Abstract
Introduction: Though individuals agree considerably on which faces they find more and less attractive, robust disagreements exist as well (Bronstad & Russell, 2007). Here, we investigate the origins of these disagreements, using a classic twin study to probe the relative contributions of genes and environments to idiosyncrasy of taste (IOT). Methods: A total of 284 twin pairs, 196 monozygotic (MZ) and 88 dizygotic (DZ), rated 200 faces on a scale of 1 ("very unattractive") to 7 ("very attractive"). We measured IOT as each individual’s divergence from the mean person’s ratings, or one minus the squared correlation of their ratings with the mean person’s ratings. In order to distinguish idiosyncrasy from inconsistent or careless responding, we repeated 60 faces and confirmed that results were robust to controlling for inconsistency of responses across the two repetitions. Results: IOT varied widely and reliably across individuals (mean=0.55, SD=0.12, reliability=0.90). As this reliability greatly exceeded the MZ twin correlation (0.26, p-value of difference <0.00001), a large amount of the reliable variation in IOT is attributable neither to the genes nor to the environmental factors that MZ twins share, but rather to the highly individual factors that make an MZ twin different from his or her co-twin. The MZ correlation was somewhat higher than the DZ correlation of 0.19, but the difference was not statistically significant, suggesting that genetic influence has, at most, a minor influence on IOT. Conclusion: We conclude that disagreements on the attractiveness of faces arise largely from highly individual past experiences of the sort that differ even between two monozygotic twins. We propose that whatever evolutionary forces operate on facial aesthetic preferences may favor the adaptability of these preferences to the particulars of one’s environment.
Meeting abstract presented at VSS 2013