Abstract Detail

 

Talk Sessions

Poster Sessions

All Presentations

Symposia

Holistic processing of faces in the composite task depends on size

33.577, Sunday, 18-May, 8:30 am - 12:30 pm, Pavilion
Session: Face perception: Whole and parts

David Ross1, Isabel Gauthier1; 1Department of Psychology, Vanderbilt University

Faces, more than other objects, are processed holistically. There is some evidence that holistic processing is strongest for faces at a conversational distance, dropping off for faces that are further away (McKone, 2009). However, the only previous study of this question used tasks that have not been directly related, theoretically or experimentally, to other measures of holistic processing. Here, we use a well validated and frequently used paradigm – the composite task – to measure the effect of viewing distance on holistic processing. Participants (n=99) judged whether halves of sequentially presented face composites matched, ignoring the other half, which could be congruent or not with the correct response. Holistic processing was defined as the congruency effect for aligned relative to that for misaligned parts. We used a within-subjects design with 4 blocked levels of size: the largest face size corresponded to a viewing distance of 2 meters, and the smallest face size corresponded to a distance of 24 meters. Holistic processing varied linearly with size (Figure 1), as supported by an interaction between alignment, congruency and linear contrast for face size, F(1, 98) = 4.36, MSE = 1.34, p <0.05, ηp2 = 0.04. There was no holistic processing at the smallest size (visual angle = 0.35o; ηp2= .000; p=.77) and considerable holistic processing at the largest size (visual angle = 4.14o; ηp2= .09; p=.003). The interaction of congruency with size was entirely driven by the aligned condition (ηp2= .05; p=.03) and was not observed in the misaligned condition (ηp2= .005; p=.47). Size influences the magnitude of holistic processing in the composite task, suggesting it should be considered when comparing across studies and carefully controlled in individual differences studies. More work will be necessary to understand if this size effect is due to experience or general constraints of selective attention.

< Back