Abstract
After adaptation to faces with contracted (or expanded) internal features, faces previously perceived as normal appear distorted in the opposite direction. These face aftereffects suggest that face-coding mechanisms adapt rapidly to changes in the spatial relations of face features and/or the global structure of the face. What role does selective attention play in the face aftereffect? Past attempts to answer this question have yielded conflicting results (Moradi, Koch & Shimojo, 2005; Murray & Yan, 2006). We report work in which adaptation to ignored faces was tested under conditions of high attentional load and where attention to the non-face task was optimal. Before and after an adaptation phase, participants rated the normality of morphed distorted faces ranging from 50% contracted through normal to 50% expanded. In the adaptation phase, participants were presented with a highly demanding rapid serial visual presentation (RSVP) of red or blue Xs and Ts flanked to the left or right by a face with 50% contracted features. The ten individual letters in the RSVP stream were each presented for 160 ms and separated by a 20 ms inter-letter interval. Forty participants either attended to the adaptation faces (attend condition), or ignored them and counted the number of red Ts in the RSVP stream (ignore condition). A monetary incentive was provided to encourage full attention to the letter-count task in the ignore condition. A reduced but significant face aftereffect was observed when adaptation faces were ignored. These results suggest that face adaptation, as a possible mechanism for rapid updating of what looks normal, occurs automatically even under conditions in which general capacity demands are very high.