Domain Specificity in Face Perception

Introduction

Extensive behavioral and neural evidence suggests that special mechanisms are engaged in the processing of faces. However, the extent to which the FFA is specific for faces and the precise nature of the processing that it carries out remain matters of active debate. The goal of the present study was to investigate whether apparently face-specific mechanisms are domain specific (i.e., engaged by faces, regardless of the type of processing), process specific (i.e., engaged in a specific process that can be applied to any stimulus class), or both stimulus and process specific (i.e., engaged in a specific process that is applied only to faces). To investigate these questions, we made use of the two most well-established markers of face processing.

Discussion

A serious effort to induce face-like processing of nonface stimuli requires a good hypothesis about how faces are processed. Here we test the widespread view that face perception critically involves configural processing, that is, the precise distances among face parts. Thus, our study specifically tested whether the FFA extracts configural information, rather than part-based information, from faces and non-faces.

Because this inversion effect is larger for faces than for many nonface stimuli, it has often been used as a behavioral marker of face-specific processing, in much the same way that the FFA has served as a neural marker of face-specific processing. Similarly, in this study we tested the domain and process specificity of face processing mechanisms by asking whether the inversion effect (i.e., the difference in performance between upright and inverted faces) is larger for faces than for houses, but comparable for the configuration and part tasks; larger for the configuration versus the part task, but similar on the face and house tasks; larger for faces than houses and for the configuration versus the part-based task; or larger for configuration versus the part-based information for faces only (Kanwisher).

It has been suggested that the face inversion effect primarily reflects the loss of configural information when faces are inverted. Studies that have tested this hypothesis have reported a larger inversion effect on configuration than part-based matching tasks on faces (Hong & Avi). However, no study to date, to our knowledge, has tested whether the inversion effect for configural information is specific for faces or whether it might occur when such information is extracted from nonface stimuli (Reyes). This is the critical condition required to test the domain-specificity and process-specificity hypotheses.

Almost all the authors who speak of the modularity of mind human, or at least of the modularity of certain parts, the specificity domain appears as a necessary condition of it, everyone seems to agree that: a cognitive mechanism is modular only if a domain-specific where domain specificity, it is usually understood in terms of certain types of restrictions on the kinds of information that this mechanism accepts (Murray et al.). So additional information, the psychology of human reasoning in the last fifteen years have achieved some important experimental results suggest that some of cognitive mechanisms underlying human inferential abilities basic cas are ...