Lineups

by

Margaret Bull Kovera, Jacqueline Katzman

• LAST REVIEWED: 24 February 2021
• LAST MODIFIED: 24 February 2021
• DOI: 10.1093/obo/9780199828340-0275

Introduction

Lineups are conducted in the course of police investigations when a crime has been witnessed by one or more people. A lineup typically consists of a person whom the police believe committed the crime (i.e., the suspect) and some number of people who are known to be innocent of the crime (i.e., fillers). When the police have developed a suspect, they show witnesses a lineup to test whether they will claim that the suspect is the person who committed the crime (i.e., the perpetrator). If so, the witness is said to have made a positive identification of the suspect. What is not clear, at least in real-world investigations, is whether that identification is correct, because sometimes suspects are guilty and sometimes they are innocent. Since the late 1970s, psychologists have conducted experiments to find lineup procedures that decrease the likelihood that witnesses will mistakenly identify innocent suspects. These experiments are typically conducted in laboratory settings in which researchers expose participants to a simulated crime, often on videotape. After the participant-witnesses have viewed the crime, they are asked to attempt an identification from a lineup. In the laboratory, researchers can vary whether the perpetrator appears in that lineup. When the perpetrator is present in the lineup (i.e., a target-present lineup), the witness can identify the suspect (a correct identification), identify a filler, or say that the perpetrator is not there (an incorrect rejection of the lineup). When the perpetrator is not present (i.e., a target-absent lineup), the witness can make a mistaken identification of the suspect, identify a filler, or correctly reject the lineup. Using this method, researchers have identified lineup procedures that decrease mistaken identifications, which are the leading cause of wrongful convictions among those who have been exonerated by DNA tests conducted after trial. This article contains sections describing comprehensive General Overviews of research on lineups, research demonstrating that Live Lineups Are Equivalent to Photo Lineups, and Policy Recommendations and Best Practice Guidelines. The remaining sections describe many of these policy recommendations, including how Lineups Are Superior to Showups, having an Evidence-Based Suspicion for placing a suspect in a lineup, unbiased Lineup Composition, Double-Blind Administration, proper Lineup Instructions, collecting witnesses’ Confidence Statements in the accuracy of their identification immediately after the initial identification, Video Recording Identification Procedures, and avoiding Repeated Lineups. An additional section addresses special issues that need to be considered when Conducting Lineups with Children.

General Overviews

Wells 1978 was the first work to highlight the importance of studying variations in lineup procedures to find ways to improve the accuracy of eyewitness identifications. This article influenced several decades of researchers to investigate the types of lineup procedures that decrease the rate of mistaken identifications. Brewer and Wells 2011 provides a very accessible overview of research on lineups and eyewitness identification. The authors argue that variables that increase mistaken identifications can do so through both general impairment of witnesses’ abilities (i.e., when variables affect memory strength or increase mistakes because they induce the witness to make a positive identification, even when the suspect is not present in the lineup) and suspect bias (i.e., when procedures encourage the witness specifically to pick the suspect). Some general overviews represent theoretical advances in understanding how witnesses make identification decisions from lineups. Clark 2003 uses computer simulations to model the eyewitness decision-making process and tests one of the first theories of eyewitness identification decisions. Wixted and Mickes 2014 offers a different theoretical explanation for eyewitness decision making from lineups. It argues that witnesses are more likely to make accurate identifications from lineups conducted with procedures that help them recognize which facial features of lineup members are diagnostic of guilt (i.e., features of the perpetrator that are shared with only one lineup member) and those that are nondiagnostic (i.e., features that many lineup members share with the perpetrator). Lee and Penrod 2019 introduces yet another model of lineup-identification decisions that, unlike Wixted and Mickes 2014, takes into account the unique information that is provided when a witness identifies a filler rather than identifying the suspect or rejects the lineup. Other general overviews discuss how to evaluate the research on identifications from lineups to make policy recommendations about what types of lineup procedures should be used to maximize correct and minimize mistaken identifications. Clark 2012 argues that many procedural reforms not only reduced mistaken identifications but also reduced correct identifications and that any preference for one outcome over the other was a value judgement. Building on this analysis, both Lampinen, et al. 2019 and Yang, et al. 2019 describe different frameworks for evaluating the costs and benefits of different proposed reforms.

• Brewer, N., and G. L. Wells. 2011. Eyewitness identification. Current Directions in Psychological Science 20.1: 24–27.

  DOI: 10.1177/0963721410389169

  Provides an overview of different types of variables that influence witness identification accuracy. Variables that influence the strength of the witness’s memory for the perpetrator, like a short exposure to the perpetrator’s face, cause a general impairment in witness memory. Other variables, like when a suspect stands out from among the fillers, create suspect bias, causing witnesses to positively identify the suspect more frequently than they otherwise would.

• Clark, S. E. 2003. A memory and decision model for eyewitness identification. Applied Cognitive Psychology 17.6: 629–654.

  DOI: 10.1002/acp.891

  Presents a model of eyewitness identification decisions that presumes that witness memory is less than perfect and that witnesses attempt to match their imperfect memories to each lineup member. Whether witnesses make an identification of a lineup member is a function of both how well a lineup member matches their memory of the perpetrator and whether that lineup member is the best matching lineup member.

• Clark, S. E. 2012. Costs and benefits of eyewitness identification reform: Psychological science and public policy. Perspectives on Psychological Science 7.3: 238–259.

  DOI: 10.1177/1745691612439584

  Reviews the empirical evidence of the effects of lineup reforms on identification decisions. Concludes that many reforms that decrease mistaken identifications also result in fewer correct identifications and therefore questions the conclusion that proposed reforms to lineup procedures have no costs.

• Lampinen, J. M., A. M. Smith, and G. L. Wells. 2019. Four utilities in eyewitness identification practice: Dissociations between receiver operating characteristic (ROC) analysis and expected utility analysis. Law and Human Behavior 43.1: 26–44.

  DOI: 10.1037/lhb0000309

  Reports a method of evaluating the benefits of different lineup reforms that takes into account the relative tradeoffs between failing to identify guilty suspects and mistakenly identifying innocent suspects and the probability that suspects are guilty.

• Lee, J., and S. D. Penrod. 2019. New signal detection theory–based framework for eyewitness performance in lineups. Law and Human Behavior 43.5: 436–454.

  DOI: 10.1037/lhb0000343

  Proposes a model of identification decisions from lineups based on signal detection theory, which posits that discriminability (i.e., the ability to detect whether a face has been seen before) and response bias (i.e., the witness’s tendency to report that a face has been seen before) will determine whether witnesses will discriminate between innocent and guilty suspects. Demonstrates that the discriminability of fillers, ignored by other models, underlies the discriminability of suspects.

• Wells, G. L. 1978. Applied eyewitness testimony research: System variables and estimator variables. Journal of Personality and Social Psychology 36.12: 1546–1557.

  DOI: 10.1037/0022-3514.36.12.1546

  Discusses the difference between estimator variables, which are variables that cannot be controlled by the justice system (e.g., lighting, distance, and other witnessing conditions) and system variables, which are under the control of the justice system (e.g., whether the lineup administrator knows which lineup member is the suspect). This article highlights the importance of studying lineup procedures to improve the justice that suspects receive.

• Wixted, J. T., and L. Mickes. 2014. A signal-detection-based diagnostic-feature-detection model of eyewitness identification. Psychological Review 121.2: 262–276.

  DOI: 10.1037/a0035940

  Presents a new theoretical model of eyewitness decisions when making identifications from lineup: the Diagnostic-Feature-Detection Model. Argues witnesses are more likely to make accurate identifications when lineups procedures help them recognize which facial features of lineup members are diagnostic of guilt.

• Yang, Y., L. Smalarz, S. A. Moody, J. J. Cabell, and C. J. Copp. 2019. An expected cost model of eyewitness identification. Law and Human Behavior 43.3: 205–219.

  DOI: 10.1037/lhb0000331

  Presents a method for evaluating the efficacy of different lineup procedures for maximizing the accuracy of witness decisions. Concludes that the superiority of different procedures depends upon the costs one assigns to filler identifications and lineup rejections in combination with the probability that suspects appearing in lineups are guilty.