Reforming Eyewitness Identification: Cautionary Lineup Instructions; Weighing the Advantages and Disadvantages of Show-Ups versus Lineups
| Author | Nancy K Mehrkens Steblay |
| Position | Professor of Psychology at Augsburg College in Minneapolis |
| Pages | 341-354 |
Page 341
-
Nancy Steblay is Professor of Psychology at Augsburg College in Minneapolis, Minnesota. She can be reached by email at steblay@augsburg.edu, or at the Department of Psychology, Augsburg College, 2211 Riverside Avenue, Minneapolis, MN 55454.
This paper was presented at the "Reforming Eyewitness Identification: Convicting the Guilty, Protecting the Innocent" symposium held at the Benjamin N. Cardozo School of Law, New York City, Sept. 12-13, 2004.
Advances in scientific psychology have generated substantial knowledge about the psychological processes that move a well-intentioned witness to a false identification, or more desirably, bring the witness to an accurate decision.1 Many factors prior to the lineup task contribute to memory accuracy, including: witness characteristics (e.g., visual acuity, attention, stress), offender attributes (disguise, unique features), event conditions (exposure time, viewing opportunity), and qualities of the time period subsequent to the crime (e.g., retention interval).2When the eyewitness finally approaches a lineup, he or she possesses a valuable asset-a memory of the event. However slim or rich that memory may be, the lineup is an investigator's chance to secure productive information about the suspect and perpetrator.
A compelling body of research reminds us that the eyewitness lineup decision is one based not only on memory, but also on witness motivation and expectation.3Indeed, one of the most intriguing lessonsPage 342 from psychological science, as applied to police practice, is the enormous power over eyewitness behavior that is exerted by the context of the lineup identification task.4Seemingly innocuous gestures or phrasing by the investigator or subtle changes in procedure can dramatically affect a witness's thinking, thus altering the course of the lineup process, the witness's retrospective recall of the memory event, and potentially an investigation and prosecution.5
Along with monitoring eyewitness accuracy, researchers have paid close attention to eyewitnesses' willingness to choose from the lineup (choice of any lineup member, whether suspect or filler).6A lineup selection can be understood in terms of a decision criterion, defined as the lowest strength or vividness of recollection the witness requires before being willing to identify a person as the perpetrator.7Thus, the selection is function of both the witness's true memory of the culprit and external influences. We do not know where on a continuum of cautiousness versus carelessness the decision criterion rests for any specific eyewitness, but laboratory research suggests that this criterion can be "pushed around" by the context and structure of the lineup.
Most recently, this is evidenced by comparisons of sequential to simultaneous lineup formats, where significant reductions in choosing rates, as well as false identifications, are produced with sequential lineup presentation.8Lineup context can also increase witness choosing rates. At the same time that witnesses may feel internal pressures-to trust in their own memory, to present themselves positively, to cooperate with investigators-the demand characteristics of an authoritative and persuasive police environment may reinforce an inclination to make a lineup choice.9The witness well may assume that investigators have put together a lineup because the culprit has been apprehended.Page 343
An effective lineup strategy must provide a careful test of memory alone-, circumventing witness motivation and expectancies that violate this central objective. Wells'10theoretical framework of eyewitness memory has provided direction for research investigations of lineup structure and delivery. Wells reasoned that a traditional lineup scenario allows the witness to engage in relative judgment (i.e., comparing lineup members for the most likely candidate in the lineup). Even a perpetrator-absent lineup can meet the criterion of a best match, potentially placing an innocent suspect at risk. Procedures that force a tighter reliance on absolute judgment (recognition based on memory itself) should produce more accurate lineup decisions.11Absolute judgment is expected to inhibit guessing, resulting in lower witness choosing rates.12
A guiding principle of the eyewitness research literature is that factors affecting identification accuracy fall into two categories. Estimator variables involve aspects of the witness's perception and encoding of a crime event, such as duration of the crime, presence of a weapon, or the race of the perpetrator. Estimator variables are not under the control of the justice system.13Other aspects of the eyewitness experience are amenable to justice system control. These system variables14provide avenues for improvement of eyewitness performance.15Lineup procedures-the context for memory retrieval provided by both structural and process properties of lineups-are among these system variables.16
Some of the earliest laboratory efforts to inhibit relative judgment produced significant effects from slight changes in lineup protocol. This early research demonstrated the subtle but powerful influence of a simple lineup instruction provided to eyewitnesses: "The perpetrator may or may not be in this lineup."17Page 344
Well-established psychological principles explain that social influence originates in an individual's reliance on others to help define or clarify a situation. For an eyewitness, the traditional lineup task is prescribed through its easily recognized prototype in popular culture, and this familiar script is facilitated by instructions from an authority that provide informational social influence (the perpetrator is in the lineup) as well as normative social influence (the acceptable response is to make a choice).18Researchers believe that the basic lineup instruction, "point to the man you saw," provides a combination of scripting, information, and advice that strongly encourages the witness to choose from the lineup.19If the eyewitness assumes the perpetrator is in the lineup, then he or she is likely to believe that the task now is to find the person who best resembles his or her memory of the perpetrator.20
An eyewitness who chooses from a lineup has seemingly met two cognitive criteria: (1) a face in the lineup is perceived as familiar, and (2) that face is judged as having appeared at the crime scene.21Normative social influence may serve to ease the acceptance of a face with familiar features; informational social influence provides the contextual criterion through the perceived claim that the crime was committed by one of the lineup members. With the traditional instruction, there is no explicit option of declining to choose. The witness may be induced to choose even with a relatively poor recollection of the culprit.22
One means to improve this situation is to communicate to the witness an expectation that allows the possibility that the lineup may not include the perpetrator. Can this small change in wording make a difference? To test the impact of lineup instructions, it is necessary to compare two lineup test conditions in which the only difference is the content of the lineup instruction. All participants must witness the same experimenter-controlled event and subsequently be tested for memory of the perpetrator. As the participant-witnesses are introducedPage 345 to the lineup task, half are provided a cautionary (unbiased) instruction: "The person you saw may or may not be in this lineup." The remaining witnesses receive the traditional biased instruction: 'Which of these is the man you saw?" Lineup choices for the two groups can then be compared for accuracy.
This comparison testing has been accomplished by multiple teams of researchers, and a meta-analytic review is available that summarizes the results of the investigations.23Meta-analysis is a statistical tool used to combine the results of multiple studies, each individual study having tested the same hypothesis. The prominent benefit of meta-analysis is to allow detection of significant patterns in data that run across studies-reliable, stubborn effects that endure across investigators, labs, and variations in method. Steblay, et al., reviewed the results of twenty-two experimental tests (i.e., comparisons of the effects of biased versus unbiased instructions under controlled laboratory conditions). The research included 2,588 participant-witnesses. The combined results demonstrated a significant difference in overall witness lineup accuracy (the combination of percentage of correct identifications in target-present lineups plus percentage of correct rejection of target-absent lineups), with the cautionary instruction producing fifty-six percent correct decisions and the biased instruction generating forty-four percent correct decisions. Thus, a simple change in instruction produced an average twelve percent increase in lineup decision accuracy.24
It is informative to separate witness decisions by lineup type, based on whether the offender (target) is present in or absent from the lineup. Table 1 indicates that cautionary instructions produced an average of twenty-five percent fewer errors for target-absent arrays, with no average loss in correct identifications in target-present lineups. The implication is immense: in 1000 target-absent lineups conducted over some period of time, we would expect 250 fewer identification errors.25A rather simple change offers significantly increased protection against identification error in a target-absent lineup.Page 346
Table 1. Lineup Decision Accuracy26
| Target-Present Lineup | ||
| Correct identification | 54% | 53% |
| Error | 46% | 47% |
| Target-Absent Lineup | ||
| Correct refection | 60% | 35% |
| Error | 40% | 65% |
The...
To continue reading
Request your trial