Become a Patron! 


Nancy Franklin and Michael Greenstein

excerpted from: Nancy Franklin and Michael Greenstein, A Brief Guide to Factors That Commonly Influence Identification and Memory of Criminal Events, 85-APR New York State Bar Journal (March/April, 2013)


By a conservative estimate, mistaken IDs lead to thousands of wrongful convictions each year. Even under the best circumstances - good perceptual conditions, no stress, and an ID made soon after exposure - people correctly identify unfamiliar faces only about half the time and make misidentifications about a quarter of the time. Several factors, many of which are characteristic of criminal situations, not only further reduce the number of correct IDs but also increase false IDs. This article provides a (by no means exhaustive) review of some of the factors known to affect ID accuracy. [Factors include: Weapon Focus, Stress, Flashbulb Memory, Confidence,  Post-Event Information/Suggestibility Effects, Social Contagion, Leading Questions, The Identification Procedure, Show-Ups, Filler Selection, Instructions, Feedback,  Identification Time,  Unconscious Transference and Mugshot Exposure, Perceptual Factors, Partial Disguise,  Consequences,  Voice Identification, Characteristics of the Witness, Age,  “Trained Observers”, Considerations Regarding Pre-trial Hearings,  Expert Agreement)

* * *


Perpetrator's Race

The best way to process a face is holistically, noting the relations between its parts. People tend toward this type of processing for faces of their own race but rely on individual features for other-race faces. This reflects a selective perceptual expertise that is already in place in infancy. The outcome is that cross-race identification is consistently worse (with both lower correct IDs and higher false IDs) than own-race identification. Most of the studies have involved black and white witnesses and target faces, but the same selective-perception effect has also been shown with many other combinations. Although it appears that having close relationships among members of the other race can reduce the effect, simple exposure within an integrated community does not help much.


* * * 

Beyond the Ken?

Laypeople have it wrong much of the time. For example, nearly 90% believe that stranger identifications are likely to be “very” or “somewhat” reliable.

Laypeople's intuitions about many of the factors affecting eyewitness memory tend to be similarly off. While it is encouraging when their intuitions point in the right direction, laypeople often substantially underestimate the degree to which an eyewitness's memory of an event is influenced by both outside and infernal factors. Laypeople may not understand that a real impairment exists (e.g., cross-race ID) or may believe that certain witnesses are not affected (e.g., the superiority of trained observers). Occasionally, lay intuitions are diametrically opposed to reality, as when the majority of people believe that the more violent the criminal act, the more accurate will be the memory of the details of that crime. Altogether, this poses a risk that jurors will evaluate the facts of the case ineffectually.

Judges are in a difficult position when they use their intuitions about whether particular topics are beyond the ken of jurors. To correctly make this assessment, they need to see past their own knowledge, a very difficult thing to do. In fact, once people become aware of a fact, they often forget what their own previous beliefs had been.

Conclusions: What Can We Say About Eyewitness Memory and Identification?

Is memory terrible? No! Perception is largely selective and constructed, and memory is selective and reconstructed. These principles allow the cognitive system to be powerful in an imperfect world where recording every detail of every moment of one's life simply isn't an option. Our cognitive system is amazingly sophisticated but ill-suited to the type of task required for courtroom examination. We're simply not equipped to disentangle our interpretations, inferences, and post-event information from a memory, and we're relatively poor judges of our own and each other's abilities. It is easy to see how sincere, honest witnesses can make false identifications and report details that didn't happen, and it is easy to see how jurors may rely heavily on this testimony when making their decisions.


* * *

For More Information About Real-World Cases

Garrett, B.L. (2011).Convicting the innocent: Where criminal prosecutions go wrong. Cambridge, MA: Harvard University Press. Brandon Garrett, a Professor of Law at the University of Virginia, examines the first 250 cases that led to DNA exonerations. Eyewitness identification had contributed to a number of these convictions.