Abstract
People can make judgments about ensemble properties of complex object features such as the average of facial identities and the diversity of car models. However, experiences of ensemble judgments for specific object categories seem to have little influence on the performance. In my previous study (Cha, Blake, & Gauthier, 2021), participants were as good at judging the diversity of car models as at judging the diversity of facial identities, even though it was unlikely that they had prior experiences of ensemble judgments for cars. The present study sought to extend the previous finding by testing whether and how much behavioral variance was shared between ensemble judgment abilities for familiar object identities (i.e., facial identities) and novel object identities (i.e., random silhouette shapes), using an individual differences approach. In Study 1, I validated the mode (the most frequent object of an ensemble) judgment task as a viable ensemble judgment task by testing the partial correlation between participants’ diversity judgment ability and mode judgment ability: I found a positive correlation (r=.45) after controlling for face recognition ability. In Study 2, I found a positive correlation (r=.38) between participants’ mode judgment ability for facial identities and novel object shapes after controlling for face recognition ability. Furthermore, a Bayesian analysis using BFpack (Mulder et al., 2019) favored the hypothesis that face mode judgment ability (r=.30) and shape mode judgment ability (r=.28) were correlated with face recognition ability to the same degree, against the hypothesis that face mode judgment ability was more strongly correlated with face recognition ability. These results suggest that ensemble judgments for object identities (i.e., facial identities, object shapes) rely on shared mechanisms and that expertise on specific object categories does not have much influence on these judgments.