Face Identity Tests and Inclusion Criteria in Self-Identified Acquired Prosopagnosia

Prosopagnosia is characterised by severe difficulties recognising facial identity. While many self-identified developmental cases fail to meet widely used cognitive task-based criteria for inclusion in research, the extent to which this is an issue in acquired prosopagnosia is unknown. We therefore assessed the sensitivity of widely used inclusion cutoffs and face-processing tests in 14 self-identified British acquired prosopagnosia cases.

We found the -2 SD cutoff did not detect impairment in 9/11 (82%) individuals on the Cambridge Face Perception Test, 5/11 (45%) of cases on the Cambridge Face Memory Test, and 2/14 (14%) on the Famous Faces Test. Requiring impairments on two memory tasks at -2 SD or -1 SD resulted in at most 8/14 (57%) and 5/14 (36%) of cases not meeting the criteria, respectively. A more liberal approach, which additionally allowed for inclusion using a single test score at -2 SD, performed better by excluding at most 2/14 (14%). Given that the Famous Faces Test alone at the -2 SD level performed identically to our liberal approach, we suggest it could be used if an objective assessment is deemed necessary.

We extend the recent suggestions in the literature that self-identified prosopagnosia cases who fail to meet cognitive test criteria warrant further investigation to explain the reasons for their self-reported status. We therefore advocate a more inclusive approach to prosopagnosia research.