Sex Differences in Affect Behaviors, Desired Social Responses, and Accuracy at Understanding the Social Desires of Other People

Evolutionary Psychology
www.epjournal.net – 2008. 6(3): 506-522
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯

Original Article
Sex Differences in Affect Behaviors, Desired Social Responses, and Accuracy
at Understanding the Social Desires of Other People

Jacob M. Vigil, Department of Psychology, University of North Florida, Jacksonville, U.S.A. Email:
j.vigil@unf.edu
Abstract: I recently proposed a socio-relational framework that suggests that phenotypic
variation in the expression of discrete affect behaviors (e.g., expressed anger vs. sadness)
may reflect two basic dimensions of behavioral response. The first dimension is the
motivation to selectively attract or avoid interactions with different people. The second
dimension is the behavioral display of either personal capacity or trustworthiness cues,
often through the expression of dominant and submissive behaviors, respectively (Vigil, in
press). Sex differences in affect behaviors (e.g., externalizing vs. internalizing displays)
may reflect developmental sensitivities to advertise capacity and trustworthiness cues
somewhat differently, due to the unique social dynamics and relational constraints under
which males and females evolved. In this study, I use a series of self-report questionnaires
to examine two basic assumptions of the framework. The first assumption is that sex
differences in nonverbal affect behaviors (aggression vs. crying) reflect the desire for, and
are effective for causing other people to either avoid or comfort the individual, respectively.
I hypothesized that males would report a greater likelihood of responding to various
distress moods with aggressive, rather than crying behaviors, and that males would desire,
and believe male peers similarly desire distancing responses from others when feeling these
moods. Instead, females were hypothesized to report more crying behaviors, and to desire,
and believe female peers desire more comforting responses from others. The second
assumption is that people are more accurate at inferring the desired social responses (i.e., to
be left alone or comforted when experiencing distress) of same-sex, rather than opposite-
sex peers. I hypothesized that people’s beliefs of same-sex peers are more similar to the
self-reported desires of the male and female participants, themselves, than their beliefs of
opposite-sex peers. The results are consistent with the first assumption, but show that only
men were more accurate at inferring same-sex, rather than opposite-sex desires. In general,
the findings are consistent with the socio-relational framework that suggests that humans
may have evolved specialized expressive behaviors and interpersonal styles for
communicating and interacting with same-sex affiliates.

Evolutionary Psychology – ISSN 1474-7049 – Volume 6.3) 2008. -506-

---

Sex differences in affect behavior

Keywords: evolution, sex differences, nonverbal behaviors, social styles, empathy
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
Recent theory on the evolution and development of social behaviors in humans
suggests that males and females evolved specialized expressive styles for communicating
and interacting with same-sex affiliates (Geary, Byrd-Craven, Hoard, Vigil, and Numtee,
2003; Geary and Flinn, 2002; Taylor et al., 2000; Vigil, 2007). Boys and men express
higher levels of behaviors that project the perception of independence and dominance (e.g.,
expressed toughness), which are in turn and broadly associated with popularity among male
peers (Arsenio, Cooperman, and Lover, 2000; Chang, 2004; Cillessen and Mayeux, 2004;
Farmer, Bishop, O’Neal, and Cairns, 2003; Levy, 2005; Rodkin, Farmer, Pearl, and Van
Acker, 2000). Girls and women, in contrast, express higher levels of behaviors, such as
crying, that are instead associated with the perception of nurturance, submissiveness, and
compassion (Montepare and Dobish, 2003; Van Tilburg, Unterberg, and Vingerhoets,
2002; Williams, 1982). In this article I describe a meta-theoretical framework that attempts
to help understand the social properties (socially relevant information) that may underlie
variation in the expression of affect behaviors (e.g., expressed anger vs. sadness), as well
as the social, evolutionary constraints that may have selected divergent expressive styles in
males and females. I then examine some basic assumptions of the framework, including the
hypotheses that sex differences in distress behaviors may be effective for causing other
people to react in systematic and distinct (e.g., sex-typical) ways, and that people are more
accurate at inferring the desired social responses (i.e., to be either left alone or consoled
when distressed) of same-sex, rather than opposite-sex peers.

The Socio-Relational Framework of Expressive Behaviors
I recently proposed a Socio-Relational Framework of Expressive Behaviors (SRFB;
Vigil, in press), which suggests that variation in the expression of discrete affect behaviors,
such as anger and sadness, can be understood in terms of two primary dimensions of
behavioral responses. The first dimension is based on the basic (“reptilian”) motivational
system designed to either approach or withdraw from environmental stimuli (Buck, 1999;
Davidson, 1993; Davidson, Jackson, and Kalin, 2000; see also, Cacioppo, Klein, Berntson,
and Hatfield, 1993; Camras, Holland, and Patterson, 1993; Gray, 2002). In humans, this
system may be especially sensitive to evaluate potential benefits and dangers of interacting
with different people and engaging in different types of relationships (Andersen and Chen,
2002; Bugental, 2000; Kenny, Mohr, and Levesque, 2001; Öhman, 2002; Rudolph,
Caldwell, and Conley, 2005). Interactions that signal the potential to provide reciprocated
investment should stimulate affiliative responses (e.g., interpersonal attraction, bonding,
protection, and comforting behaviors) and associated emotions (e.g., joy, sadness, and
sympathy). Instead, interactions that are perceived to be non-reciprocal, and thus
potentially exploitive, should stimulate avoidant responses (e.g., expressed hostility) and
emotions (e.g., fear, disgust, and anger) designed to protect the individual. In this sense, all
social (expressive) behaviors should be identifiable along an affiliative/avoidant dimension
of behavioral response, depending on whether the behavior usually results in social
bonding or in social distancing, across individual relationships. Importantly, some affect
Evolutionary Psychology – ISSN 1474-7049 – Volume 6(3). 2008. -507-

---

Sex differences in affect behavior
behaviors such as sadness are pleiotropic (resulting in multiple outcomes), and can result in
both social bonding (e.g., among intimate affiliates) and in social distancing (e.g., among
less familiar affiliates), depending on the context of the relationship (Vigil, in press).
The second dimension that may help explain variation in expressive behaviors is the
display of the two most fundamental properties of reciprocity potential, or attractiveness as
a social partner (e.g., mate, friend, community member, etc.). Specifically, I previously
hypothesized that that these essential properties can be conceptualized as the individual’s
(a) perceived capacity to either help or hurt other people, and their (b) perceived
trustworthiness or probability of actually reciprocating altruism (Vigil, 2007). These two
elementary concepts (capacity and trustworthiness) may be the most parsimonious
properties (irreducible characteristics) by which social evaluations and expressive
behaviors are based. This is because insufficient levels of either of these constructs (e.g.,
having a lot to offer but being unwilling to do so, or being willing but having nothing to
offer) results in little impact to others. This thesis is supported by recent research showing
that people are particularly sensitive to process cues of competency and trustworthiness in
human faces (Todorov, 2008, Todorov, Mandisodza, Goren, and Hall, 2005; Willis and
Todorov, 2006).
From the perspective of the SRFB, humans should also be sensitive to advertise
cues of capacity and trustworthiness to other people through myriad expressive behaviors,
and especially through behaviors that are perceived as dominant or submissive. Traits (e.g.,
healthiness and independence) and affect behaviors that are perceived as dominant, such as
expressed joy and anger, may be functional for signaling the capacity to reciprocate with
other people (affiliation) and the ability to protect oneself (avoidance), respectively. Traits
(e.g., feebleness and dependence) and affect behaviors that are perceived as submissive
(e.g., fear and sadness) may instead operate (in part) to exaggerate the display of
vulnerability in order to reduce the perception of threat (avoidance) and to signal deference
(affiliation), and hence the trustworthiness component of reciprocity potential to other
people (Vigil, in press).
Research on social relationships suggests that people adjust preference for
(affiliative) capacity attributes (e.g., physical healthiness, intelligence, material resources)
and trustworthiness attributes (e.g., kindness, responsibility) in relation to the expected
duration of the relationship. For example, research on mate preferences shows that both
men and women place greater emphasis on preference for high capacity traits rather than
high trustworthiness traits when anticipating short-term relationships. In contrast, people
increase preference for higher levels of trustworthiness traits than personal capacity
attributes when seeking long-term or more exclusive partners (e.g., Cottrell, Neuberg, and
Li, 2007; Geary, Vigil, and Byrd-Craven, 2004; Li et al., 2002; Vigil, Geary, and Byrd-
Craven, 2006).
Other research shows that people adjust the display of affect behaviors in relation to
social situational factors, such as the perceived social status, group size, familiarity, and
relationship history of audience members (e.g., Chapman, 1973; Fridlund, 1991; Kraut and
Johnston, 1979). Several studies, for example, have shown that people increase the display
of externalizing behaviors (e.g., aggression) and overall capacity cues (e.g., expressed
confidence) in larger group settings, and instead increase the expression of internalizing
behaviors (e.g., sadness and worrying) and overall trustworthiness cues (e.g., expressed
compassion) in more intimate (e.g., dyadic) social settings (Benenson et al., 2002;
Evolutionary Psychology – ISSN 1474-7049 – Volume 6(3). 2008. -508-

---

Sex differences in affect behavior
LaFrance, Hecht, and Paluck, 2003; Leaper and Smith, 2004; Vingerhoets, Cornelius, Van
Heck, and Becht, 2000). Collectively, these studies suggest that subtle detection and
expression of capacity and trustworthiness cues of reciprocity potential may be specialized
and functional for facilitating the formation of strategic, fitness-enhancing relationships in
humans (Vigil, in press).

Evolution of Sex Differences in Affect Behaviors
From the perspective of the SRFB, phenotypic variation in the expression of
capacity and trustworthiness cues are predicted to covary with the fundamental dynamics of
individuals’ social networks (Vigil, in press). These dynamics can be summed up as the
quantity, quality, and expected extent (e.g., short-term vs. long-term) of interpersonal
investment across one’s relationships. Research in the social developmental literature
describes similar distinctions in the prototypical social networks and relationship styles of
males and females. These studies show that girls and women spontaneously form, and
report a preference for fewer, but more intimate relationships with their same-sex peers.
Males, in contrast, evidence the opposite pattern by usually forming and reporting a
preference for larger, but less intimate social networks on average (Geary et al., 2003;
Lever, 1978; Maccoby, 2002; Rose and Rudolph, 2006; Vigil, 2007).
One model that has the potential to explain these sex differences in relationship
dynamics is based on a human evolutionary history characterized by male-biased
philopatry and male-male coalitional competition. In this type of social system, males tend
to remain in closer proximity to their male-kin—this allows them to form strong, kin-based
coalitions—while females tend to emigrate into the social networks of their husbands upon
marriage (see Geary, 1998, 2002; Geary and Flinn, 2002; Geary et al., 2003; Wrangham
and Peterson, 1996). With this system, males would have been exposed to and reliant upon
more daily interactions with kin. In contrast, females would have been dependent upon
more daily interactions with non-kin or distantly related kin (de Waal, 1993; Geary, 2002;
Geary et al., 2003).
Relationships between non-kin operate differently than relationships among kin,
because the former are based on reciprocal altruism (equal exchange of resources), whereas
the latter are based on inclusive fitness (equal sharing of genes; Hamilton, 1964; Trivers,
1971). In the context of a SRFB, the necessity for females to solicit more committed and
secure relationships among non-kin may have co-evolved with the sensitivity to exaggerate
the display of submissive behaviors vis-à-vis explicit demonstrations of vulnerability (e.g.,
expressed sadness) and appeasement (e.g., expressed compassion). These behaviors may
therefore reflect an expressive bias for females to signal higher levels of the trustworthiness
component of reciprocity potential than males (Vigil, 2007, in press; see also Geary and
Flinn, 2001, 2002; Geary et al., 2003; Taylor et al., 2001). Formation of smaller social
networks in females may in turn enable greater allocation of intimate investment behaviors
(requiring time and empathy) and thus trustworthiness cues into fewer, but more reliable
relationships (Geary and Flinn, 2001, 2002; Geary et al., 2003; Vigil, 2007; Vigil and
Geary, 2008).
Instead, as a result of more daily exposure to kin-related peers, there is predicted to
be an overall relaxation of the selection pressures for males to exaggerate the expression of
vulnerability and intimacy behaviors within their relationships (Geary et al., 2003; Vigil,
2007). In theory, this would have enabled men to form a greater number of total
Evolutionary Psychology – ISSN 1474-7049 – Volume 6(3). 2008. -509-

---

Sex differences in affect behavior
relationships, and thus larger and more functional coalitions (Geary and Flinn, 2001, 2002).
At the same time, men may have evolved a sensitivity to rely more heavily on the
behavioral advertisement of dominance behaviors (e.g., aggression, denial of pain, flashy
body movements) and thus the capacity component of reciprocity potential (Vigil, 2007, in
press). Capacity cues are more observable (visible from a distance and through limited
interactions) than trustworthiness cues, and may be more effective for maintaining larger,
less intimate (e.g., time-investing), and less exclusive peer networks among males. From
this perspective, sex differences in the propensity express higher levels of dominance
displays (e.g., impulsivity, risk-taking, inflated self-evaluations, and physical aggression) in
males and submissiveness displays (e.g., cautiousness, conscientiousness, modest self-
descriptions, and expressed sadness) in females may reflect asymmetries in the social
ecologies and relationship demands in which humans evolved. More specifically, these
behaviors may reflect the associated sensitivity to interchange the capacity and
trustworthiness components of reciprocity potential within these social ecologies, among
same-sex peers (Vigil, 2008; see also Geary and Flinn, 2002; Geary et al,. 2003; Vigil,
2007).

Current Study
The plausibility of the SRFB for understanding variation in sex-general and sex-
typical expressive behaviors rests on several testable assumptions. One assumption, for
example, is that males and females express distress through behaviors that project the
perception of dominance and submissiveness, which may ultimately operate (in part) to
cause peers to either distance from, or to comfort the individual, respectively. Expressed
anger is associated with the perception of dominance, whereas fear and sadness are
associated with the perception of submissiveness (e.g., Marsh, Adams, and Kleck, 2005;
Montepare and Dobish, 2003; Zuroff, Fournier, and Moskowitz, 2007; see also Mondillon
et al., 2005). Higher levels of physical aggression among males may signal the individual’s
strength, prowess, and hence capacity to hurt others and to defend oneself. Anger facial
expressions are inversely associated with the likelihood of receiving sympathetic responses
from others (Strayer and Roberts, 2004), which may further enhance the perception of
capacity (e.g., toughness) by causing peers to distance themselves from the individual and
thus demonstrate the ability to defend oneself in times of stress. Instead, sadness behaviors
among females may display high trustworthiness cues (e.g., via expressed vulnerability) in
order to evoke reciprocal displays of trustworthiness (e.g., via expressed compassion) and
hence investment into the relationship from peers. Trustworthiness interchanges may be
especially adaptive under conditions in which individuals experience decreased capacity
resources (i.e., negative life events), when they are unable to effectively advertise these
qualities to others, and when reliable social support is most essential. Another assumption
of the framework is that prototypical male and female expressive styles (e.g., masculine
and feminine distress behaviors) are used to communicate with same sex affiliates.
According to this assumption, people should be better at inferring (understanding) the
social desires of same-sex, rather than opposite-sex peers.
In the current study, I examine these assumptions through a series of self-report
questionnaires designed to assess 4 predictions. According to the first assumption above,
people should (a) be more likely to respond to dominant mood states (e.g., anger and
frustration) with aggressive behaviors and to respond to submissive moods (e.g., sadness,
Evolutionary Psychology – ISSN 1474-7049 – Volume 6(3). 2008. -510-

---

Sex differences in affect behavior
fear, worrying, embarrassment) with crying behaviors. These responses should differ by
sex such that males and females should exaggerate reports of aggression and crying
behaviors, respectively. Likewise, people should (b) desire others to respond to one’s
dominant distress moods with distancing (avoidant) reactions and to respond to submissive
moods with comforting (affiliative) reactions. These desires should again differ by sex,
such that males should desire more distancing reactions from others, whereas females
should desire more comforting reactions. In order to support the hypothesis that sex-typical
distress behaviors are effective for communicating the desire to evoke distancing and
comforting responses from peers, I also needed to demonstrate that people (c) believe their
peers similarly desire to be left alone and comforted when feeling dominant and submissive
distress moods and expressing aggression and crying behaviors. These beliefs should differ
according to the sex of the respondent and sex of the peer. Female respondents are expected
to believe that peers desire more comfort than believed of male respondents, and both sexes
should infer that female peers desire more comfort than male peers. Finally, to examine the
second assumption, I hypothesized that (d) people are better at inferring the social desires
of same-sex, rather than opposite-sex peers. I examine this hypothesis by directly
comparing the believed social desires of same-sex and opposite-sex peers (to receive
comfort vs. distancing when experiencing distress moods) and the reported desires of the
male and female participants, themselves.
Materials and Methods
Participants and Procedures
Two hundred participants (125 males) were recruited from the University of
Missouri-Columbia (mean age for men = 19.1, SD = 0.9 years; mean age for women =
18.9, SD = 1.3 years). In a 30 to 40 minute testing session, participants completed a series
of questionnaires designed to measure “stress behaviors” in a classroom setting.

Materials
The questionnaires were designed specifically for this study, and consisted of 56
items across three broad sections. Items in the first section were designed to assess the
association between common nonverbal distress behaviors (crying vs. physical aggression)
and three dominant mood states (anger, frustration, and hot-temperedness) and five
submissive moods (sadness, embarrassment, worry, fright, and stress). Participants were
told to think about their behaviors when they felt different emotions, and were given eight
forced-choice items that asked “when I am feeling (mood state), I am most likely to
______” (choices: “start crying” vs. “hit, bang, or slam a nearby object”). The second part
of this section consisted of 10 items designed to measure desired social responses from
other people when experiencing the eight mood states and nonverbal distress behaviors
(crying and hitting). Participants were asked to imagine how they wish other people would
respond to their emotions, and were given eight forced-choice items that asked “when I am
feeling (mood state), I wish people would ______” (choices: “leave me alone for awhile”
vs. “come over and give me comfort”). Two additional items asked participants how they
wished others would respond when the participant is “crying” and when they are “hitting,
slamming, or banging objects”.

The second section was designed to assess how people respond to negative life-
Evolutionary Psychology – ISSN 1474-7049 – Volume 6(3). 2008. -511-

---

Sex differences in affect behavior
events, the perceived impact of the events on individuals’ life, and how they wish others
would react to them, under the conditions. Participants were asked to imagine experiencing
six negative life-events that consisted of 1) stepping in a mud puddle, 2) receiving an
official letter from the Undergraduate Dean stating that the participant was accused of
cheating on an exam and would be expelled from the University, 3) locking their keys
inside their car or house, 4) receiving a phone call from the police saying that their parent
had been killed in a car accident, 5) receiving a failing grade on an important exam, and 6)
stubbing their toe on a metal chair. For each of the six hypothetical events, participants
were asked three questions; the first question asked participants to report what their “first
reaction would most likely be” (choices: “start crying” vs. “hit, bang, or slam a nearby
object”); the second question asked how much “this event would impact my life” (choices:
“not at all”, “a little”, or “very much”); and the third question asked how they “wish people
would respond” to their reaction (choices: “leave me alone for a while” vs. “coming over to
give me comfort”).

The third section was designed to assess knowledge of desired responses of same-
sex vs. opposite-sex peers. In the first part of this section, participants were asked to “think
about how you think a male friend might want you to respond to his emotions”.
Participants were then given eight questions that asked to “imagine a close male friend is
feeling very (each of the eight emotions from Section 1), he probably wants me to ______”
(choices: “leave him alone for awhile” vs. “go over and give him comfort”). Two
additional questions asked how they think their male peers would want the participant to
respond if the peer was crying and hitting nearby objects. The entire set of 10 questions
was then repeated to refer to the social desires of female peers.
Results
Prediction (a) People respond to dominant moods with physical aggression and to
submissive moods with crying behaviors, and these responses are exaggerated in males
and females, respectively. Percentages of males and females that reported either crying or
hitting behaviors in response to the 8 mood states (sadness, anger, frustration,
embarrassment, worry, hot-temperedness, fright, and stress) are presented in the first set of
columns of Table 2. Nonverbal responses across mood states were assessed by McNemar’s
chi square tests; for these comparisons, a conservative cutoff point (? = .002) was used to
test mood differences. As shown in Table 2, males reported the highest likelihood of hitting
in response to dominant moods that included feelings of hot-temper, anger, and frustration.
In contrast, females reported the highest likelihood of crying in response to submissive
moods that included feelings of stress, fright, embarrassment, worry, and sadness. Chi
square tests revealed sex differences in nonverbal responses for each of the mood-states (ps
< .001), due to higher reports of crying behaviors in women and aggressive behaviors in
men. These findings indicate a male-bias to respond to dominant moods with aggressive
behaviors that are suggestive of capacity displays, and a female-bias to respond to
submissive moods with sadness behaviors that are suggestive of trustworthiness displays
(Vigil, in press).
Evolutionary Psychology – ISSN 1474-7049 – Volume 6(3). 2008. -512-

---

Sex differences in affect behavior
Table 2. Sex Differences in Affect Behaviors, Desired Social Responses, and Believed Social Desires of Peers

Nonverbal Behaviors

Desired Social Responses

Beliefs about Males

Beliefs about Females

Males Females
Males Females Males Females Males Females

Cry
Hit Cry Hit D C D C
D C D C
D C D C
Dominant
(Capacity)
Moods



Hot-temper
1%a
99% a
32% a
68% a
86% a 14% a 89% a 11% a 90% a 10% a 96% a 4% a

60% a
40% a
80% a 20% a






Anger
2% a
98% a
51% b
49% b
84% a 16% a 83% a b 17% a b
86% a 14% a 91% a b 9% a b 46% b 54% b 51% b 49% b




Frustration
6% a
94% a
73% c
27% c
67% b 33% b
65%bc 35%bc 69% b 31% b 76%bc 24%bc 24%c 76%c 35%bc 65%bc




Submissive
(Trustworthiness)
Moods




Stress
19%b
81%b
89% d
11% d
49%c 51%c 56%cd 44%cd 50%bc 50%bc
59%cd 41%cd

10%de
90%de
27%bc 73%bc






Fright
34%c
66% c
96% d
4% d

26% d
74% d
8%e 92%e 30% d 70% d 21%e 79%e 3%ef 97%ef 0% d 100% d






Embarrassment
37%cd
63%cd
92% d
8% d
53%bc 47%bc 49%cd 51%cd

56%bc
44%bc
80%bc 20%bc 18%cd 82%cd 25%bc 75%bc






Worry
49%d
51% d
93% d
7% d

29% d
71% d
16%e 84%e 19% d 81% d 9%e 91%e 5%ef 95%ef 3% d 97% d






Sadness
53% d
47% d
99% d
1% d
38%cd 62%cd 39% d 61% d 46%c 54%c 45% d 55% d 1%f 99%f 1% d 99% d




Nonverbal
Behaviors



Crying -
-
-
-

54%
40% 60%
53% 47% 51% 49%
2% 98% 3% 97%

46%
Hitting
-
- -
-

82% 18% 76% 24%
82%
18%
95% 5%
46%
54%
65% 35%
Note. Frequencies of the responses of male and female participants are shown for crying and hitting behaviors (first set of columns), for desiring distance (D) or
comfort (C) responses from others (second set of columns), and for the believed desires of male and female peers (third and fourth sets of columns). Underlined
values indicate significant differences (p < .05) between the responses of male and female participants. Different superscript letters indicate significant mood
differences in crying/hitting and distancing/comforting responses (? = .002).

Evolutionary Psychology – ISSN 1474-7049 – Volume 6.3) 2008. -513-

---

Sex differences in affect behavior
Prediction (b) People desire others to react to one’s dominant distress behaviors
with avoidant responses and to react to one’s submissive distress behaviors with
comforting responses, and these desires are moderated by sex. Percentages of males and
females that desired either distancing or comforting responses from others are shown in the
second set of columns in Table 2. Examination of the social desires across mood states (? =
.002) revealed that, collectively, males and females report the greatest desire for others to
avoid rather than comfort them when feeling hot-temper and anger, followed by frustration,
stress and embarrassment, sadness, worry, and fright. Chi square tests revealed sex
differences in desired responses for fright, worry, and crying behavior (ps < .05), due to
greater desires of females to receive more social comforting than males.
In order to further examine the hypothesis that males and females differentially
prefer others to react to one’s distress with avoidant and comforting responses,
respectively, I assessed distress behaviors and desired social responses to six negative life-
events (stepping in mud, expulsion from school, locked keys in car, parental death, failing
grade, and stubbed toe). Chi square tests, again, revealed a higher likelihood of males
responding to each negative life-experience with hitting than crying behaviors and females
with more crying than hitting behaviors (ps < .001), except for parental death (p = .17), of
which both sexes reported a greater likelihood of crying to this event. These group
differences emerged despite no sex differences in the perceived impact of the events on
participant lives (ps > .29). Examination of desired social responses revealed sex
differences for each event; males desired others to react with more avoidant than
comforting reactions and females desired others to provide more comforting than avoidant
behaviors (ps < .05), except for parental death (p = .13) and getting a failing grade (p =
.30), of which both sexes reported a greater likelihood of desiring comfort and avoidance
(respectively) to these events. Collectively, these findings indicate that people desire others
to respond to self-expressed anger and hot-temper with avoidant reactions and respond to
self-worry and especially fright with comforting reactions and that males and females
differentially prefer avoidant and comforting social responses from others.
Prediction (c) People believe their peers similarly desire others to respond to the
peer’s dominant and submissive distress behaviors with avoidant and comforting
responses, respectively, and these beliefs differ by sex of the respondent and sex of the peer.
The believed social desires (avoidance vs. comforting) of male and female peers are shown
in the third and fourth set of columns in Table 2. Examination of believed social desires of
male peers for each of the mood states (? = .002) revealed that both male and females
participants believed male peers desired to be distanced from when experiencing dominant
moods such as hot-temper and anger, and desired to be comforted when experiencing
submissive moods such as fright and worry. Examination of the believed social desires of
female peers again revealed that both male and female participants believed female peers
desired distance when experiencing dominant moods such as hot-temperedness, and desired
to be comforted when experiencing submissive moods such as sadness, fright, and worry.
Chi square tests revealed differences in believed desires of male and female peers
among male and female participants. Chi square tests revealed that more males than
females believed that male peers desire comfort when experiencing embarrassment and
hitting behaviors, and that female peers desire comfort when experiencing hot-
temperedness, stress, and hitting behavior. These tests also revealed that both male and

Evolutionary Psychology – ISSN 1474-7049 – Volume 6.3) 2008. -514-

---

Sex differences in affect behavior
female participants believed that female peers desired more social comforting than male
peers when experiencing each of the 8 mood states and distress behaviors (ps < .05), except
for worrying among women (p = .06). These findings indicate that people believe their
peers want others to respond to the peer’s dominant moods with avoidant reactions, and to
respond to submissive moods with comforting reactions. Both sexes believe female peers
desired more comforting behavior than male peers; however, male participants were more
likely to infer that peers desire more comfort than female participants for some of the
moods and aggressive behavior.
Prediction (d) Males and females are more accurate at inferring the social desires
of same-sex, rather than opposite-sex peers. This hypothesis was examined by comparing
participants’ self-desired social responses (for others to avoid or provide comfort; second
set of columns in Table 2) and participants’ believed social desires of male and female
peers (third and fourth set of columns of Table 2), separately by sex. Due to the possibility
of carry-over effects in reporting self- and other-desired responses, McNemar’s chi square
tests were used to test significant (p < .05) differences. These tests revealed a same-sex
advantage at inferring the social desires of others among males, but not among females.
Specifically, male participants believed their male peers’ social desires were similar to their
own social desires for every mood state, crying, and hitting behavior (ps > .05). In contrast,
male participants believed female peers desired more comforting responses than the male
participants, themselves, self-desired for every mood state, crying, and hitting behaviors (ps
< .001) as described above.
For female participants a different pattern emerged, such that female participants
believed their female peers’ social desires were only similar to their own social desires for
hot-temper (p = .09) and hitting behavior (p = .15). For the remainder of the mood states
(except for fright) and for crying behavior, female participants self-desired less comforting
responses than they believed their female peers desired (ps < .001). Comparisons for fright
could not be computed due to the uniformity of belief that females' peers desired comfort
when frightened. In contrast, female participants self-desired more similar responses to the
believed desires of male peers for sadness, anger, frustration, worry, hot-temper, stress, and
crying behavior (ps > .17); and self-desired less distancing responses than believed of male
peers for embarrassment, fright, and hitting behavior (ps < .05).
Further examination of the bivariate correlations between self-reported desires
(coded 1 for social comfort and 2 for social distancing) and inferred social desires of same-
sex peers confirmed a male advantage at within-sex inferences. Males evidenced significant
correlations between self-reported desires and inferred desires of male peers for each of the
8 mood states and crying behavior (rs ranged from .19 to .42, ps < .05). In contrast, female
participants’ only evidenced significant correlations between self-reported desires and
inferred social desires of female peers for hot-temper (r = .37, p < .01), frustration (r = .30,
p < .05), and hitting behavior (r = .25, p < .05); for the remainder of the mood states and for
crying behavior, the correlations were non-significant (ps > .05).
Discussion
According to the socio-relational perspective, sex differences in distress behaviors,
such as higher levels of physical aggression in males and crying behaviors in females may
reflect the selective advantage of advertising the capacity (via dominant behaviors) or
Evolutionary Psychology – ISSN 1474-7049 – Volume 6(3). 2008. -515-

---

Document Outline

• ??
• ??
• ??
• ??

---