Abstract
Previous studies have demonstrated that perceptual training can improve face processing in individuals with developmental prosopagnosia (DP). However, training face memory has only been attempted in a few prosopagnosia cases, with mixed success. We recently showed that face recollection, the all-or-none recognition with associated contextual information, is deficient in DPs. Further, we have found evidence for the existence of two subgroups of DP: perceptually impaired (PI-DP) and perceptually unimpaired (PU-DP), suggesting that memory impairments may underlie impairment in the latter group. To investigate whether training could improve face recollection and whether this differed between DP subgroups, in the current study we used a novel four-week-long repetition-lag training program targeting face recollection following four weeks of perceptual training in 14 individuals with DP (Age: M=42.8, SD=12.9). Participants completed an extensive battery of tests assessing face perception and memory prior to the eight weeks of training and assessment immediately after as well as six weeks post-training. Immediately following training, we observed significant group-level improvements in the Cambridge Face Memory Test (Pre: M=.54, SD=.09, Post: M=.61, SD=.13; p=.002). Face recollection in a remember/know task improved numerically but not significantly (Pre: M=.51, SD=.38, Post: M=.75, SD=.25; p=.088). Notably, PU-DPs showed a numerical improvement in face recollection (Pre: M=.59, SD=.43, Post: M=.88, SD=.23) that reached levels comparable to the recollection mean of typically developing controls (TD: M=.88, SD=.12). PU-DPs also showed significantly improved source recall on the trials where they correctly remembered a face (Pre: M=.63, SD=.34, Post: M=.95, SD=.09; p=.049). Face recollection improvements persisted in the entire sample at the 6-week post-training follow-up (M=.80, SD=.33). Together, these findings provide important 'proof of concept' that repetition-lag training can improve face recollection, a critical memory deficit in DPs. The effects were particularly apparent in PU-DPs and suggest applying different treatment approaches to PI-DPs vs. PU-DPs.