| Abstract: | Contemporary face composite systems require scrutiny of many alternative features during the construction phase and judgments of similarity based on viewing features divorced from a facial context. The present study assessed the role of these factors in limiting composite production accuracy. In Exp I, 60 students observed a target face and made similarity judgments on sample eyes or mouths drawn from a Photo-fit Kit before attempting to identify the same features of the target. Exp II with 60 Ss was a replication, except that the interpolated judgments were made on the features in the context of a complete Photo-fit face resembling the target. In neither experiment did the Photo-fit components significantly interfere with recognition of the corresponding features of the target. In Exp III, the similarity ratings from memory provided by Ss in the previous experiments were compared with parallel sets of ratings furnished by 30 Ss in the presence of the target. Mean judgments of similarity made from memory on isolated features were discrepant from all other ratings. Results are interpreted as suggesting that whereas interference is not a major problem, judging resemblance from features seen in isolation may be a serious source of distortion in composite production. (20 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved) |
