§ 26.01 EYEWITNESS IDENTIFICATION: THE PROBLEM AND POTENTIAL SAFEGUARDS

• Jurisdiction: North Carolina

§ 26.01. Eyewitness Identification: The Problem and Potential Safeguards¹

[A] The Problem

[1] Overview of the Problem

Eyewitness misidentification of suspects has long been recognized as one of the most serious problems in the administration of justice. In 1967, in United States v. Wade,² the Supreme Court warned that "[t]he vagaries of eyewitness identification are well known," and "the annals of criminal law . . . rife with instances of mistaken identifications." And, in the Court's most recent pronouncement on eyewitness identification procedures,³ the justices cited a brief of the petitioner, which in turn "cit[ed] studies showing that eyewitness misidentifications are the leading cause of wrongful convictions." The Court reiterated that "[w]e do not doubt . . . the fallibility of eyewitness identifications."

There are ample studies to support the Court's observations. A survey in 1932 reported that among 65 cases in which innocent persons were convicted, 29 of the improper verdicts were attributed to improper eyewitness identification.⁴ Another study of wrongful convictions in the late 1980s estimated that eyewitness misidentifications accounted for 52 percent of the cases surveyed.⁵

Two more recent studies demonstrate that misidentifications remain a serious and common problem. A 2005 survey of all reported exonerations of convicted defendants between 1989 and 2003 (totaling 340) concluded that "[t]he most common cause of wrongful convictions is eyewitness misidentification"; 64 percent of the wrongful convictions involved an eyewitness misidentification, including 90 percent of the rape cases.⁶ And in a 2008 study of the first 200 persons exonerated by post-conviction DNA testing in the United States, all of whom had been convicted of either rape or murder, 79 percent were convicted with incorrect eyewitness testimony.⁷

Despite the severity of the problem, the Supreme Court has expended relatively little time devising constitutional safeguards against false convictions based on misidentification. Ironically, the Court has adopted far more rigorous rules, resulting in the exclusion of far more reliable evidence, in the search-and-seizure and police interrogation areas than it has in the identification field, where attention is needed.

[2] Details of the Problem

Some eyewitness misidentifications result from the intentional use of suggestive techniques by police officers, such as by displaying a suspect in a lineup with persons who do not look like him or fit the witness's description of the criminal.⁸

Most misidentifications, however, are attributable to conditions beyond the control of the police, namely, the "inherent unreliability of human perception and memory and . . . human susceptibility to unintentional, and often quite subtle, suggestive influences."⁹ As two observers have summarized the research data, "people quite often do not see or hear things which are presented clearly to their senses, see or hear things which are not there, do not remember things which have happened to them, and remember things which did not happen."¹⁰

A person can simultaneously perceive only a limited number of stimuli from his environment, even if his attention level is high.¹¹ As a consequence, it is difficult for a witness to observe concurrently the height, weight, age, and other features of a suspect at the time of a crime. Reliability is further reduced because the encounter between the witness or victim and the criminal is often brief, in poorly lit conditions, and in highly stressful circumstances. There is also substantial evidence of unreliability in cross-racial identification cases, i.e., in which a person of one racial group seeks to identify a perpetrator of a crime who is a member of a different racial group.¹²

Another serious problem is that people tend to see what they expect to see. A hunter looking for deer may identify the silhouette of a human being as a deer, even though a non-hunter nearby, observing the same moving figure, will realize it is a human being.¹³ This phenomenon is aggravated by stereotyping and prejudice: "[I]n particularly complex or ambiguous situations, individuals will necessarily structure events to make them understandable, and in cases of ambiguity may be guided more by past experiences, needs, and expectations than by the stimuli themselves."¹⁴ In such circumstances, a person may unconsciously "see" a person who fits the viewer's expectation of a criminal.¹⁵

Beyond the difficulties that arise from the original observation, human memory decays over time. Worse still, "memory . . . is an active, constructive process,"¹⁶ that is, a person tends to fill in memory gaps with new "information." People have a psychological need to reduce uncertainty and to render consistent what is not; therefore, witnesses often unconsciously fill in memory holes with details that are inaccurate.

The police identification process itself can aggravate the situation. By its nature, a lineup "is a multiple-choice recognition test"¹⁷ in which the eyewitness-participant often believes (or is led to believe) that there is no "none of the above" option. The effect of this is that the witness picks the "most correct answer," i.e., the person who most resembles his memory of the culprit.¹⁸ Thus, in one study,¹⁹ after a staged crime, eyewitnesses were asked to select the offender from a lineup in which the "wrongdoer" was absent. When the witnesses were warned that the culprit might not be in the lineup, misidentifications occurred in 33 percent of the cases; without the warning, the error rate was 78 percent.²⁰

Witnesses are also subject to socio-psychological pressures that can render identifications untrustworthy. For example, a subtle comment or action by a police officer (the authority figure), such as a change in voice intonation or even the hint of a smile, may inadvertently suggest the "right" choice in a lineup.²¹

Small-group pressures also influence the process. If witnesses are in each other's presence during the identification process,²² there is a significant risk that once a person is identified as the perpetrator by some witnesses, others will feel an...