Facial composite production : the usability of therReverse correlation image classification technique for perpetrator recognition

Current identification techniques are based on the reproduction of memory in two ways; (1) By letting an eyewitness actively describe and select features (recall) to create a face composite or (2) by letting an eyewitness recognize someone in a lineup (recognition). As a result of holistic processing, recall of features is poor. This research tries to construct a composite by matching the preferred retrieval method to the holistic information. We test the usability of the reverse correlation image classification technique for composite purposes. In the first part of this study participants constructed face composites of either a Caucasian or a Moroccan target perpetrator. Resemblance judgments of independent participants showed that the constructed face composites resembled their targets well. Furthermore, in a line-up the majority of the participants selected the correct perpetrator. Recommendations for future research are suggested.