Effects of spatial caricaturing and anti-caricaturing on face learning

Abstract

Introduction The learning of facial identities can be influenced by several factors. In addition to the learning situation and individual abilities of the observer, the face stimulus itself can facilitate or impede learning. According to Multidimensional Face Space (MDFS [1]) Models, faces can vary on several dimensions and are characterized by their individual position in this n-dimensional space (for a simplified example, see Fig 1). Similar faces lie close to each other and typical faces cluster around an average or the origin. Distinctiveness refers to the fact that a face lies outside this cluster, due to its deviation on one or more dimensions, and therefore pops out. Distinctiveness can be artificially manipulated by spatial caricaturing [2] or anticaricaturing, which increases or decreases spatial differences relative to an “average” face. As the processing of familiar and unfamiliar faces is supposed to be qualitatively different [3], and distinctiveness seems to improve especially the processing of unfamiliar faces [4], we aimed at exploring the effects of increased and decreased spatial distinctiveness on unfamiliar face learning and its neural correlates (ERPs). To measure real face learning instead of picture learning, we used different pictures of the same person at learning and test.