UCLA-UCSB Conference on
Human Nature and Society
Spring Quarter
Location: UCLA Anthropology Department
352 Haines Hall, UCLA
(driving directions below)
Domain-Specific Enhancements and Reversals
in the Actor-Observer Effect and the Fundamental Attribution Error
Much research on the fundamental attribution error and the actor-observer effect suggests that people attribute others’ behavior, more often than their own behavior, to dispositional rather than situational factors. These effects are assumed to be domain-general, applying equally to the attribution of all kinds of attitudes and personality traits. However, based on social contract theory (Cosmides, 1989) and reasoning about deep engagement relationships (Tooby & Cosmides, 1996) it seems likely that people should be particularly vigilant about inferring dispositional traits associated with dishonesty to potential cheaters as well as attributing honesty to deep-engagement partners. It was therefore hypothesized that dispositions associated with dishonesty would display stronger FAE and actor-observer effects than other negative traits. It was also hypothesized that for dispositions associated with honesty, the FAE and actor-observer effects would be reversed. These hypotheses were tested in two experiments using methodology adapted from previous attribution research in which participants rated traits of the self and others. As predicted, participants gave higher ratings to others on dishonesty than the self, whereas ratings for the self and acquaintance were similar for other negative traits showing a stronger actor-observer effect for dishonesty. The reverse pattern was observed for positive traits. Comparing attributions for friends and acquaintances showed that the FAE is also trait-dependent as well as target-dependent. These results support the hypothesis that some attribution processes may be linked to evolved, domain-specific inferential mechanisms.
Dominance and Heterosexual Preferences Updated
Previous research has indicated that women preferentially select as mates men who are high in dominance, presumably because dominance may imply social status, good economic prospects, and ambition. However, if dominance is marked by aggressiveness, authoritarianism, oppression, and domineering behaviors, dominant mates may also pose adaptive problems for women and their children. Evidence suggests that women’s evaluations of potential partners vary according to the context of men’s dominant behaviors.
Human Mate Choice for Risky Behavior
Human
risk-taking shows strong sex differences, suggesting that risk-taking may
be sexually selected. We investigated whether young heterosexual women might
use risk-taking as a cue for mate choice. As measured on a domain-specific
risk inventory, women reported finding male risk-taking in recreational and
social domains attractive, perhaps because it acts in those domains as a cue
for physical quality and social status respectively. Risk-taking in three
other domains was unattractive. We found that men know which risks women find
attractive and unattractive. Finally, couples were positively assorted for
risk-taking, suggesting a "likes attract" rather than "opposites
attract" mate-choice mechanism.
Cheater Detection Under Uncertainty
Cheater detection under uncertainty, especially in social exchange conditions based on trust and voluntary contributions, is a cognitively complex but ecologically valid problem. Using game theory and an experimental design incorporating elements of trust, voluntary cooperation, and challenge –subject to punishment– I have developed a game of strategic interaction between anonymously paired individuals. Participants’ behavior in this game will inform predictions of cheating behavior and cheater detection under uncertainty. Because social exchange relations in the real-world are subject to reputational discrimination, I predict that initial game behaviors will be characterized by higher levels of challenges and lower levels of cheating than expected by income-maximizing predictions. Once reputations are established, I predict that subjects attempting to maximize their returns and minimize the other person’s cheating behavior will engage specific intuitive statistical tools designed to compare observed distributions of claims to expected distributions of returns. Because, as I argue, humans do not act randomly, I expect observed responder strategies to differ significantly from a normative strategy based on a proposer’s random strategy.
Male Muscularity as a Good-Genes Indicator:
Evidence from women’s preferences for short-term
and long-term mates and reports of past sexual behavior
Prior research suggests
that some traits (e.g., facial masculinity and symmetry) may be indicators
of good genes. We propose that male muscularity is also a sexually-selected
indicator of good genes. We predicted that women would prefer greater levels
of muscularity in short-term mates than in long-term mates. We also predicted
that women would perceive muscular men as more dominant, sexually attractive,
and less investing in their long-term relationships than non-muscular men.
In a series of 4 studies, we examined (1) women’s preferences for muscularity
in a short-term and long-term mate, (2) women’s ratings of the personality
traits of muscular and non-muscular men, (3) women’s ratings of the muscularity
of their past short-term and long-term mates, and (4) the relationship between
men’s self-reported level of muscularity and their past sexual behaviors.
The results to date provide preliminary evidence for male muscularity as a
good-genes indicator.
3:30–4:00
The Form and Function of the Quarterly UCLA-UCSB Conference Series
Sponsored
by the
Map of the UCLA campus (Haines Hall in northeast quadrant at coordinates F, 3) - Driving directions.