Holistic Facial Composite Construction and Subsequent Lineup Identification Accuracy: Comparing Adults and Children

ABSTRACT

When the police have no suspect, they may ask an eyewitness to construct a facial composite of that suspect from memory. Faces are primarily processed holistically, and recently developed computerized holistic facial composite systems (e.g., EFIT-V) have been designed to match these processes. The reported research compared children aged 6–11 years with adults on their ability to construct a recognizable EFIT-V composite. Adult constructor's EFIT-Vs received significantly higher composite-suspect likeness ratings from assessors than children's, although there were some notable exceptions. In comparison to adults, the child constructors also overestimated the composite-suspect likeness of their own EFIT-Vs. In a second phase, there were no differences between adult controls and constructors in correct identification rates from video lineups. However, correct suspect identification rates by child constructors were lower than those of child controls, suggesting that a child's memory for the suspect can be adversely influenced by composite construction. Nevertheless, all child constructors coped with the demands of the EFIT-V system, and the implications for research, theory, and the criminal justice system practice are discussed.

Keywords:

• eyewitness identification
• child witnesses
• holistic facial composite
• EFIT-V
• video lineup

Author Notes

Josh P. Davis is a senior lecturer in the Department of Psychology, Social Work, and Counselling at the University of Greenwich. His research interests include eyewitness identification and individual differences in face recognition ability. Sarah Thorniley is a former student of the University of Greenwich who currently works in the Faculty of Architecture, Computing, and Humanities at the university. Stuart Gibson is a Lecturer in the School of Physical Sciences at the University of Kent. His research interests include facial identification, digital image and signal processing, and machine learning. Chris Solomon is a Reader in the School of Physical Sciences at the University of Kent. His research interests include digital image processing, evolutionary methods, face models, and search optimization techniques.

Acknowledgments

This research was approved by the University of Greenwich Research Ethics Committee following guidelines issued by the British Psychological Society. Parts of this research were presented at the Identifying the Suspect: Improving Facial Composites Workshop, Institute of Psychological Sciences, University of Leeds, January 2013.

The authors would like to thank the following for their help with data collection: Andreea Maigut, Rebecca Fell, Ima Fagerbakke, Ionela Jurj, Michael Sansom, Corrado Ranalli, Beckie Hogan, and Kelty Battenti. They would also like to thank the anonymous reviewers of an earlier version of this article.

Solomon and Gibson are faculty members of the University of Kent and directors of VisionMetric Ltd. VisionMetric Ltd. markets the EFIT-V and E-FIT facial composite systems. Solomon and Gibson's contribution to this work was to train the operator, the development of the software, and provision of software support. Data collection, analysis, and interpretation were performed by Davis and Thorniley.

Notes

1. A Pearson's correlation test found a strong positive relationship between the composite-suspect similarity ratings provided by the suspect-acquaintance assessors, and those provided by the suspect-unfamiliar assessors to the 57 EFIT-Vs, r(57) =.89, p <.001, suggesting that regardless of familiarity, these ratings measured the same construct and therefore for brevity the two groups were pooled for all further analyses.

2. A Pearson's correlation test found a strong positive relationship between constructors’ self-assessments of EFIT-V recognizability and their belief their composite would be subsequently recognized, r(57) =.52, p <.001. As these data were therefore essentially measuring the same construct, these data were combined for all further analyses by calculating each constructor's mean composite-suspect likeness rating from the two scales.

3. The composite-suspect similarity ratings significantly differed between the EFIT-Vs constructed of the two suspects, suggesting it may have been easier to construct a likeness of one than the other. However, this suspect variable did not interact with any other on any analyses and therefore these suspect data were pooled.

4. All reported analyses on lineup outcomes were conducted with and without the inclusion of the three “not sure” response participants. There were no differences in conclusions.