Kin recognition signals in adult faces , Vision

10 Maloney and Dal Martello (2006) reported that similarity ratings of pairs of related and unrelated children were almost perfect predictors of the probability that those children were labeled as being siblings by a second group of observers. Surprisingly, similarity ratings were not good predictors of whether a sibling pair was same-sex or opposite-sex or how close a pair was in age, suggesting that people ignore cues that are uninformative about kinship when making similarity judgments of faces. Here we replicate this study using two sets of adult sibling pairs. In both sets, similarity ratings were very good predictors of the probability of being judged siblings. In contrast to the findings for child faces, similarity ratings for same-sex pairs were significantly higher than for opposite-sex pairs, suggesting that similarity judgments of adult faces are not entirely synonymous with kinship judgments. Additionally, Dal Martello and Maloney (2006) found that the kinship information observable in either the upper and lower halves of the face alone predicted the information observable in the full face. They concluded that the spatial relationship between features in the upper and lower halves of the face (configural information) is not used in kinship judgments. However, here we find evidence suggesting that redundant kinship information exists in the upper and lower halves of the face, calling this previous interpretation into question.