Ensemble Face Adaptation Effects: Evidence from RSVP

Abstract: In our daily lives, we are often confronted with a plethora of diverse and intricate adaptive stimuli. Prior research has predominantly focused on investigating the processing mechanisms of individual faces from the perspective of adaptation effects, yielding relatively consistent findings. However, controversies persist regarding the formation process of average representations in ensemble face adaptation effects, accompanied by a dearth of comprehensive explorations into ensemble face processing mechanisms. Attractiveness, as a crucial and distinct dimension in face perception, exhibits a phenomenon known as the averageness attractiveness effect, wherein faces perceived as more average tend to be rated as more attractive. This effect serves as a tool to examine the generation process of ensemble face adaptation effects and to further elucidate the processing of ensemble faces. This study employed a rapid serial visual presentation paradigm to investigate ensemble face adaptation effects from the perspective of cognitive load, aiming to advance our understanding of ensemble face processing mechanisms. Experiment 1 employed a within-subject design featuring three adaptation conditions (high attractiveness faces, low attractiveness faces, mixed faces). Results indicated that after adapting to low (high) attractiveness faces, individuals exhibited increased (decreased) judgments towards new faces, thus demonstrating adaptation effects. This outcome substantiates the notion that individuals process ensemble faces via holistic averaging. In Experiment 2, cognitive load manipulation was introduced with a mixed design of 2 (adaptation type: high attractiveness faces, low attractiveness faces) × 2 (load level: high load, low load). Results revealed that under both high and low cognitive load conditions, individuals exhibited adaptation effects after adapting to low (high) attractiveness faces, thus indicating that individuals employ an automated processing mechanism for ensemble faces. This study, by utilizing rapid presentation of single faces to form face ensembles, demonstrated that under conditions of rapid serial presentation, adaptation effects are based on the holistic averaging of ensemble faces, unveiling the processing of ensemble faces. Furthermore, by manipulating cognitive load, this study for the first time investigated the processing mode of ensemble faces and the role of cognitive resources in ensemble face processing. The findings corroborate the existence of ensemble face adaptation effects even under conditions of limited cognitive resources, suggesting that individuals process ensemble faces in an automated manner, thereby contributing to a more comprehensive understanding of face processing research.