Abstract
It is widely believed that face identity is achieved via holistic processing of the representation integrated over parts of a face. In contrast, processing of facial expressions can be undertaken via analytic processing based on representation of separate parts. The composite face task (CFT) is a commonly used paradigm to investigate the underlying mechanism of face processing. In the most diagnostic condition of CFT, two identical top halves joined by two different bottom halves are difficult to judge whether or not they are identical when the top and bottom halves are aligned than when they are misaligned. Such failure of selective attention has been construed as evidence of holistic processing for processing both face identity and facial expression. However, using the CFT, some recent studies have argued that both identity and expression are processed analytically. Hence, whether identity and expression of a face requires analytic or holistic processing remains controversial. To resolve the controversy, the present study combined the redundant target design and the CFT within the framework of systems factorial technology (SFT) and investigated the nature of capacity coefficients for processing identity and expression. In Experiments 1 and 2, participants were asked to employ either the self-terminating rule or the exhaustive stopping rule for discriminating face identities. In Experiment 3 and 4, they were asked to adopt the two stopping rules for discriminating facial expressions, respectively. Results of Experiments 1 and 3 revealed that a majority of participants exhibited unlimited or limited capacity, which is consistent with analytic processing. In contrast, results of Experiments 2 and 4 revealed super capacity for most participants, which is consistent with holistic processing. Taken together, these findings suggest that the face identity and facial expression can be processed either holistically or analytically depending upon the use of specific stopping strategy.
Meeting abstract presented at VSS 2017