Set-size effects for complex objects in developmental prosopagnosia: Visual search for faces, words and cars

Whether developmental prosopagnosia is specific to faces is an ongoing controversy. One difficulty that may contribute to discrepant results is differences in tests used to assess face and object processing. Here, we devised a search test to measure the efficiency of processing for faces, cars, and words. We had 14 subjects with developmental prosopagnosia and 26 controls perform three search tests. For face-search, subjects saw an array ranging from three to seven faces and had to answer whether all depicted the same person or if one was a different person. Similar tests were constructed for cars and words. The main variables were response time and the set-size effect, which refers to the additional time required for every additional stimulus in the array. At the group level, the prosopagnosic subjects were impaired in face search but no different from controls on car or word search. However, at a single-subject level, only 20% had evidence of a putative classical dissociation to support claims that their recognition disorder is face-specific, while 30% had abnormally prolonged car-search times, and 50% had indeterminate results. The results align with a meta-analysis that concluded there was heterogeneity in object recognition in developmental prosopagnosia.