Having to Make Quick Decisions Helps Witnesses Identify the Bad Guy in a Lineup

Eyewitness identification evidence is often persuasive in the courtroom and yet current eyewitness identification tests often fail to pick the culprit. Even worse, these tests sometimes result in wrongfully accusing innocent suspects. Now psychological scientists are proposing a radical alternative to the traditional police lineup that focuses on eyewitnesses’ confidence judgments.

In a new article forthcoming in Psychological Science, a journal of the Association for Psychological Science, Neil Brewer of Flinders University and colleagues report a new type of lineup in which witnesses are presented with lineup photos one at a time and are simply asked to rate how confident they are that the person in each photo is the culprit. Importantly, witnesses are not given time to mull over their assessments; they must respond within a few seconds.

Brewer and his colleagues have tested this technique, called “deadline confidence judgments,” with more than 900 participants across several experiments conducted over the past three years.

In each experiment, volunteers watched short films depicting crimes or a mundane event in which one person was prominent.

Either five minutes later or a full week later, half of the participants were shown a series of individual pictures from a lineup of 12 people and asked to make a confidence decision about each face within three seconds of it appearing on the computer screen. They were asked to choose one of 11 options, ranging from “absolutely confident that this is the culprit” to “absolutely certain this is not the culprit.”

The other half of the participants were shown the same faces but given as long as they liked to answer whether each face was or was not the culprit (yes or no). Sometimes the photos included the culprit and sometimes they did not.

Using an algorithm to infer a decision from participants’ confidence ratings, the researchers found that overall classification accuracy was 20 to 30 percent higher for the new lineup than the conventional one.

Moreover, the researchers were able to identify particular patterns of confidence judgments that showed either a very high or very low likelihood that an individual witness’s judgments were accurate.

The finding that eyewitnesses’ judgments were more accurate under a deadline fits with previous research, which has shown that accurate eyewitness identifications are made significantly faster than inaccurate ones, and that a number of outside factors are removed with a short deadline.

“A weakness of the traditional test lies in the fact that it requires a witness to make a single ‘yes’ or ‘no’ decision about a lineup, with plenty of time to reflect on their decision,” says Brewer. “But the time lapse from the initial viewing to the response often mitigates against witnesses making accurate decisions, as does an array of external factors.”

These external factors include the conditions under which people view lineup photos, constraints on attention, and social cues that bias the witness towards a positive identification. For example, a witness might think that she should know they answer just because she viewed a photo for a long time.

According to Brewer, traditional identification tests often fail because the witness feels pressure to identify a guilty party. This new study suggests that witnesses are more likely to make accurate identifications when they do not have to be so precise.

With the rising number of DNA exonerations and the frequent failure of witnesses to identify the true culprit, Brewer believes that there is a compelling case for a new system of lineups.


APS regularly opens certain online articles for discussion on our website. Effective February 2021, you must be a logged-in APS member to post comments. By posting a comment, you agree to our Community Guidelines and the display of your profile information, including your name and affiliation. Any opinions, findings, conclusions, or recommendations present in article comments are those of the writers and do not necessarily reflect the views of APS or the article’s author. For more information, please see our Community Guidelines.

Please login with your APS account to comment.