A Critical Examination of the Photofit System For Recalling Faces

Abstract

Three experiments are reported which investigate the Photofit System for face recall. In the first experiment it was found that constructions of the videotaped face of a man reading a factual text did not differ among groups of subjects who saw the face for either 15 s or 2-5 min. Further, instructions to attend to face or to the passage resulted in equally good constructions although those who attended specifically to the passage scored significantly higher on a subsequent comprehension test. The second experiment employed an experienced Police Photofit operator as well as our own research operator. Both supervised compositions made while the videotape ran during construction. The results from the two operators' subjects did not differ. Nor were these constructions judged to differ from those made from memory in experiment 1. Finally, the ability of subjects to sketch faces was compared with their ability to construct them under two conditions, working from memory after a 10 s exposure to the face, or recalling in the presence of the target face. There was a slight advantage to Photofit in the memory condition and a larger advantage to sketching when the face was present. Again there was no difference between Photofit constructions made from memory and those made while the face remained on view. These observations are attributed to an inherent limitation in the efficiency of the Photofit System.