With the advent of computer systems, there is an opportunity to create an ever more sophisticated method of reconstructing faces to aid the police (Bruce, Frowd, and Hancock). However, such devices while impressive in their ability to create something that looks like someone they still depend upon the reliability of the minds cognitive processes to accurately recall unique information about the features of the face, was that a Roman nose, were they olive shaped eyes? Research suggests humans have a natural inclination to process and recognise faces above all other information from an evolutionary perspective this makes a lot of sense. However reconstructing from recall those faces does not come as easily. Factors such as ‘Weapon Focus’ (Loftus, 87) and post witness identification influences such as confirming feedback (Wells and Bradfield 1998) call into question the accuracy of any eye-witness testimony. Pawan Sinha published an influential article on key factors that impact upon the accuracy of facial reconstruction read it here. Try a facial recognition test here.
The variability of someone’s ability to accurately recall a face can be seen evidently below…..
Believe it or not the above image contributed to the perpetrator being caught…
The difficulty of constructing faces from our memory has been known for over 30 years (e.g. Davies, 1978). We are not good at the tasks required – describing and selecting individual facial features – instead we process faces ‘holistically’, more as a complete entity (e.g. Young et al., 1987). For example, the perception of facial features changes in the presence of other features (e.g. Tanaka & Farah, 1993), and so the features and their position on the face are both important. Modern facial composite systems, where witnesses choose individual features in the context of a complete face, apply this idea to some extent.
However Evofit can and does prove accurate and, therefore useful, and the infamous ‘Beat of Bozeat’ case was such an example. Frowd and Bruce used this as part of their research.
The use of ‘Holistic’ based software such as EvoFit allows factors such as trustworthiness or aggressiveness can be added in the face as Sinha identifies recall of faces tend to be greater when we have associated an emotional component to them.
Watch below how the Police are using ‘EvoFit’ recognition, of faces rather than actively trying to recall features in their work with witnesses.
Frowd & Bruce also conducted some different research attempting to investigate the importance of internal and external features when recreating faces. Do we have cognitively process internal and external features differently based upon our familiarity of the individual or is there a difference in accuracy just because internal features are more difficult to replicate? –Read the study here.