this as having too sma ll a sample size, as there may not have been enough statistical power to generate significa nce. Furthermore, they note that this particular sample was somewhat unusual, as males an d females were not found to differ significantly in their digit ratios (in contrast to most studies). Lastly, they speculate that the lack of significance in males may have been leg itimate, and due to a ceiling effect, where all males may have received a sufficient amou nt of androgen exposure in the womb to elicit aggression in a more uniform manner than in females, making it more difficult to detect differences in aggression betwe en males. Even if this is the case, however, previous studies still stand in contrast t o this one. In support of Benderlioglu and Nelsons (2004) find ings but having used different measures of aggression, Voracek and Stieger (2009) measured the relationship between both implicit and explicit aggression and 2D:4D rat io in two separate samples, the first consisting of 114 men and the second consisting of 101 men. To measure explicit aggression, they had participants complete the Buss and Perry Aggression Questionnaire as well as an aggression feeling thermometer (where participants rated their aggression level on a scale from 1-100). To measure implicit a ggression, an Implicit Association Test (IAT) was administered. Digit ratio was measur ed using scans of both the right and left hand. Results showed no association between di git ratio and either implicit or explicit
33 aggression. However, unlike Benderlioglu and Nelson (2004), Voracek and Stieger (2009) found, in both samples, that men had smaller 2D:4D ratios than women, rendering these particular samples more typical than Benderli oglu and Nelsons sample. Therefore, it is unlikely that the lack of association emerged as a result of an atypical sample, at least in this particular sense. It must be noted, however that Voracek and Stieger found no significant sex differences between men and women i n either implicitly or explicitly measured aggression, except for finding that men sc ored significantly higher on physical aggression on the BPAQ than women did in Sample 1 b ut not in Sample 2. This is atypical considering the usual trend for men to be more aggressive than women, and may possibly explain the lack of relationship between 2 D:4D ratio and aggression, especially considering all of the studies that did show a rela tionship between those variables. On the other hand, it could signify that 2D:4D ratio may n ot be the best marker if there is a biological component to aggression. In women. In addition to studying the relationship between di git ratio and aggression in men, Hampson and colleagues (2008) al so looked at the relationship between digit ratio and aggression in 77 undergradu ate women. Again, they measured participants 2D:4D ratios, for both the right and left hands, then administered the Buss and Perry Aggression Questionnaire. Results showed that, in the women, a lower right hand digit ratio predicted total aggression as well as three measures of aggression: physical aggression, anger, and hostility. However, the relationship between 2D:4D ratio and verbal aggression merely approached significanc e ( p = .089). This stands in stark contrast to the results obtained for the men, for w hom only verbal aggression was negatively correlated with 2D:4D ratio.
34 Benderlioglu and Nelson (2004) also studied the rel ationship between digit ratio and aggression in women. Again, they measured digit ratio and directional asymmetry in 49 female university students (controlling for sali vary testosterone levels, age, handedness, height, and weight) and used two separa te behavioral measures of aggression. Participants made charity calls and wer e presented with first a mildly provoking response followed by a highly provoking r esponse. The force used in hanging up and the tone of the follow-up letter were measur ed. Unlike the non-significant results found for males, results for females showed that, in the high provocation condition, DA was neg atively correlated with reactive aggression as measured by the force used to hang up the phone, and that 2D:4D ratio was negatively correlated with reactive aggression as m easured by the tone of the follow up letters. In other words, the more masculine the 2D: 4D ratio in females, the greater the amount of reactive aggression. It is interesting to note that a significant effect was found in women, but not in men. Again, the authors specul ate that the lack of significance in males may have been due to a ceiling effect, wher e all males may have received a sufficient amount of androgen exposure in the womb to elicit aggression in a more uniform manner than in females, making it more diff icult to detect differences in aggression between males than in females. Either wa y, however, in this study, females were shown to demonstrate greater aggressive tenden cies when they had a more masculine digit ratio, in support of results regard ing females in Hampson and colleagues (2008) study. It is also interesting to note that n o significant results were found for the low provocation condition, demonstrating the possib le importance of context cues in making the relationship between 2D:4D ratio and agg ression visible.
35 Coyne, Manning, Ringer, and Bailey (2006) studied t he relationship between digit ratio, directional asymmetry (defined as the right digit ratio minus the left digit ratio) and both direct and indirect humor, in women only. They had 100 undergraduates complete the BPAQ as a measure of direct aggression, and the Indirect Aggression Questionnaire (Forrest, Eatough, & Shevlin, 2005) as a measure of indirect aggression. The IAQ asks participants to rate the extent to which each state ment is characteristic of themselves on a scale of 1 to 5, but only for behavior that has occ urred in the last 12 months. It measures three different types of indirect aggression soci al exclusion, malicious humor, and guilt induction, and scores on these subscales can be com bined to form a composite indirect aggression score (InAg). Results showed that lower values of DA (more male typical) predicted all types of indirect aggression, but not direct aggression. However, no relationships were found between either right or le ft hand digit ratio and direct or indirect aggression. Some studies found no relationship between 2D:4D ra tio or DA and aggression in females. Kuepper and Hennig (2007), for example, al so included females in their study. Again, they had all participants complete the FPI-R as well as a modified version of the Taylor paradigm. Unlike males, females demonstrated no relationship between digit ratio and either the self-report (FPI-R) or behavioral (T aylor paradigm) measure of aggression. Furthermore, Bailey and Hurd (2004) also included f emales in their study, which consisted of completing the FPI-R. Unlike males, fe males demonstrated no relationship between digit ratio and self-reported aggression. L astly, Voracek and Stieger (2009) also included females in their study, which consisted of measuring implicit and explicit
36 aggression, and found that, as was the case in male s, 2D:4D ratio in females did not correlate with either measure of aggression. Millet and Dewitte (2007) also looked at the relati onship between digit ratio and aggression in women, in addition to men. Again, the y speculated that aggressive context cues might mediate the relationship between 2D:4D r atio and aggression, possibly accounting for inconsistencies in the literature. U sing the same experimental setup for 49 females as they did with the males, they found that in contrast to the results for males, female 2D:4D ratio did not predict aggression in ei ther the aggressive or non-aggressive music video conditions. In other words, for women, the relationship between digit ratio and aggression was never apparent, even after being exposed to an aggressive context cue. Thus, it is possible that differences between the s exes in the type of humor they use may be accounted for, at least in part, by biol ogical factors. Whereas some studies have shown that there is a biological component to humor preferences, none of these in particular have taken participant sex into account, and studies that have taken participant sex into account examined biological measures that could fluctuate throughout a persons life and, thus, cannot imply causation. Furthermore no studies have looked at the possible influence of prenatal hormone exposure on humor preferences, let alone on differences between the sexes in their humor prefer ences. 2D:4D ratio may be used at least partly as a way of determining the effect of prenatal hormone exposure on humor preferences. As no studies exist that examine the r elationship between any of the humor types with 2D:4D ratio, the relationship between 2D :4D ratio and parallel traits was reviewed, and findings were mixed. However, the pos sibility that 2D:4D ratio and gender
37 role are related which would blur the line betwee n sociocultural and biological influences on humor type still remains. The Relationship between Gender Role and 2D:4D Rati o Csath, Osvth, Bicsk, Kardi, Manning, and Kllai (2003) looked at the relationship between 2D:4D ratio and gender role in 46 women. Participants completed the Bem Sex Role Inventory (BSRI). Results showed t hat lower (more masculine) 2D:4D correlated positively with masculine scores on the BSRI, indicating that male-typical 2D:4D ratios predict masculine gender roles in wome n. It is interesting to note, however, that higher (more feminine) 2D:4D ratios did not co rrelate positively with feminine scores on the BSRI. Regardless, this study has impo rtant implications, as it indicates the possibility that 2D:4D ratio and gender role scores may co-vary, and must therefore be analyzed using regression to untangle each ones po ssible contribution to humor type preference. The Current Study Based on previous research, it is clear that men an d women exhibit different preferences toward the type of humor they use in th eir daily lives, with men tending to use aggressive humor more than women and women tend ing to use cohesive humor more than men (Greengross & Miller, 2008; Martin et al., 2003; Robinson & Smith-Lovin, 2001). However, no studies to date have looked at t he extent to which biological and sociocultural factors account for these differences Achieving an understanding of gender differences in humor styles is important both theor etically and practically, and it is also important to understand the ways in which social an d biological factors could mask one another in this context.
38 As such, the current study attempted to dem onstrate the extent to which gender role accounts for differences in humor type preferences. In addition to the participants gender role, this study examined one biological factor s pecifically, 2D:4D ratio that also possibly accounts for these differences in humor ty pe preferences between men and women. Again, 2D:4D ratio can be used as a measure of biological influence on gendertypical behaviors. 2D:4D ratio is defined as the ra tio between the length of the index and ring finger, which is putatively determined by (and therefore representative of) sexually organizing effects of testosterone exposure on the fetuss developing brain (Romano et al., 2005; Voracek et al., 2007). In other words, 2 D:4D ratio is a putative marker of inherent biological differences between the sexes, already present at birth. However, as of today, no studies have correlated 2D:4D ratio with humor type preference, rendering such a study necessary. On the other hand, gender role, defined as a persons preference for/adoption of behavioral characteristics or endorsement of person ality traits that are linked to cultural notions of masculinity and femininity, can be used as a measure of sociocultural influence on gender-typical behaviors (Yarber et al ., 2010). While some studies have correlated gender role with humor type preference ( Martin et al., 2003; Mio, 2009), none have looked at possible interactions with biologica l factors. Due to these problems, a study that evaluated possible interactions between biological and sociocultural factors was needed. As previously stated, then, the present study attem pted to demonstrate the extent to which one biological factor, 2D:4D ratio, and on e sociocultural factor, gender role,
39 individually account for differences between men an d women in humor type preference, as well as each factors interaction with the other s. To do so, university students and faculty members completed the Humor Styles Questionnaire (HSQ; Martin et. al., 2003), and the short-form revised Bem Sex-Role Inventory (BSRI-R; Bem, 1978). The HSQ revealed the participants own personal humor style usage, while the BSRI-R revealed the pa rticipants gender role. Following completion of these questionnaires, a photocopy of the participants left and right hand was obtained, from which each hands respective 2D: 4D ratio was later measured. There were several anticipated outcomes a s well as some questions: 1) Aggressive humor: a. In line with Martin and colleagues (2003) and Greengross and Miller (2008), it was hypothesized that overall, men would demonstrate mo re aggressive humor use than women. b. Because the literature is inconsistent rega rding a correlation between 2D:4D ratio and aggression, (Hnekopp, 2011; Kuepper & Hennig, 2007; Bailey & Hurd, 2005; Millet & Dewitte, 2007; Hampson et al., 2008; Butov skaya et al., 2010; Voracek & Stieger, 2009; Benderlioglu & Nelson, 2004; Coyne e t al., 2007), I explored whether a relationship existed between 2D:4D ratio and aggres sive humor style. However, if a relationship was found, it was hypothesized to be a negative one, such that a lower, more male-typical digit ratio would predict greater aggr essive humor use. c. Because a more male-typical gender role or ientation predicts aggressive humor use (Mio, 2009; Martin et al., 2003), it was hypothesiz ed that there would be a negative correlation between gender role score and aggressiv e humor style, such that a lower,
40 more male-typical gender role score predicts greate r use of aggressive humor. d. As results in the literature are inconsiste nt for correlations between 2D:4D ratio and gender role (Csath et al., 2003; Rammsayer & Troch e, 2007; Troche et al., 2007), I explored whether the relationship between gender ro le and aggressive humor would decrease significantly once the other factors (2D:4 D ratio) were held constant. 2) Self-deprecating humor: a. I explored whether men would differ from w omen in their self-deprecating humor use, as results in the literature have been contrad ictory (Martin et al., 2003; Greengross & Miller, 2008). However, if they did, it was hypothe sized that men would demonstrate more self-deprecating humor use than women. b. Furthermore, I explored whether a correlati on would exist between 2D:4D ratio and self-deprecating humor use. If a relationship did e xist, it was hypothesized that a lower, more male-typical digit ratio would predict greater use of self-deprecating humor. c. Because a more masculine gender role and a less feminine gender role predicts self-deprecating humor use (Martin et al., 2003), i t was hypothesized that there would be a negative correlation between gender role score an d self-deprecating humor style, such that a lower, more male-typical gender role score p redicts greater use of self-deprecating humor. 3) Affiliative humor: a. I explored whether men would differ from women in their affiliative humor use, as results in the literature have been contradictory ( Martin et al., 2003; Greengross & Miller, 2008; Robinson & Smith-Lovin, 2001). b. I also explored whether a correlation would ex ist between 2D:4D ratio and affiliative
41 humor use. c. I explored whether gender role would predict a ffiliative humor use. 4) Self-enhancing humor: a. I explored whether men would differ from women in their self-enhancing humor use. b. Furthermore, I explored whether a correlation would exist between 2D:4D ratio and self-enhancing humor use. c. Because a more masculine gender role and a les s feminine gender role predicts selfdeprecating humor use (Martin et al., 2003), it was hypothesized that there would be a negative correlation between gender role score and self-enhancing humor style, such that a lower, more male-typical gender role score predic ts greater use of self-enhancing humor. Method Participants University students and faculty members (24 males, 28 females) in southwest Florida, ages 19 68, were recruited on the univer sity campus. Only right-handed, heterosexual, Caucasian individuals were included i n the analysis, for the purpose of maintaining homogenous data1, resulting in a final sample of 16 males and 19 fe males. Materials Participants completed three different su rveys, the Humor Styles Questionnaire (HSQ; Martin, Puhlik-Doris, Larsen, Gray, & Weir, 2 003), the short-form Bem Sex-Role Inventory Revised (BSRI-R; Bem, 1977), and a demo graphic questionnaire. Humor Styles Questionnaire. The first survey participants completed was the Humor Styles Questionnaire (Martin et al., 2003), a self-report measure used to
42 determine the extent to which participants utilize four possible humor styles in daily life: Affiliative humor, Self-enhancing humor, Aggressive humor, and Self-deprecating humor. The questionnaire consisted of four differen t subscales one per humor type with 8 items in each subscale for a total of 32 ite ms. Each item was rated on a Likert scale of 1-7, with a 1 indicating the trait did not at all describe the participant and a 7 indicating the trait strongly described the partici pant. Sample questions include Even when Im by myself, Im often amused by the absurdi ties of life and If someone makes a mistake, I will often tease them about it. (The entire HSQ can be found in Appendix A). Sum scores were calculated separately for each subs cale (please refer to Appendix A for items in each subscale). Scores for each subs ection of the HSQ were determined by adding the values of non-reverse keyed items to rev erse keyed items, yielding a score ranging from 8 to 56 for each subscale. Thus, a hig her score indicated greater use of that particular humor style by the participant. Bem Sex-Role Inventory Revised. The second survey participants completed was the short-form Bem Sex-Role Inventory Revised (BSRI-R; Bem, 1977), a revision of the original Bem Sex-Role Inventory (BSRI; Bem, 1974). The short-form BSRI-R is a self-report measure consisting of 30 personality ch aracteristics on which participants are to rate themselves, using a Likert scale from 1 (ne ver or almost never true) to 7 (always or almost always true). Of these items, 10 are cons idered to be stereotypically masculine (comprising the masculine subscale), 10 are stereot ypically feminine (comprising the feminine subscale), and the remaining 10 are filler items. In the current study, rather than classifying participants into groups, continuous Tscores were created to indicate how
43 male-typical or female-typical each participant was with a score of 12 indicating most male-typical and a score of 88 indicating most fema le-typical. This was done by subtracting each participants standard masculinity scale score from his or her standard femininity scale score to obtain the standardized d ifference score, which was then transformed into a standardized T-score. Demographic questionnaire. The final questionnaire completed was a demographic questionnaire. Participants were asked to provide their gender, age, sexual orientation, race/ethnicity, and handedness. 2D:4D ratio materials. In addition to completing the surveys, participants also had their left and right hands photocopied using a portable HP Photosmart C4795 photocopier. A vernier caliper was used to measure the length of the index finger and the length of the ring finger for each hand, from the b asal crease to the tip of the finger. The 2D:4D ratio was obtained by dividing the length of each hands index finger by the length of the ring finger, with a lower ratio (long er ring finger than index finger) being more typical of males. Procedure Participants were recruited in several ways. For so me, an announcement was made during their class regarding the purpose of the stu dy, and a sign-up sheet was passed around, which the experimenter used to contact inte rested students. For others, an announcement was made during their class regarding the purpose of the study, and flyers were passed around so that interested students coul d contact the experimenter. Additionally, flyers were posted around the univers ity so that interested individuals could contact the experimenter. Lastly, the experimenter set up a table in a central location in
44 the school, put up signs on the table stating that individuals interested in partaking in a humor study could do so right then and there, and a llowed individuals to approach the table of their own accord without being prompted by the experimenter. Upon the participants completion of the informed consent form, a packet containing the HSQ, the short-form BSRI-R, and demo graphic questions was then administered and participants could take as much or as little time as they wanted to complete it. The HSQ was the first survey they saw, followed by the short-form BSRI-R and then the demographic section. This order was ch osen to prevent any possible effect that answering questions about gender role could ha ve on HSQ responses. Again, there was no time limit for this questionnaire. Following completion of the surveys, a photocopy wa s obtained of the participants left and right hands (as done by Bail ey & Hurd, 2005; Hampson et al., 2008; Kuepper & Hennig, 2007; Millet & Dewitte, 2007; etc ). Both surveys and the photocopy were labeled with the same coded ID so they could b e matched later with no master file linking the participants name to them. These were then were placed in a sealed envelope to preserve anonymity. Finally, participants were d ebriefed, were told not to discuss the study with others, and were compensated with a $5 S tarbucks gift card. Additionally, students from one of the classes signed their name on a single sheet acknowledging they had completed the experiment, which was to be used by the professor to award extra credit. Results First, t-tests were run to determine over all sex differences, including differences in humor style, gender role, and digit ratios. Followi ng this, zero-order correlations were
45 determined separately within males and females for the variables in question. Finally, regression analyses were run on the sample as a who le. Although no predictions were made regarding age, it was included in the correlat ion and regression analyses. See Table 1 for complete descriptive statistics on humor styl es, digit ratios, and gender role scores. Overall Sex Differences Several t-tests were conducted to determi ne differences in humor style, gender role, and digit ratios between the sexes. Humor style. As predicted, there was a significant effect of sex on aggressive humor, t (33) = -2.27, p = .03, with males ( M = 29.31, SD = 11.09) receiving higher scores than females ( M = 22.26, SD = 7.19). There was no effect of sex on self-deprecating humo r, t (33) = -1.44, p = .16, with males ( M = 30.19, SD = 8.05) and females ( M = 25.63, SD = 10.30) not scoring significantly differently. However, because I notic ed that a particularly helpful participant was sending coworkers my way toward the end of my s tudy, analyses were run to ensure that the smaller, untainted sample ( n =20 sample) was similar to the full sample which likely contained a number of non-independent observ ations ( n =35). Substantial differences were found, such that in the n =20 sample, a significant effect for selfdeprecating humor was found, t (18) = -2.22, p = .04, with males ( M = 32.18, SD = 8.73) receiving higher scores than females ( M = 23.11, SD = 9.52). There was no effect of sex on affiliative humor, t (33) = -1.37, p = .18, with males ( M = 49.38, SD = 5.26) and females ( M = 46.21, SD = 7.87) not scoring significantly differently.
46 Likewise, there was no effect of sex on self-enhanc ing humor, t (33) = -0.96, p = .34, with males ( M = 44.81, SD = 5.27) and females ( M = 42.68, SD = 7.41) not scoring significantly differently. Gender role. There was a significant effect of sex on gender rol e, t (33) = -2.22, p = .03, with males ( M = 49.75, SD = 11.66) receiving lower, more male-typical scores than females ( M = 57.58, SD = 9.19). Digit ratios. There was a significant effect of sex on right digi t ratio, t (33) = 2.74, p = .01, with males ( M = 0.964, SD = 0.033) demonstrating lower, more male-typical ratios than females ( M = 0.998, SD = 0.039). However, no significant effect was found for left d igit ratio, t (32) = 0.49, p = .63, with males ( M = 0.9691, SD = 0.0397) and females ( M = 0.9750, SD = 0.0300) not scoring differently. Zero-Order Correlations Correlations were run separately for males and fema les (see Tables 2 and 3 for intercorrelations.) The primary focus was to determ ine whether digit ratios or gender role could predict humor style. Additionally, correlatio ns were run to determine whether digit ratios could predict gender role. Predictors of humor styles. In males. Right 2D:4D ratio did not predict use of aggressive humor in males, r (14) = -0.32, p = .23. However, the relationship between left 2D:4D and aggressive humor approached significance with a strong negative correlation, r (14) = -0.50, p = .06. In other words, the lower and more male-typ ical the left 2D:4D ratio, the greater the use of aggressive humor. In contrast with what was expected, gender role score did not predict aggressive humor use, r (14) = -0.28, p = .30.
47 Neither right 2D:4D ratio, r (14) = 0.00, p = 1.0, nor left 2D:4D ratio, r (14) = -0.35, p = .21, predicted use of self-deprecating humor. Fu rthermore, in contrast with what was expected, gender role score did not predict self-de precating humor use, r (14) = 0.28, p = .29. Likewise, neither right 2D:4D ratio, r (14) = -0.08, p = .78, nor left 2D:4D ratio, r (14) = -0.03, p = .92, predicted use of affiliative humor. Gender role score did not predict affiliative humor use, r (14) = -0.08, p = .78. Lastly, neither right 2D:4D ratio, r (14) = -0.02, p = .94, nor left 2D:4D ratio, r (14) = 0.41, p = .13, predicted use of self-enhancing humor. In c ontrast with what was expected, gender role score did not predict self-en hancing humor use, r (14) = 0.07, p = .80. In females. Right 2D:4D ratio approached significance in predic ting use of aggressive humor in females with a moderate positiv e correlation, r (17) = 0.44, p = .06. Furthermore, the relationship between left 2D:4D an d aggressive humor was significant with a strong positive correlation, r (17) = 0.49, p = .03. In other words, the higher and more female-typical the right and left 2D:4D ratio, the greater the use of aggressive humor. In contrast with what was predicted, gender role score did not predict aggressive humor use, r (17) = -0.23, p = .34. Neither right 2D:4D ratio, r (17) = 0.03, p = .90, nor left 2D:4D ratio, r (17) = -0.08, p = .76, predicted use of self-deprecating humor. Fu rthermore, in contrast with what was expected, gender role score did not predict self-de precating humor use, r (17) = 0.19, p = .44. Neither right 2D:4D ratio, r (17) = 0.31, p = .20, nor left 2D:4D ratio, r (17) = -0.08,
48 p = .75, predicted use of affiliative humor. Likewis e, gender role score did not predict affiliative humor use, r (17) = -0.04, p = .86. Lastly, neither right 2D:4D ratio, r (17) = 0.05, p = .84, nor left 2D:4D ratio, r (17) = -0.22, p = .36, predicted use of self-enhancing humor. In c ontrast with what was expected, gender role score did not predict self-en hancing humor use, r (17) = 0.19, p = .44. Relationship between digit ratio and gender role. In males. Right 2D:4D ratio did not predict gender role score, r (14) = 0.10, p = .71. However, left 2D:4D ratio approached significance in predicting gender role s core, r (14) = -0.46, p = .08. In other words, the lower and more male-typical the digit ra tio, the higher and more femaletypical the gender role score. In females. Neither right 2D:4D ratio, r (17) = -0.13, p = .58, nor left 2D:4D ratio, r (17) = -0.19, p = .43, predicted gender role score. Regression Analyses Linear multiple regressions were run to determine w hether gender role, digit ratios, age, or sex predicted any of the four humor styles while controlling for said variables. The sexes were not analyzed separately due to a sma ll sample size and therefore insufficient statistical power. Furthermore, linear multiple regressions were run to determine whether sex, digit ratios, or age predict ed gender role while including each of these variables in the equation. Finally, linear mu ltiple regressions were run to determine whether sex and age predicted digit ratios. Predictors of humor style. Neither gender role, b = -.20, t (31) = -1.18, p = .25, nor right digit ratio, b = .04, t (29) = .20, p = .84, nor left digit ratio, b = -.08, t (29) = -.46,
49 p = .65, predicted aggressive humor. Furthermore, se x did not predict aggressive humor, b = .17, t (31) = .98, p = .34. Neither gender role, b = .25, t (31) = 1.38, p = .18, nor right digit ratio, b = .05, t (29) = .23, p = .82, nor left digit ratio, b = -.10, t (29) = -.53, p = .60, predicted self deprecating humor. Furthermore, sex did not predict self deprecating humor, b = .26, t (31) = 1.42, p = .17. Neither gender role, b = -.03, t (31) = -.15, p = .89, nor right digit ratio, b = .24, t (29) = 1.12, p = .27, nor left digit ratio, b = -.11, t (29) = -.56, p = .58, predicted affiliative humor. Furthermore, sex did not predict affiliative humor, b = .16, t (31) = .82, p = .42. Neither gender role, b = .08, t (31) = .40, p = .69, nor right digit ratio, b = -.02, t (29) = -.10, p = .92, nor left digit ratio, b = .08, t (29) = .40, p = .69, predicted self enhancing humor. Furthermore, sex did not predict self enhanc ing humor, b = .22, t (31) = 1.10, p = .28. Predictors of gender role. Even controlling for age and digit ratios, sex was a significant predictor of gender role score, b = -.41, t (30) = -2.18, p = .04, with males scoring in a lower, more male-typical direction tha n females. Neither right digit ratio, b = -.06, t (30) = -.30, p = .77, nor left digit ratio, b = -.27, t (30) = -1.57, p = .13, predicted gender role score. Furthermore, age did not predict gender role score, b = -.20, t (30) = 1.14, p = .27. Predictors of digit ratios. After controlling for age, sex still significantly predicted right digit ratio, b = -.48, t (32) = -3.01, p = .01, with females possessing higher right digit ratios than males.
50 Conversely, after controlling for age, se x still did not predict left digit ratio, b = .05, t (30) = -.28, p = .78. Age Age was analyzed in both the zero-order correlation s and the regressions. For the correlations, the female non-systematic (sm aller) sample approached a significant negative correlation with aggressive hu mor, r (7) = -0.62, p = .08, such that younger females exhibit greater use of aggressive h umor. However, the female systematic (larger) sample did not approach signifi cance in this regard, r (17) = -0.38, p = .11. No trend was evident in males. Age did not cor relate with any of the other humor types in either males or females. Age did, however, correlate with right digit ratio in females, r (17) = -0.59, p = .01, and approached a significant relationship with left digit ratio in males, r (14) = 0.45, p = .09. In other words, older females exhibited lowe r, more male-typical digit ratios, whereas older males exhibited higher, more female-t ypical ratios. As for the regressions, age did predict aggressive humor, b = -.41, t (31) = -2.50, p = .02. In other words, as age increased, aggressive humor use decreased. However, age did not predict any of the other humor styles. Discussion Aggressive Humor Thus, it is difficult to draw any solid c onclusions regarding aggressive humor use from this study. This is due primarily to statistic al tests being inconsistent with one another, and to possible issues with the sample its elf (both discussed below).
51 Hypothesis 1: men will demonstrate more aggressive humor use than women. It is unclear whether males and females actually di ffered in their use of aggressive humor. On the one hand, a simple t -test comparing male and female means found that males did, in fact, utilize aggressive humor more t han females (in line with Martin et al., 2003, and Greengross & Miller, 2008). On the other hand, once gender role score, age, right digit ratio, and left digit ratio were all co ntrolled for in the regression analysis, this difference between males and females was no longer found. This disparity could have occurred for a number of reasons. One possible explanation is that males and females really do dif fer in their use of aggressive humor, but there may not have been enough participants, an d, thus, statistical power, to yield statistical significance in the regression analysis ; there were only 35 participants, yet five variables were being controlled for. Another possible explanation for this dis parity is that males and females really do not differ in their use of aggressive humor. It may be that in controlling for a possible confound age, which correlated negatively with ag gressive humor use the regression revealed a lack of significant differences between the sexes. The fact that the mean age of male participants was 40 whereas the mean age of fe male participants was 47 lends additional support to this idea; males may have com e out as more aggressive than females in the t-test because their average age was younger and younger age is correlated with greater use of aggressive humor. Thus, it is unclea r whether these data supported the hypothesis that men would demonstrate greater use o f aggressive humor than women. Considering that no previous studies have found a l ack of significant differences, though,
52 it is likely that the disparity occurred due to the first reason. Still, no conclusions can be drawn without further studies. Non-directional hypothesis 2: does 2D:4D ratio pred ict aggressive humor use? Furthermore, it is unclear whether right or left di git ratios predicted aggressive humor use. On the one hand, there were several sign ificant (or near-significant) zeroorder correlations for both males and females. On t he other hand, once gender role score, age, sex, and both digit ratios were controlled for these relationships between digit ratios and aggressive humor use were no longer found. In males. For the zero-order correlation, although neither ri ght nor left digit ratios predicted aggressive humor use, left digit ratio ap proached significance in the hypothesized direction, with more male-typical rati os being strong predictors of greater aggressive humor use. It is not unlikely that havin g a sample size larger than the current 14 degrees of freedom would have made this finding a significant one. On the other hand, once gender role score, age, sex, and digit ratio w ere controlled for, neither right nor left digit ratio even approached significance in predict ing aggressive humor use. Again, one possible explanation for this disparity is that left digit ratio does predict aggressive humor use, but that the sample s ize was merely too small to obtain significance in the regression. However, it is also possible that left digit ratio really does not predict aggressive humor use, and that age, or gender role, was confounding the correlation results. This may be supported by the f inding that left digit ratio in males increased with age (which is problematic in and of itself; see section on Overarching Limitations below) and older age is associated with lesser use of aggressive humor. It may also be supported by the finding that gender ro le score and left digit ratio were
53 almost significantly moderately correlated, such th at a more male-typical left digit ratio was predictive of a more female-typical gender role score. Thus, it is unclear whether digit ratios were actually predictive of aggressive humor use in males. If they were, then that would indicate that aggressive humor use has a t least some biological specifically, prenatal hormonal component to it; if they were n ot, then that would indicate that aggressive humor does not have a prenatal hormonal component to it (though this does not exclude other possible biological components). In females. For the zero-order correlation, left digit ratio wa s strongly predictive of aggressive humor use in the opposite direction o f what was hypothesized, with a higher, more female-typical digit ratio predicting greater use of aggressive humor. Furthermore, right digit ratio approached significa nce in predicting aggressive humor use, also in the opposite direction of what was hypothes ized. This finding likely would have reached significance with a greater sample size. On the other hand, once gender role score, age, sex, and digit ratio were controlled fo r, neither right nor left digit ratio even approached significance in predicting aggressive hu mor use. Again, one possible explanation for this discrepanc y is that left and right digit ratio do predict aggressive humor use, but that the sample size was merely too small to obtain significance in the regression. However, it is also possible that both digit ratios really do not predict aggressive humor use, and tha t age was confounding the correlation results. This may be supported by the finding that right digit ratio in females decreased with age (which is problematic in and of itself; se e section on Overarching Limitations below) and older age is associated with lesser use of aggressive humor. However, this would not explain the discrepancy for left digit ra tio. Thus, it is unclear whether digit
54 ratios were actually predictive of aggressive humor use in females. If they were, then that would indicate that aggressive humor use has at lea st some biological specifically, prenatal hormonal component to it; if they were n ot, then that would indicate that aggressive humor does not have a prenatal hormonal component to it (though this does not exclude other possible biological components). Hypothesis 3: a more male-typical gender role score will predict greater use of aggressive humor. In males. This hypothesis was not supported by the data, with no relationship being found between male gender role s core and aggressive humor use (in contrast with Mio, 2009; Martin et al., 2003). Howe ver, the direction of the effect was in the hypothesized direction; perhaps a greater sampl e size would have yielded significance. In females. Likewise, no relationship was found between gender role score and aggressive humor use. However, as was the case with males, the direction of the effect was in the hypothesized direction, with a more male -typical gender role score being predictive of greater use of aggressive humor. Thus, while it is likely safe to assume that there is at least some social influence on aggressive humor use, this study did not pick up on it. The most probable explanation seems to be that gender role score does predict agg ressive humor use, but that the sample size was too small to yield the necessary statistic al power, especially since previous studies have found a relationship between these var iables (Mio, 2009; Martin et al., 2003). On the other hand, it may be that this gende r role questionnaire was not sensitive enough, though this is unlikely considering that th e sexes scored differently on it. More likely is the possibility that only certain facets of masculinity or femininity are associated
55 with aggressive humor use, especially seeing as Mar tin and colleagues found significance when looking at both positively and negatively vale nced masculinity and femininity, whereas the BSRI-R does not take positive and negat ive valence into account. Another possibility is that only other non-gendered sociocu ltural factors might influence aggressive humor, or even that familial, rather tha n sociocultural, values might influence aggressive humor use. However, this is unlikely con sidering Martin and colleagues significant gendered findings. Self-Deprecating Humor Non-directional hypothesis 4: do men and women diff er in their use of selfdeprecating humor? As was the case with aggressive humor, it is unclea r whether males and females differ in their use of self-deprecating humor. This uncertainty primarily stems from major differences between the N=35 sampl e and N=20 sample. In the former, no significant difference was found between men and women in their use of selfdeprecating humor, in line with Greengross and Mill er (2008). However, in the latter (which may have been a more representative sample; see section on Overarching Limitations below), males scored significantly high er on their use of self-deprecating humor than women, in line with Martin and colleague s (2003). This is especially striking considering the smaller sample size, as it is more difficult to obtain significance with such a size. Thus, it is difficult to draw any soli d conclusions regarding differences between males and females in their use of self-depr ecating humor without conducting further studies using larger and more representativ e samples. Non-directional hypothesis 5: does 2D:4D ratio pred ict self-deprecating humor use? In males. Neither right nor left digit ratio predicted self-d eprecating humor
56 use, although a greater sample size may have yielde d a significant relationship between left digit ratio and self-deprecating humor use (as the left p-value was lower than the right p-value by about 0.8 significance points, and the m ajority of digit ratio correlations in males were with the left one). This trend was in th e hypothesized direction, with a more male-typical left digit ratio being predictive of g reater self-deprecating humor use. Ultimately, however, it is difficult to draw any so lid conclusions here without conducting further studies using a larger sample size. By exte nsion, it is difficult to say whether exposure to prenatal hormones is predictive of self -deprecating humor use in males. In females. Digit ratio did not appear to predict self-depreca ting humor use in females, even if a larger sample size were to be us ed. It is somewhat safe to say that in females, prenatal hormone levels are not predictive of self-deprecating humor use. Hypothesis 6: a more male-typical gender role score will predict greater use of self-deprecating humor. In males. The hypothesis was not supported, as gender role score did not predict self-deprecating humor use in males. In females. The hypothesis was not supported, as gender role sc ore did not predict self-deprecating humor use in females. However, the N=35 and N=20 sample differed substantially in terms of not only effect size a difference of 0.55 but direction of effect size, with the N=20 sample yielding a modera te (albeit non-significant) effect with more female-typical gender role score predicting le sser use of self-deprecating humor, in line with the hypothesis. There is reason to believ e that the N=20 sample was more representative of the population, and having a larg er and more representative sample size may have yielded a significant relationship between the variables.
57 Thus, in males and females, it is unclear why no re lationship was found between gender role score and self-deprecating humor use. I t may be that prenatal hormone exposure does predict self-deprecating humor use, b ut that the small sample prevented significance from occurring; it is also possible th at, again, only certain facets of masculinity or femininity are associated with selfdeprecating humor use, especially seeing as Martin and colleagues found significance when looking at both positively and negatively valenced masculinity and femininity, whe reas the BSRI-R does not take positive and negative valence into account. Further more, it may be that only other nongender-related sociocultural factors influence self -deprecating humor, though this is unlikely considering Martin and colleagues signifi cant gendered findings. Non-directional hypothesis 7: do males and females differ in their use of affiliative humor? Males and females did not differ in their use of a ffiliative humor, in line with Greengross and Miller (2008) and in contr ast with Robinson and Smith-Lovin (2001)(who found that females used affiliative humo r more than males), though the latters data may very well be outdated, as it was from the early 1980s. However, it is possible that a larger sample size would have yield ed a significant difference, with males scoring higher on affiliative humor use than female s, which is somewhat in line with Martin and colleagues (2003), who found that males scored higher than females on affiliative humor use, but with a nearly insubstant ial effect size which may only have reached significance due to the large sample size. Non-directional hypothesis 8: does 2D:4D ratio pred ict affiliative humor use? Digit ratio did not predict affiliative humor use i n either males or females, though it is possible that a larger sample size would have yield ed a significant positive correlation
58 between affiliative humor and right 2D:4D in female s, such that a more female-typical digit ratio would be predictive of greater affiliat ive humor use. However, this would be in stark contrast to the previous possibility of males scoring higher than females on affiliative humor use. Thus, it appears unlikely that prenatal hormone exp osure is predictive of affiliative humor use in either males or females. Non-directional hypothesis 9: does gender role scor e predict affiliative humor use? Gender role score does not appear to predict affili ative humor use in males or females. Again, it is possible that non-gendered rather than gendered, sociocultural factors would predict affiliative humor use in male s and females. Self-Enhancing Humor Non-directional hypothesis 10: do males and females differ in their use of self-enhancing humor? Males and females do not appear to differ in their use of selfenhancing humor. Non-directional hypothesis 11: does digit ratio pre dict self-enhancing humor use? Digit ratio does not appear to predict self-enhanci ng humor use in either males or females, though it is possible that a larger sample size would have yielded a significant positive correlation between self-enhancing humor a nd left 2D:4D ratio in males, such that a more male-typical digit ratio would be predi ctive of lesser self-enhancing humor use. Ultimately, however, it is probably safe to sa y that exposure to prenatal hormones is not predictive of self-enhancing humor use in males or females. Hypothesis 12: a more male-typical gender role scor e will predict greater use of self-enhancing humor. The hypothesis was not supported, as gender role sc ore does
59 not appear to predict self-enhancing humor use in m ales or females. Again, it is possible that only non-gendered, rather than gendered, socio cultural factors would predict selfenhancing humor use in males and females, though th is is unlikely considering Martin and colleagues significant gender role findings. I t is also possible that only certain facets of masculinity or femininity are associated with se lf-enhancing humor use, especially seeing as Martin and colleagues found significance when looking at both positively and negatively valenced masculinity and femininity, whe reas the BSRI-R does not take positive and negative valence into account. Overarching Limitations Several possible overarching limitations exist, which may have affected the findings. For one, the small sample size was likely a large limitation. This limitation may apply especially to the correlations, which were ru n separately in males and females and therefore were even smaller. The same may be true f or the regressions, which controlled for several variables and therefore needed a larger sample size to achieve the optimum level of statistical power. Several trends were app roaching significance or had p-values on the lower end, and many of these may have reache d significance with a larger sample size. Another limitation was the puzzling tr end of males and females differing only in their right digit ratio, yet the majority of signif icant (or near-significant) correlations being with the left digit ratio. It is difficult to say what could have caused this. Perhaps it is due to the next limitation: a problematic sample (though the exact means by which it would be due to this is also unclear). This sample may have been problematic in several regards. For one, the sample
60 was not entirely random. Traditionally, in analyses data points are assumed to be independent of one another, when in reality, they m ay not have been. This is especially true for the second half of the sample (the N=35 sa mple), since a particularly helpful participant, who was an employee at the university, sent coworkers down to partake in the study. Not only did these coworkers interact wi th each other on a daily basis, but they all may have had similar qualities due to the fact that they all worked in similar jobs for the same employer. This is supported by the fact th at there were some substantial differences in significance and effect size between the N=20 and the N=35 sample, especially in self-deprecating humor, aggressive hu mor, and gender role score. The data points may also have not been independent of each o ther even in the first half of the sample (the N=20 sample), as groups of friends ofte n participated. Another possible limitation of the sample is that it may have been nonrepresentative of the general population in a subst antial way. This is supported by the fact that digit ratio was found to increase in men as th ey got older, and to decrease in women as they got older, which is entirely contrary to al l of the longitudinal research on digit ratio stability. This may indicate that participant s with certain personality traits, which could be reflected by digit ratio, may have been mo re likely to participate, creating the illusion that digit ratio changes as one ages. Spec ifically, in younger males of this sample, digit ratio was found to be more male-typical. Stud ies have shown that digit ratio can mediate males reactions to female experimenters, w ith lower/more male-typical digit ratios lead to greater interest. This raises the po ssibility that younger males with more male-typical digit ratios (and therefore stronger r eactions to female experimenters) may have been more likely to approach the experimenter and participate in the study. The fact
61 that several of the younger male participants flirt ed with the experimenter (and one asked her on a date) may lend support to this explanation Older participants with more maletypical digit ratios, on the other hand, may have f ound it less appropriate to approach a female experimenter with that intention in mind due to the age difference (or being married), leading to older males with more female-t ypical digit ratios (possibly with a greater inclination to help others?) approaching th e experimenter more frequently. Similarly, in younger females of this sam ple, digit ratio was found to be more female-typical; this may reflect the possibility th at older women who return to school may be more assertive and have more male-typical di git ratios. Likewise, older women in the workforce may be more assertive than the averag e woman of similar age, due to generational differences (women working less back t hen). Thus, these atypical trends between age a nd digit ratio may actually reflect a selfselected sample, which very well could have impacte d the generalizability and validity of the results in a substantial way. However, it is al so possible that digit ratio does change as one ages, though this is highly unlikely considerin g the multitude of research that has found the opposite. It is worth noting that despite these iss ues with the sample itself, several trends that have been consistently noted were found such as m ales having lower digit ratios than females, and males having more male-typical gender role scores than females possibly indicating that the sample was not entirely non-rep resentative. These trends emerged significantly even in the regression analysis, whic h was more conservative in that it controlled for the other variables. Either way, how ever, it seems plausible that this
62 unusual sample affected findings that would have ha d a smaller (and therefore, harder-todetect) effect size. Directions For Future Studies For one, future studies should use a diff erent recruitment method that would collect a more representative sample. This would include co llecting participants not affiliated with any particular institution, only allowing one member of a group of friends (or of a romantic affiliation) to participate, and reducing self-selection by more aggressively asking people to partake in the study. Additionally, participants should be included only if they are within a particular age group, as age was found to correlate with humor style, making it difficult to properly analyze the results. It would be interesting to see whether the aging process itself affects humor style, or whether it is simply a cohort effec t. Furthermore, it would be wise to use a mu ch larger sample size to detect somewhat smaller effects, perhaps somewhere around 100 parti cipants. Lastly, perhaps it would make sense to an alyze humor differences not in terms of the raw amount of a certain type of humor type used but rather, the percentage of a persons total humor that is comprised by each of t he four types; this would make sense because in previous studies, males have been shown to use more of all the humor types than females, making it difficult to draw any real conclusions about female humor use (other than the fact that they use it less than mal es). This proposed method may be a better measure of differences between the sexes in their use of humor. Only after understanding the causes of a behavior is it possible to control said
63 behavior, which could have possible implications to ward reducing aggressive humor toward others as well as promoting more cohesive re lations. Although this study was not experimental in nature, and therefore could not imp ly causality, it did provide information that could be used to determine which d irection further studies especially experimental ones should head in. It would be use ful to determine a way to conduct such experimental studies in an ethical manner, tho ugh this may present itself as a challenge. Thus, although this study did not provide strong su pport for the hypotheses, it also did not disprove them, and further studies should b e conducted to draw further conclusions.
64 References Bailey, A. A., & Hurd, P. L. (2005). Finger length ratio (2D:4D) correlates with physical aggression in men but not in women. Biological Psychology 68 215-222. doi:10.1016/j.biopsycho.2004.05.001 Berenbaum, S.A., Bryk, K.K.,Nowak,N., Quigley, C.A. ,Moffat, S., 2009. Fingers as a marker of prenatal androgen exposure. Endocrinology, 150, 51195124. Bem, S. L. (1977). On the utility of alternative pr ocedures for assessing psychological androgyny. Journal Of Consulting And Clinical Psychology, 45, 196-205. doi:10.1037/0022-006X.45.2.196 Ben-Hur, H., Thole, H.H.,Mashiah, A., Insler, V., B erman, V., Shezen, E., Elias, D., Zuckerman, A., Ornoy, A. (1997). Estrogen, progeste rone and testosterone receptors in human fetal cartilaginous tissue: immu nohistochemical studies. Calcified Tissue International., 60, 520526. Breedlove, S. M., Cooke, B. M., and Jordan, C. L. ( 1999). The orthodox view of brain sexual differentiation. Brain Behavior and Evolution,. 54 814. Bressler, E. R., Martin, R. A., & Balshine, S. (200 6). Production and appreciation of humor as sexually selected traits. Evolution and Human Behavior, 27 121-130. doi:10.1016/j.evolhumbehav.2005.09.001 Brown, W. M., Hines, M., Fane, B. A., & Breedlove, S. (2002). Masculinized finger length patterns in human males and females with con genital adrenal hyperplasia. Hormones And Behavior, 42 380-386. doi:10.1006/hbeh.2002.1830
65 Cameron, D. (1997). Performing gender identity: you ng mens talk and the construction of heterosexual masculinity. Language and Masculinity Blackwell, Oxford, pp. 4764. Chapell, M., Batten, M., Brown, J., Gonzalez, E., H erquet, G., Massar, C., & Pedroche, B. (2002). Frequency of public laughter in relation to sex, age, ethnicity, and social context. Perceptual and Motor Skills, 95 doi:10.2466/PMS.95.7.746-746 Ciumas, C. C., Hirschberg, A., & Savic, I. I. (2009 ). High fetal testosterone and sexually dimorphic cerebral networks in females. Cerebral Cortex, 19 1167-1174. doi:10.1093/cercor/bhn160 Coates, J. M., Gurnell, M., & Rustichini, A. (2009) Second-to-fourth digit ratio predicts success among high-frequency financial traders. Proceedings of the National Academy of Sciences of the United States of America 106 623628. Coyne, S. M., Manning, J. T., Ringer, L., & Bailey, L. (2007). Directional asymmetry (right-left differences) in digit ratio (2D:4D) pre dict indirect aggression in women. Personality and Individual Differences, 43 865-872. doi:10.1016/j.paid.2007.02.010 Crawford, M. (2003). Gender and humor in social con text. Journal of Pragmatics, 35 1413-1430. doi:10.1016/S0378-2166(02)00183-2 Csath, ., Osvth, A., Bicsk, ., Kardi, K., Man ning, J., & Kllai, J. (2003). Sex role identity related to the ratio of second to fourth d igit length in women. Biological Psychology, 62 147-156. doi:10.1016/S0301-0511(02)00127-8 De Vries, G. J., Rissman, E. F., Simerly, R. B., Ya ng, L., Scordalakes, E. M., Auger, C. J., & ... Arnold, A. P. (2002). A Model System for Study of Sex Chromosome
66 Effects on Sexually Dimorphic Neural and Behavioral Traits. The Journal Of Neuroscience, 22 9005-9014. Fisher, H. E., Rich, J., Island, H. D., & Marchalik D. (2010). The second to fourth digit ratio: A measure of two hormonally-based temperamen t dimensions. Personality and Individual Differences, 49 773-777. doi:10.1016/j.paid.2010.06.027 Fink, B., Neave, N., Laughton, K., Manning, J.T. (2 006). Second to fourth digit ratio and sensation seeking. Personality and Individual Differences, 41 12531262. Funny Web Jokes. Retrieved April 11, 2012, from http://funnywebjokes.com/funny6.html Galis, F., Ten Broek, C. M. A., Van Dongen, S., & W ijnaendts, L. C. D. (2010). Sexual dimorphism in the prenatal digit ratio (2D:4D). Archives of Sexual Behavior, 39 5762. Graeven, D. B., & Morris, S. J. (1975). College hum or in 1930 and 1972: An investigation using the humor diary. Sociology & Social Research, 59 406-410. Greengross, G., & Miller, G. F. (2008). Dissing one self versus dissing rivals: Effects of status, personality, and sex on the short-term and long-term attractiveness of selfdeprecating and other-deprecating humor. Evolutionary Psychology, 6, 393-408. Hampson, E., Ellis, C. L., & Tenk, C. M. (2008). On the relation between 2D:4D and sex-dimorphic personality traits. Archives of Sexual Behavior, 37 133-144. doi:10.1007/s10508-007-9263-3 Hines (2002). Sexual Differentiation of Human Brain and Behavior. Hormones, Brain, and Behavior (4), 425-462.
67 Hnekopp, J. (2011). Relationships between digit ra tio 2D:4D and self-reported aggression and risk taking in an online study. Personality and Individual Differences, 51 77-80. doi:10.1016/j.paid.2011.03.010 Hnekopp, J., Bartholdt, L., Beier, L., & Liebert, A. (2007). Second to fourth digit length ratio (2D:4D) and adult sex hormone levels: New dat a and a meta-analytic review. Psychoneuroendocrinology 32 313-321. doi:10.1016/j.psyneuen.2007.01.007 Hnekopp, J., & Schuster, M. (2010). A meta-analysi s on 2D:4D and athletic prowess: Substantial relationships but neither hand out-pred icts the other. Personality and Individual Differences, 48 410. Jackson, D. E., & Jackson, K. A. (1997). Gender, ob ject, and humor. Current Psychology, 15 286-290. doi:10.1007/s12144-997-1006-z Johnson, A. (1992). Language ability and sex affect humor appreciation. Perceptual and Motor Skills, 75 571-581. doi:10.2466/PMS.75.5.571-581 Kalliny, M., Cruthirds, K. W., & Minor, M. S. (2006 ). Differences between American, Egyptian and Lebanese Humor Styles: Implications fo r International Management. International Journal of Cross Cultural Management, 6 121-134. doi:10.1177/1470595806062354 Kempel, P., Gohlke, B., Klempau, J., Zinsberger, P. Reuter, M., Hennig, J. (2005). Secondtofourth digit length, testosterone and spa tial ability. Intelligence, 33 215230. Kohn, N. N., Kellermann, T. T., Gur, R. C., Schneid er, F. F., & Habel, U. U. (2011). Gender differences in the neural correlates of humo r processing: Implications for
68 different processing modes Neuropsychologia, 49 888-897. doi:10.1016/j.neuropsychologia.2011.02.010 Kuepper, Y., & Hennig, J. (2007). Behavioral aggres sion is associated with the 2D:4D ratio in men but not in women. Journal of Individual Differences, 28 64-72. doi:10.1027/1614-0001.28.2.64 Loehlin, J.C., Medland, S.E., Martin, N.G. (2009). Relative finger lengths, sex differences, and psychological traits. Archive of Sexual Behavior, 38 298305. Lutchmaya, S. S., Baron-Cohen, S. S., Raggatt, P. P ., Knickmeyer, R. R., & Manning, J. T. (2004). 2nd to 4th digit ratios, fetal testoster one and estradiol. Early Human Development, 77 23-28. doi:10.1016/j.earlhumdev.2003.12.002 Malas, M. A., Dogan, S., Evcil, E. H., & Desdiciogl u, K. (2006). Fetal development of the hand, digits and digit ratio (2D:4D). Early Human Development, 82 469475. Manning, J. T., Churchill, A. G., & Peters, M. (200 7). The effects of sex, ethnicity, and sexual orientation on self-measured digit ratio (2D :4D). Archives Of Sexual Behavior, 36 223-233. doi:10.1007/s10508-007-9171-6 Manning, J. T., & Peters, M. M. (2009). Digit ratio (2D:4D) and hand preference for writing in the BBC Internet Study. Laterality: Asymmetries Of Body, Brain And Cognition, 14 528-540. doi:10.1080/13576500802637872 Manning, J.T., Scutt, D., Wilson, J., Lewis-Jones, D.I. (1998). The ratio of 2nd to 4th digit length: a predictor of sperm numbers and conc entrations of testosterone, luteinizing hormone and oestorgen. Human Reproduction, 13 3000-3004. Manning, J.T., Trivers, R.L., Singh, D., Thornhill, R., 1999. The mystery of female beauty. Nature, 399 214-215.
69 Martin, R. A., & Kuiper, N. A. (1999). Daily occurr ence of laughter: Relationships with age, gender, and Type A personality. Humor: International Journal of Humor Research, 12 355-384. doi:10.1515/humr.19188.8.131.525 Martin, R. A., Puhlik-Doris, P., Larsen, G., Gray, J., & Weir, K. (2003). Individual differences in uses of humor and their relation to psychological well-being: Development of the Humor Styles Questionnaire. Journal of Research in Personality 37 48-75. doi:10.1016/S0092-6566(02)00534-2 Millet, K., & Dewitte, S. (2007). Digit ratio (2D:4 D) moderates the impact of an aggressive music video on aggression. Personality and Individual Differences, 43 289-294. doi:10.1016/j.paid.2006.11.024 Mio, J. (2009). Metaphor, humor, and psychological androgyny. Metaphor and Symbol, 24 174-183. doi:10.1080/10926480903028128 Mulkay, M. (1988). On Humor Basil Blackwell, New York. kten, A., Kalyoncu, M., & Yari, N. (2002). The ratio of secondand fourth-digit lengths and congenital adrenal hyperplasia due to 2 1-hydroxylase deficiency. Early Human Development, 70 47-54. doi:10.1016/S0378-3782(02)00073-7 Rizwan, S., Manning, J. T., & Brabin, B. J. (2006). Maternal smoking during pregnancy and possible effects of in utero testosterone: evid ence from the 2D:4D finger length ratio. Early Human Development, 83 87-90. Robinson, D. T., & Smith-Lovin, L. (2001). Getting a laugh: Gender, status, and humor in task discussions Social Forces, 80 123-158. doi:10.1353/sof.2001.0085 Romano, M., Rubolini, D., Martinelli, R., Alquati, A., & Saino, N. (2005). Experimental manipulation of yolk testosterone affects digit len gth ratios in the ring-necked
70 pheasant (Phasianus colchicus). Hormones And Behavior, 48 342-346. doi:10.1016/j.yhbeh.2005.03.007 Talaroviov, A., Krkov, L., & Blaekov, J. (2009). Testo sterone enhancement during pregnancy influences the 2D:4D ratio and open field motor activity of rat siblings in adulthood. Hormones And Behavior, 55 235-239. doi:10.1016/j.yhbeh.2008.10.010 Trivers, R., Manning, J., & Jacobson, A. (2006). A longitudinal study of digit ratio (2D:4D) and other finger ratios in Jamaican childre n. Hormones and Behavior, 49 150156. Vernon, P. A., Martin, R. A., Schermer, J., Cherkas L. F., & Spector, T. D. (2008). Genetic and environmental contributions to humor st yles: A replication study. Twin Research and Human Genetics, 11 44-47. doi:10.1375/twin.11.1.44 Veselka, L., Schermer, J., Martin, R. A., & Vernon, P. A. (2010). Laughter and resiliency: A behavioral genetic study of humor sty les and mental toughness. Twin Research And Human Genetics, 13 442-449. doi:10.1375/twin.13.5.442 Wilson, G. D., Rust, J., & Kasriel, J. (1977). Gene tic and family origins of humor preferences: A twin study. Psychological Reports, 41 659-660. Yarber, W.L., Sayad, B.W., Strong, B. (2010). Gende r and Gender Roles. In Human Sexuality: Diversity in Contemporary America (7th e d.) (p. 127). New York: McGraw Hill. Yip, J. A., & Martin, R. A. (2006). Sense of humor, emotional intelligence, and social competence. Journal Of Research In Personality, 40 1202-1208. doi:10.1016/j.jrp.2005.08.005
71 Footnotes 1Ethnicity, handedness, and sexual orientation have all been shown to affect 2D:4D ratio (Manning et al., 2007; Manning et al., 2000; McFadden et al., 2005), which could possibly create inconsistencies within the da ta.
72 Table 1 Descriptive Statistics Measure Mean Minimum Maximum St. Dev. Aggressive Humor 25.49 11.00 55.00 9.71 Self-Deprecating Humor 27.71 10.00 53.00 9.49 Affiliative Humor 47.66 27.00 56.00 6.90 Self-Enhancing Humor 43.66 30.00 56.00 6.52 Right 2D:4D Ratio 0.98 0.92 1.07 0.04 Left 2D:4D Ratio 0.97 0.90 1.03 0.03 Gender Role Score 54.00 29.00 83.00 11.00
73 Table 2 Intercorrelations Between Humor, Digit Ratio, Gende r Role, and Age in Females Measure 1 2 3 4 5 6 7 8 1. Aggressive Humor -0.44* 0.08 -0.31 0.44* 0.49* -0.23 -0.38 2. Self-Deprecating Humor --0.23 -0.02 0.03 -0.08 0.19 -0.25 3. Affiliative Humor ---0.08 0.31 -0.08 -0.04 -0.10 4. Self-Enhancing Humor ----0.05 -0.22 0.19 -0.25 5. Right 2D:4D Ratio -----0.06 -0.13 -0.5 9** 6. Left 2D:4D Ratio -------0.19 -0.22 7. Gender Role Score -------0.10 8. Age --------Note. p < .10 ** p < .05
74 Table 3 Intercorrelations Between Humor, Digit Ratio, Gende r Role, and Age in Males Measure 1 2 3 4 5 6 7 8 1. Aggressive Humor 0.56** 0.17 -0.45* -0.32 -0.50 -0.28 -0.40 2. Self-Deprecating Humor 0.18 -0.62** 0.00 -0.35 0.28 -0.27 3. Affiliative Humor -0.21 -0.08 -0.03 -0.08 -0.28 4. Self-Enhancing Humor -0.02 0.41 0.07 0.30 5. Right 2D:4D Ratio 0.53** 0.10 0.16 6. Left 2D:4D Ratio -0.46* 0.45* 7. Gender Role Score -0.38 8. Age Note. p < .10 ** p < .05
75 Appendix A *Items marked with an asterisk are reverse keyed (8 -score) Affiliative 1. I usually dont laugh or joke around much with o ther people.* 5. I dont have to work very hard at making other p eople laughI seem to be a naturally humorous person. 9. I rarely make other people laugh by telling funn y stories about myself.* 13. I laugh and joke a lot with my closest friends. 17. I usually dont like to tell jokes or amuse peo ple.* 21. I enjoy making people laugh. 25. I dont often joke around with my friends.* 29. I usually cant think of witty things to say wh en Im with other people.* Self-Enhancing 2. If I am feeling depressed, I can usually cheer m yself up with humor. 6. Even when Im by myself, Im often amused by the absurdities of life. 10. If I am feeling upset or unhappy I usually try to think of something funny about the situation to make myself feel better. 14. My humorous outlook on life keeps me from getti ng overly upset or depressed about things. 18. If Im by myself and Im feeling unhappy, I mak e an effort to think of something funny to cheer myself up. 22. If I am feeling sad or upset, I usually lose my sense of humor.* 26. It is my experience that thinking about some am using aspect of a situation is often a very effective way of coping with problems. 30. I dont need to be with other people to feel am used I can usually find things to laugh about even when Im by myself. Aggressive 3. If someone makes a mistake, I will often tease t hem about it. 7. People are never offended or hurt by my sense of humor.* 11. When telling jokes or saying funny things, I am usually not very concerned about how other people are taking it. 15. I do not like it when people use humor as a way of criticizing or putting someone down.* 19. Sometimes I think of something that is so funny that I cant stop myself from saying it, even if it is not appropriate for the situation 23. I never participate in laughing at others even if all my friends are doing it.*
76 27. If I dont like someone, I often use humor or t easing to put them down. 31. Even if something is really funny to me, I will not laugh or joke about it if someone will be offended.* Self-Deprecating 4. I let people laugh at me or make fun at my expen se more than I should. 8. I will often get carried away in putting myself down if it makes my family or friends laugh. 12. I often try to make people like or accept me mo re by saying something funny about my own weaknesses, blunders, or faults. 16. I dont often say funny things to put myself do wn.* 20. I often go overboard in putting myself down whe n I am making jokes or trying to be funny. 24. When I am with friends or family, I often seem to be the one that other people make fun of or joke about. 28. If I am having problems or feeling unhappy, I o ften cover it up by joking around, so that even my closest friends dont know how I reall y feel. 32. Letting others laugh at me is my way of keeping my friends and family in good spirits.