Facial Composite Techniques May Hinder CCTV Identification

Main Category: Psychology / Psychiatry
Article Date: 21 Nov 2008 - 5:00 PST

email to a friend   printer friendly   view / write opinions   rate article

Composite photographic images, similar to that used in the 'de Menezes case', may hinder rather than help suspect identification. This is the finding of a study presented today, 21st November, at The British Psychological Society Scotland's Annual Conference.

Ailsa Strathie and Dr Allan McNeill from Glasgow Caledonian University asked 24 people to look at a series of photo pairs. The participants had to decide whether the photos in each pair were images of the same person, or of two different people.

Each participant viewed 150 pairs, and there were three different types of image pair. Fifty pairs were made up of two full-face photos, 50 pairs were made up of two separated half face photos, and the final 50 photo pairs were composite images - where the right half of one photograph and the left half of another were combined to produce one image. Half of the photo pairs in each condition consisted of two different photos of the same person, and half the pairs consisted of photos of two different people.

Ailsa Strathie said: "As you might expect, we found overall accuracy was highest when participants made identity judgements from two full face photos, rather than from either type of half face photo pair. Furthermore, when participants were shown two half-face photos combined to form a composite, they were far more likely to make errors.

"In this condition, participants only correctly identified that the photos were of two different people in 57% of trials, compared with 87% in the full-face condition. This suggests that by combining photos in a composite you reduce the chance that photos will be correctly judged.

"This study is important because within the legal system, there is growing acceptance of the use of facial composite techniques to establish a person's identity from CCTV footage. Our results cast doubt on their reliability and we support the need for these procedures to be investigated scientifically," Ailsa concluded.

The British Psychological Society