National security and crime prevention sometimes rely on the ability of security personnel to establish the identity of individuals and to check they are who they claim to be. Identity is often established using photo ID documents such as passports, but past research has demonstrated that the job of checking the validity of photo ID documents is more difficult than most people imagine. This is a concern, as at a border crossing point, a failure to detect the fraudulent use of invalid photo-ID could have serious consequences and may even represent a risk to national security.

One of the reasons this task is so difficult is that it requires the operator to match two images of unfamiliar faces, and research has demonstrated that when looking at unfamiliar faces we attend more to the external features of the face such as hairstyle, and less to the internal features such as the eyes. This is problematic because the external features are more easily changed and are therefore an unreliable cue to identity.

Kemp, Caon and Brooks have shown that when participants are required to decide whether two images are of the same person, they are more accurate when only the internal features of the faces are shown than when the whole faces are visible. This may be because when the external features are masked the participants are forced to compare the internal features of the faces which are more reliable indicators of identity. This research could have a practical implication for the training of security personnel such as those validating passports at airports.

Research Contact

Dr Richard Kemp (r.kemp@unsw.edu.au)

Homepage Articles

Click here for a complete list of homepage articles.