Steblay, N., Dysart, J., Fulero, S. & Lindsay, R.C.L. (2003). Eyewitness accuracy rates in police showup and lineup presentations: a meta-analytic comparison. Law and Human Behavior, 27, 523-540.

Time for another Meta-Analysis kids!

The Basics:

Lineups consist of a group of photos or live individuals, with the key aspect being the collection is multiperson.

Showups consist of a single person, whether photo or live, that is presented to the witness.

Research up to this point has been dominated by lineup research, which may be disproportional to the amount of lineups conducted in the real world. Various reports cited by Steblay and colleagues show that the range of percentage of cases in which showups are used can range from 30% to 77%.

As we remember from past readings, there are two basic types of lineups, simultaneous and sequential. Since sequential lineups use a more absolute judgment process, Steblay and friends propose that a showup may just be a special case of sequential lineup. Unfortunately, a showup does not give an innocent suspect any protection. The showup provides no alternative choices and may have an inherent level of suggestibility, or so would say the Supreme Court.

A rough review of the past literature led the writers to four main points:

1. There has been little research comparing showups to lineups.

2. The research so far has shown inconsistent results.

3. The measure of interest in each study may have affected the interpretation of the results.

4. Opposing predictions may be justified based on the interpretation of the current literature.

This meta-analysis takes the early stance that witnesses in the showup conditions will show a higher rate of false identifications when compared to those in the lineup conditions. To test this idea the research used a set of three basic comparisons: correct identification rates overall, correct identification rates from target present lineups, misidentification rates from the target absent lineups.

Method:

Basic computer search of PsychINFO followed up with some correspondence with authors gave the experimenters their data.

Data for the following were incorporated:

1. correct decision (both identifications and rejections)
2. correct identifications
3. false rejections
4. selection of foil (target–absent only)
5. correct rejections
6. identification of foil (any)
7. Selection of designated innocent

Eight papers with 12 studies total were used, with 3013 participants (range of 59 – 565). Community resident made up 41% of the sample with 44% college students and 15% children.

Results and Discussion:

Overall correct choices

• Showups produced a significantly higher amount of correct choices (69% vs. 51%).

Frequency of choosing

• In target absent conditions, a witness is twice as likely to choose from a lineup as from a showup (54% vs. 27%)
• In a target-present lineup witness chose 71% of the time, while witness in the showup chose 46% of the time.

Decision-Making in Target-Present Conditions

• Correct identification is higher in the target-present showups than the target-present lineups (47% vs. 45%, p=0.08).
• The findings are similar with miss rates in the two conditions. 53% for showups and 55% for lineups.
• False rejections fewer in the lineup condition (34% vs. 58%)
• Considering only choosers, lineups produce more choosers (71%) but a lower hit rate (64%) when compared to showups (46% chooser, 100% hit rate by default).
• Overall we get a showup lineup comparison of 46% vs. 45% correct identification rate.

Decision-Making in Target-Absent Conditions

• Showups yielded higher correct rejection rate (85% vs. 57%).
• Showup yielded a lower error rate than lineups (15% vs. 43%).
• This 15% id pretty dangerous as that is a guaranteed mistake vs. a potential obvious mistake in the selection of a foil as in a lineup.

Thus, corrections are in order. Foil identification was teased out of the data.

• Now they have 15% vs. 10% showup vs. lineup in the dangerous error.
• Overall exclusion of foils choices yields 84% correct rejection and 16% false IDs of the dangerous variety for lineups and 85% correct rejection and 15% false Ids for showups.
• “Innocent Suspect” studies (purposeful inclusion of a description matched foil) yielded lower error rates in showups (23% vs. 45%), but higher dangerous false identifications (23% vs. 10%) than lineups.

Comparison of Showups to Sequential and Simultaneous

• Table One

Additional Discussion

• Lineup foils are important because they give a credibility measure to the witness (24% foil ID in target-present and 31% in target-absent).
• Past research has shown children are more likely than adults to pick a foil from a target-absent lineup, despite warnings. The current data shows children were better able to perform in the target-absent showups than target-absent lineups, similar to adults. Results were parallel for target-present lineups.
• There may be an inherent bias in innocent suspects who match a perpetrator description.

---

University of Arkansas	Department of Psychology	Graduate Program in Experimental Psychology
Lampinen Lab	False Memory Reading Group	False Memory Reading Group Spring 2004