This blog post was written by Brad Duchaine and has been crossposted from the Psychonomic Society blog.
For every cognitive ability, there are individual differences, and in the new special issue of Cognitive Research: Principles and Implications (CRPI), the articles are focused on individual differences in face recognition. Given the universality of individual differences, the existence of differences in the ability to recognize faces isn’t news, but what has surprised researchers are findings in recent years showing that individual differences in face recognition ability run the gamut. These individual differences have implications in a number of important law enforcement, security, and legal settings, yet these implications have had little impact on practices in these areas.
People with developmental prosopagnosia have great difficulty recognizing facial identity. Most people fail to recognize faces from time-to-time, but developmental prosopagnosics frequently fail to recognize faces, including the faces of their family and friends, and sometimes cannot even recognize their own face in photographs. Below are a few quotes from developmental prosopagnosics who have contacted my research website (http://www.faceblind.org) that illustrate how poor their face recognition can be.
“While my son was in boot camp, I studied the Marine website daily because they would post photos from training. When my son came home, he was amused to see a picture of a strange marine taped to the refrigerator.”
“During a psychological evaluation for a child custody decision, I mistook a blonde girl for my daughter. The psychiatrist questioned my attachment to my daughter, and as a result, I lost the custody action. My children were abused by their stepfather, and I have carried guilt for the event ever since.”
On the other end of the distribution are super recognizers—they recognize faces seen only briefly years before. In fact, their abilities are so good that many super recognizers find it best to not reveal that they’ve recognized someone from a fleeting encounter, because it’ll appear they are overly interested in that person (imagine: “Hello, I saw you two years ago at a concert in the park”).
To take advantage of super recognizer’s skills, London’s Metropolitan Police created a unit of super recognizers. This small group of officers makes a disproportionate number of the Met’s suspect identifications and has inspired similar units in other police departments. It’s easy to imagine the advantages of employing super recognizers in other occupations that place strong demands on face recognition like passport officers and security screening personnel at airports.
In between these two extremes lie the great majority of people, and the differences in their abilities to recognize faces are stable and result primarily from genetic differences. The evidence for the heritability of face recognition comes primarily from several twin studies carried out in the last ten years. It’s unclear, though, what cognitive and neural factors drive the pronounced differences in face recognition ability. A number of factors have been found which modestly correlate with performance on face recognition tests; it seems likely that a substantial number of factors contribute to these individual differences.
The existence of such massive differences in face recognition raises many questions. We can ask fundamental questions such as:
- How are different face processing abilities related to each other?
- Can face recognition be improved with training?
- How can we take advantage of individual differences to learn about the organization and operation of the mechanisms that carry out face processing?
Because face recognition is critical in many security and law enforcement roles, differences in face recognition ability must have major effects on job performance. In addition, face recognition is critical in consequential decisions like eyewitness identifications in legal proceedings. As a result, numerous use-inspired research questions need to be addressed by psychologists:
- What tests should be used to identify super recognizers in police departments?
- How much more effective would passport inspections be if face recognition ability was considered when selecting border control officers?
- Can assessments of face recognition be used by legal systems to allow juries and judges to better gauge the credibility of eyewitness identification?
- What are the best procedures for combining judgments about facial identity from different individuals?
For most abilities, practice improves performance. Is that true for face recognition? In a study published in 2014, David White and his colleagues investigated this question with passport officers. The passport officers tested receive training in how to match faces and spend half their waking hours matching live faces with passport photographs, so if practice improves performance with faces, their face matching should be better than average. However, in three studies, passport officers did no better with face matching than university students or participants drawn from the general population. In the special issue of CRPI, Megan Papesh reports another study examining this question in which she tested more than 800 individuals from another occupation requiring matching live faces to ID photographs – professional notaries. The results from the new study are consistent with the study of passport officers. Notaries were no better than control participants, and years on the job was not associated with performance. So much for “practice makes perfect”; these studies suggest it doesn’t improve face matching ability at all!
The articles in the special issue tackle questions like these, and the findings provide the field with a deeper understanding of a number of critical issues. Just as importantly, I’m hoping the special issue raises public awareness of individual differences in face recognition so that more organizations that depend on effective face recognition appreciate the impact of these differences and take steps to address them. If such organizations want to do that, psychologists will be happy to lend a hand.