News
COLUMN – Science HEAT

by Brent Snook

An investigator called me recently to ask whether he should use a “simultaneous” or “sequential” lineup for a file he was working on. We agreed that simultaneous lineup would involve asking an eyewitness to determine if they recognize the culprit from a set of photographs (or people) presented all at once – often referred to as a photo array.

Guidelines would accompany this approach to avoid making the lineup suggestive (e.g., tell the eyewitness that the culprit may or may not be present in the lineup, use similar looking people/backgrounds in pictures).

We also agreed that someone conducting a sequential lineup would present one photograph (or person) at a time to the eyewitness and that some administrative guidelines would need to be followed (e.g., no side-by-side comparisons, decision must be made prior to moving on to the next photo and not allowing the person to look through the pictures again).

It was my understanding that sequential lineups were “superior” but I certainly did not want to provide outdated or incorrect advice. I also wanted him to make a decision based on his viewing of the best available evidence. My search of the scientific literature revealed a summary of the research findings on the topic, which I relayed to the officer.

The summary was actually a meta-analysis published in <Psychology, Public Policy and Law> by Nancy Steblay, Jennifer Dysart and Gary Wells. The study was a statistical summary of the tests that compared the effectiveness of the two types of lineups.

As you can imagine, researchers around the world vary how they conduct their research (e.g., use live people versus photographs, vary the age of their participants, use different statistical techniques). The great thing about a meta-analysis is that it cuts through all of the differences in research approaches and potential biases and combines all of the results from the studies so that the central pattern(s) or message(s) is revealed.

Steblay and her colleagues (2011) found 49 studies (varying in quality) containing 72 comparisons (using just over 13,000 participants) of the two types of lineups. The authors converted all of the results from the different comparisons into an effect size – a measure of the strength of the difference (in accuracy) between the lineups. The authors then averaged the effect sizes to get an overall measure of the difference between sequential and simultaneous lineups.

Their analyses showed that:

(1) when the actual culprit is present in a lineup, more correct identifications occur with simultaneous lineups (14 per cent difference in identifications);

(2) when the culprit is not in the lineup, more mistaken identifications occur with simultaneous lineups (22 per cent difference); and

(3) when an innocent person matching the description of a culprit is included in a lineup, more false identifications are made with simultaneous lineups (14 per cent difference).

Most importantly, however, a calculation of the ratio of correct to incorrect identifications for each type of lineup revealed a sequential-superiority effect – a witness’ decision made from a sequential lineup is more probative of guilt. This finding was enhanced when the same analysis was performed using only the highest-quality studies.

Based on the information gained by averaging the 72 sequential-simultaneous comparisons, the inquiring officer was able to make an informed decision (and support it if asked to defend it) as to which type of lineup would be best for his/her investigation.

Both status quo and the fact that a simultaneous lineup would be more likely to get a positive identification (a tempting option) convinced the officer to follow the best available data on the issue. He ultimately agreed with the authors of the meta-analysis that the sequential lineup is a higher standard and more rigorous test and that he could trust the identification results more if a culprit is identified in the lineup.

BIO

Brent Snook, B.A., M.Sc., Ph.D., is an Associate Professor in the Psychology branch of Memorial University in St. Johns, Newfoundland. Contact him at bsnook@mun.ca or 709 864-3101 for more information.