Abstract
In a typical composite face task, participants judge whether the top (bottom) of a test face is the same as the top (bottom) of a study face, while ignoring whether the bottom (top) is the same or different. The standard finding is that the irrelevant part reliably affects responses about the relevant part. The question that arises is whether this reflects some type of perceptual holism. Wenger and Ingvalson (2002, 2003) used a different approach to this question. In their studies, participants simultaneously judged whether both the top and bottom of a test face were the same or different from the study face (a complete identification task). This task allowed them to analyze their results using the tools of general recognition theory (GRT, Ashby & Townsend, 1986), which makes a distinction between perceptual and decisional sources of stimulus integrality. Wenger and Ingvalson found consistent evidence for decisional influences, with very limited evidence for perceptual holism. The present experiment relates the composite face task to the complete identification task empirically and theoretically. Three participants completed five sessions of both a complete identification task (judging both parts simultaneously) and a variant of the more standard composite task (judging only one part at a time). Results for all participants for both tasks show consistent violations of decisional separability, with some violations of perceptual separability. Critically, there was no strong evidence for violations of perceptual independence. This work suggests needed refinements of the concept of holism in face perception and memory.
This work is supported by a grant from the James S. McDonnell Foundation to the Perceptual Expertise Network