The spatial spread of information constrains face discrimination

Abstract

Previous work in our lab and others has suggested that humans use only a small proportion of the available information in face identification tasks. However, we still do not understand what limits our ability to encode information in faces. Here we examine whether observers' performance is affected by the spatial spread of information. We constructed pairs of faces that had equal amounts of discrimination information, but that varied in the spatial spread of that information (e.g., concentrated around the eyes, or spread across the entire face). Accuracy was consistently better when the information was spatially concentrated. The correlation between threshold and Syvajarvi et. al. (1999) measure of spatial spread (originally applied to contrast in Gabor stimuli) ranged from −0.71 to −0.82. Additional experiments demonstrated that varying stimulus contrast did not alter efficiency significantly when the information was spatially distributed, but that efficiency improved at high contrasts when information was concentrated. Together, these results suggest that the spatial spread of information places significant constraints on face discrimination.

Footnotes

 Supported by NSERC Discovery Grants 42133 & 1054994, the Canadian Research Chair Program, and an NSERC PGS-A Scholarship awarded to JSH.