A New Test for Individual Differences Research in Face Recognition

 

Outcome:
Researchers in the NSF-sponsored Temporal Dynamics of Learning Center have developed a new task for measuring individual differences in holistic face processing, The Vanderbilt Holistic Face Processing Test (VHPT-F). Holistic processing measured in this new test correlates with the standard measure of holistic processing, but shows better psychometric properties (e.g., better internal consistency and test-retest reliability), providing a useful tool for studying face recognition and individual differences in high-level vision.

Impact/Benefit:

Holistic processing is a key behavioral signature of expert face perception. In the standard task used to measure holistic processing, participants are instructed to attend to one face half (e.g., top), while ignoring the other face half (e.g., bottom), but are unable to do so; the to-be-ignored half interferes with performance because faces are processed as wholes. While the standard task performs very well in group studies, it has poor psychometric properties for measuring individual differences. The VHPT-F is freely available (http://gauthier.psy.vanderbilt.edu/resources) to facilitate work on individual differences in high-level vision, and can easily be adapted for use with clinical populations.

Background/Explanation:
Tasks used in group studies are designed to achieve an optimal level of difficulty, usually to avoid floor and ceiling performance levels. To create a more reliable measure better suited for individual differences work, VHPT-F trials vary in difficulty, and in the extent to which they should tap into holistic processing, allowing more precise measurement of holistic processing across the full continuum of ability. For example, trials where the target part is a very small part of the face (e.g., eyes only) should be processed holistically by most subjects except those with the least holistic tendencies; trials where the target part is a very large part of the face (e.g., top 2/3) should not be processed holistically except by those with the most holistic tendencies.


Richler, J.J., Floyd, R.J., & Gauthier, I. (2014). The Vanderbilt Holistic Face Processing Test: A short and reliable measure of holistic face processing. Journal of Vision, 14, 1-14.
Wang, C.-C., Ross, D.A., & Gauthier, I., & Richler, J.J. (in preparation) Validation of the Vanderbilt Holistic Face Processing Test.