The role of familiarity in view transferability of face identity adaptation

The aftereffects induced by complex shapes like faces have provided insights into the nature of high-level visual representations. Previous studies have shown that face adaptation survives two-dimensional affine transformations. More recent studies report varying degrees of three-dimensional view-transferability for high-level aftereffects, with divergent interpretations (Jiang et al., in press; Jeffery et al., in press). The viewpoint transferability of face adaptation, however, may depend on the familiarity of the face—with more familiar faces having more robust representations (Roark et al., in press). In the present study, we tested the role of familiarization in the view transferability of face identity adaptation. During learning, we varied participants' experience with the original faces and their .35-level anti-caricatures. The familiarity manipulation consisted of three pure repetition conditions (2, 4, or 8 exposures) and one multiple-view presentation condition (4 exposures to the frontal view and 4 exposures to the 30-degree view). In all conditions, we measured the effect of “anti-face” adaptation on identifying anti-caricatures of faces (Leopold et al., 2001). We tested identification of faces from the .10-level anti-caricatures, both within and across-viewpoint. High levels of familiarity, developed from 8-exposure repetition and the multiple-view presentation conditions significantly improved the view-transferability of face identity adaptation. This indicates that a more robust representation of faces can be developed from familiarization, which in turn benefits recognition across viewpoints.