A recipe for mistaken convictions: why federal rule of evidence 403 should be used to exclude unreliable eyewitness-identification evidence.

• Author: Shell, Eva G.

"[E]yewitness testimony is likely to be believed by jurors, especially when it is offered with a high level of confidence, even though the accuracy of an eyewitness and the confidence of that witness may not be related to one another at all. All the evidence points rather strikingly to the conclusion that there is almost nothing more convincing than a live human being who takes the stand, points a finger at the defendant, and says 'That's the one!"' (1)

1. INTRODUCTION

   Eyewitness misidentification is the leading cause of mistaken convictions in the United States. (2) Presently, more than 300 people have been exonerated through postconviction DNA evidence. (3) The average sentence served in these cases, at the time of exoneration, was 13.6 years. (4) Out of these 300 exonerations, eyewitness misidentification played a part in nearly 75% of the convictions. (5) Scientific studies of eyewitness identification began to develop in the late 1970s, frequently the subject of psychological rather than legal journals. (6) The Supreme Court has not directly revisited the topic of eyewitness identification since 1977, despite significant progress in related scientific research over the past thirty years. (7)

   Defense attorneys have attempted to use expert testimony to combat the prejudicial effects of unreliable eyewitness-identification evidence, with mixed results regarding admissibility. (8) Some courts provide jury instructions on the potential unreliability of eyewitness identifications, and many jurisdictions are beginning to update their identification procedures to better align with current scientific understanding of memory. (9) While these remedies are important, they do not provide a sufficient safeguard against the potential for wrongful convictions caused by eyewitness misidentifications. (10)

   In order to prevent wrongful convictions, judges should use their gatekeeping powers under Federal Rule of Evidence (FRE) 403 to exclude unreliable eyewitness-identification evidence from use at trial. (11) FRE 403 is currently used to exclude classifications of evidence where juries are likely to be irrationally persuaded. (12) Additionally, trial judges already have significant gatekeeping responsibilities under the Federal Rules of Evidence with regard to other areas of evidence, including character evidence and expert testimony. (13)

   This Note will discuss some of the most thoroughly studied causes of eyewitness misidentification and suggest that trial judges use their discretion under FRE 403 to exclude unreliable eyewitness-identification evidence. (14) The first part of this Note will provide an overview of scientific research from the last thirty years relating to memory and eyewitness identification. (15) The second part of this Note will discuss current procedures for combating misidentifications, analogous situations where FRE 403 is used for exclusion, and gatekeeping capacities of trial judges. (16) Lastly, the third part of this Note will present arguments for using FRE 403 to exclude unreliable eyewitness-identification evidence. (17)

2. History

   Recent developments in scientific research have shown that eyewitness-identification accuracy depends on many situational factors, both within and outside of law enforcement agencies' control. (18) In recognition of the potential for unreliability, some jurisdictions have updated their eyewitness-identification procedures in order to produce more reliable identifications. (19) Additionally, some courts allow eyewitness-expert testimony and jury instructions on the topic of eyewitness-identification unreliability. (20) The Supreme Court's framework for excluding unreliable eyewitness-identification evidence has not been updated since 1977. (21)

   1. Thirty Years of Research Shows that Memory Is Malleable

      While scientists have studied the accuracy of eyewitness identifications for more than 100 years, the development of extensive, programmatic research in this area did not begin until the late 1970s. (22) Scientists have since identified different variables that are significant to the evaluation of eyewitness-identification evidence. (23) These variables are generally divided into two categories: system variables and estimator variables. (24) System variables are those in the eyewitness-identification process that are within the control of the criminal-justice system, while estimator variables are those that the criminal-justice system cannot control. (25) The New Jersey Supreme Court recently appointed a Special Master to extensively probe the past thirty years of scientific research on memory and eyewitness identification, and incorporated the research into a decision that changed the state's treatment of eyewitness-identification evidence. (26) This Note utilizes the wealth of research collected by the Special Master, specifically with regard to the identified variables for determining eyewitness-identification reliability. (27)

      1. System Variables

         Eight system variables are relevant to a discussion of the evaluation of eyewitness-identification evidence. (28) The first three system variables are: blind versus nonblind lineups, preidentification witness instructions, and lineup construction. (29) Studies show that blind and double-blind lineups are far more reliable than nonblind lineups. (30) A blind lineup occurs when the lineup administrator knows the identity of the suspect but does not know where the suspect is located within the lineup or photo array. (31) Double-blind lineups, where the lineup or photo array administrator does not know the identity of the suspect, are shown to produce the most accurate identifications. (32) Eyewitness identifications are also more accurate when the eyewitness is told prior to seeing the lineup or photo array that he or she need not pick a suspect. (33) The lineup-construction variable includes the percentage of nonsuspect persons in the lineup, the number of suspects in the lineup, and the relative physical similarity between the suspects and nonsuspects. (34) Studies show that eyewitness identifications are more likely to be accurate when the lineup contains only one suspect, when the suspect looks similar to other members of the lineup, and when the lineup consists of at least five nonsuspects, called "fillers." (35)

         The fourth, fifth, and sixth system variables are: postidentification feedback, the witness's level of confidence in the identification, and whether the eyewitness viewed the suspect multiple times during the investigatory process. (36) Lineup administrators may intentionally or unintentionally signal to the eyewitness that he or she identified the suspect, thereby providing postidentification feedback. (37) This behavior can cause the eyewitness to have increased confidence in his or her decision, regardless of whether the eyewitness actually identified the perpetrator. (38) Contrary to common belief, an eyewitness's confidence in his or her identification does not necessarily correlate with accuracy. (39) Identifications are, however, more likely to be accurate if the eyewitness has not previously viewed investigatory photographs of any members of the lineup. (40)

         The final system variables are: whether a facial composite or a "showup" was used in the identification process. (41) While it is unclear whether creating a composite has an effect on the eyewitness's memory, research suggests that composites are not likely to bear a strong resemblance to the actual perpetrator. (42) A showup occurs when an eyewitness views a single suspect, presented by the police for purposes of identification, usually at the scene of the crime and shortly after the crime was committed. (43) Timing is an important factor for the showup procedure because showups are more likely to produce unreliable results when conducted more than two hours after the commission of a crime. (44)

      2. Estimator Variables

         Nine estimator variables are relevant to this discussion. (45) The first four estimator variables are: stress, weapon focus, duration of the eyewitness's encounter, and lighting and distance. (46) Research shows that a high level of stress during an encounter negatively impacts an eyewitness's ability to make an accurate identification at a later time. (47) Similarly, when a weapon was present during an encounter, the eyewitness is likely to be approximately ten percent less accurate when later attempting to make an identification. (48) Not surprisingly, research shows that eyewitnesses generally make more accurate identifications after a longer encounter with the perpetrator. (49) Eyewitnesses also make more accurate identifications when they were closer to the perpetrator during the encounter, and when lighting conditions during the encounter were favorable. (50)

         The remaining estimator variables relevant to this analysis are: witness characteristics, perpetrator characteristics, memory decay, race bias, and actions of private actors. (51) Research shows that factors such as eyewitness intoxication have significant negative effects on identification accuracy. (52) Disguises and differences in the perpetrator's facial hair between the encounter with the eyewitness and the lineup also produce less accurate identifications. (53) Additionally, research confirms the common notion that memories irreversibly decay with time. (54) Eyewitnesses are significantly less likely to make accurate identifications when they are of a different race than the perpetrator. (55) In addition to law-enforcement officers, nonstate actors can also interfere with eyewitnesses' memories and increase the risk of inaccurate identifications. (56)

   2. Current Attempts to Remedy the Misidentification Problem

      Many members of the legal community have recognized that eyewitness misidentifications pose a significant problem. (57) Some jurisdictions have taken steps to adjust police procedures to align with current scientific data on how to prevent...