Abstract: There is a general consensus that mechanisms involved in face processing are "special". It is unclear, however, that what exactly they are specialized for. According to the face-specificity hypothesis, these mechanisms are selectively involved in processing faces per se. According to the expertise hypothesis, on the other hand, they are selectively involved in processing any class of visual stimuli that share the same basic configuration and for which the subject has gained substantial visual expertise. Here we reviewed recent behavioral, neuroimaging and neuropsychological studies concerning this issue, and we further proposed a novel approach to address this debate by orthogonally manipulating two factors that are centered in both hypotheses: cognitive process (holistic versus part-based analysis) and stimulus (faces versus nonface objects). Specifically, three new hypotheses derived from this manipulation were proposed to replace the original two for future experiments: a) faces are special (i.e. domain specific), b) the holistic process is special (i.e. process specific), and c) a hybrid of the first two: the holistic process of faces is special. Finally, we suggest that studies from genetic perspectives may provide final solution to this long lasting debate

Key words: face-specificity hypothesis, expertise hypothesis, faces processing, behavioral genetis