Background: Many studies have reported an association between observers' self-attractiveness and their preference for sexual dimorphism across different physical domains, including the face, voice, and body. However, the results of these studies are inconsistent. Here, a meta-analysis was conducted to estimate the association between observers' own attractiveness and their dimorphic preference. Methods: Major electronic databases including PsycINFO, Web of Science, PubMed, ProQuest, and Google Scholar were searched during April 2017 (the first time) and April 2018 (the second time). The effect size computation and moderating effect analyses were conducted separately for masculine and feminine preferences. Results: We identified 5,359 references, of which we included 25 studies (x = 55, x = number of the effect size) with 6,853 participants in the meta-analysis. Across these studies, the correlation between observers' own attractiveness and their sexual dimorphic preference was 0.095 (x = 55) and that for preference for masculinity (x = 39) and femininity (x = 16) were 0.102 and 0.076, respectively. The results of the funnel plot, Egger's regression method, and fail-safe number suggested that there was no obvious publication bias. The relationship depended on the relationship context (short or long-term), opposite or same sex (the gender of the observer and host), measures of observers' self-attractiveness (subject or objective), and preference task (e.g., attractiveness rating, forced-choice, and face sequence test). Furthermore, for female participants, using a hormonal contraceptive also influenced their masculinity preference. The effect size for the preference for a masculine body and voice was larger than that for facial masculinity. Conclusion: We found a small but significant correlation between self-attractiveness and physical dimorphic preference, the relationship was moderated by the relationship context, same/opposite-sex, and contraceptive using. These three moderating effects represented the observer's trade-off on good genes, good provider and good father (3Gs) consistent with the life history strategies. Besides, measurement of observers' attractiveness, type of preference task and stimuli may also involve the relationship.
Keywords: femininity; masculinity; meta-analysis; self-attractiveness; sexual dimorphism.