Abstract
Faces are processed holistically and the composite paradigm is widely used to quantify holistic processing (HP) but there is debate regarding the appropriate design and measures in this task. Important theoretical conclusions hinge on which measure of HP is adopted because different approaches yield qualitatively different results. We argue that some operational definitions of HP are problematic because they are sensitive to top-down influences, even though the underlying concept is assumed to be cognitively impenetrable. Participants matched one half of two sequentially presented face composites while trying to ignore the irrelevant half. In the often-used partial design the irrelevant halves are always different, and HP is indexed by higher hit rates or d' for misaligned vs. aligned composites. Here, we used the complete design, which also includes trials where irrelevant halves are the same. We told one group of subjects that the target half would remain the same on 75% of trials, and another group that it would change on 75% of trials. The true proportion of same/different trials was 50% - groups only differed in their beliefs about the target halves. We assessed the effect of beliefs on three measures of HP: the difference in hit rate for aligned vs. misaligned trials (the standard measure used in the partial design), d' for aligned vs. misaligned trials based only on partial design trials, and the interaction between congruency and alignment, which can only be obtained from the complete design. Critically, beliefs influenced response biases and altered both partial design measures of HP while the complete design measure was unaffected. Thus, top-down biases, in addition to stimulus transformations (Cheung et al., 2008), can complicate partial design measures of HP. Many claims about face processing depend only on partial design measures should be re-examined with more valid measures of HP.
This research was supported by grants to the Temporal Dynamics of Learning Center (NSF Science of Learning Center SBE.