Eyewitness Identification: General Acceptance in the Scientific Community

This research paper focuses on the degree to which experts and others are persuaded that each of a number of factors influences the accuracy of eyewitness identifications. Supreme Court cases, among them United States v. Amaral (1973) and Daubert v. Merrell Dow Pharmaceuticals (1993), have opened avenues of research addressing how the influence of various factors on the judgments of eyewitnesses is perceived by different parties in the legal system. Reflecting their familiarity with the literature, experts substantially agree on the extent to which many variables influence identifications. Research indicates that jurors do not agree with the experts on many of these influencing factors. The use of legal processes that will help jurors make better decisions in cases that involve eyewitness identifications, such as having experts testify in these cases, is thus justifiable. Those who serve as law enforcement personnel show unexpected patterns of agreement with experts, though this tendency may change as a result of eyewitness reform at the state level.

The Rationale

The issue of whether or not to allow scientific findings into the courtroom continues to evolve in the United States. The Supreme Court established the admissibility of eyewitness research in United States v. Amaral (1973). The later Daubert v. Merrell Dow Pharmaceuticals (1993) ruling established criteria that had to be demonstrated for scientific testimony to be entered into a trial. One of these criteria was that the basis for the testimony should be generally accepted by the scientific community. The Daubert decision renewed interest in what eyewitness factors are in fact generally accepted by the scientific community. The first survey focusing on the acceptance of eyewitness factors was published in 1989. Research since then has greatly expanded psychologists’ understanding of how members of the scientific, legal, and lay communities accept the findings reported in the eyewitness literature and how this acceptance has changed over time.

It is now common for members of the legal psychology community to distinguish between what are known as system variables and estimator variables. System variables are those that are under the control of the legal system and that can potentially bias an eyewitness during the course of a criminal identification procedure. For example, bias could enter into an identification procedure through the techniques used to construct the criminal lineup or by the use of leading identification instructions given to an eyewitness. In contrast to system variables, estimator variables are those that encompass eyewitness and crime scene characteristics that are not under the direct control of the legal system. Examples include the length of time afforded to the eyewitness to view the crime or the presence of multiple perpetrators at the crime scene. The provision of expert testimony in a trial in which variables such as these are relevant may serve to highlight potential biases in the identification procedure that otherwise may have escaped consideration by the judge or jurors.

The Opinions of Experts

Survey research demonstrates that many phenomena experts overwhelmingly reported as being sufficiently reliable to introduce under oath in 1989 continue to be viewed as reliable influences on the accuracy of eyewitness identifications more than a decade later. There appears to be considerable consensus among experts as to the reliability of the research evidence regarding the wording of questions, the construction of lineups, and the role of witnesses’ attitudes and expectations in influencing their identifications, and on the relationship between witnesses’ confidence in their identification and their identification accuracy. Furthermore, experts agree on the existence of other variables that reliably influence eyewitness identifications, such as the rate at which memories decay, the impact of exposure time on memory and subsequent identifications, and the unconscious transference of the memory of a familiar face from one situation to another. Appreciable increases were observed between 1989 and 2001 in the percentage of experts who agreed that human attention is likely to be focused on a weapon rather than on a perpetrator’s face (a weapon focus effect) and the impact of hypnotic suggestibility. Both changes in consensus were attributed to the respective increases in interest and scholarship on both topics in the years following the publication of the first expert survey.

Later research would investigate the general acceptance of eyewitness factors not addressed in the original 1989 survey of experts. Attesting to the expanding corpus of literature in the eyewitness field, a substantial majority of experts agreed on the malleability of eyewitness confidence, the suggestibility of the child eyewitness, and the tendency of eyewitnesses to choose suspects from a lineup previously encountered in mug shot arrays. Other factors agreed on by a majority of experts included the impact of alcohol on the eyewitness, the difficulty in making identifications of perpetrators not of the same race as the eyewitness, and the reduction in false identifications due to the use of sequential rather than simultaneous lineups. Other phenomena that were supported by at least two thirds of the experts included the inferior accuracy of the child eyewitness when compared with the adult eyewitness, the potential for misleading memories recovered from childhood, and how the use of similar foils (here foil refers to an innocent person in a police lineup) in a lineup increases eyewitness accuracy.

The Opinions of Judges

Although individual jurors are ultimately responsible for interpreting the testimony of an expert witness and applying their insight to the facts of the case at hand, judges alone determine whether the expert witness meets the Daubert criteria for inclusion in the trial. Judges, like jurors, may rely on common sense when interpreting eyewitness evidence in the absence of formal psychological training. Eyewitness identification errors have been cited in many cases of wrongful conviction, although a survey reported that only 43% of judges believed that such errors have been made in half of the reported cases of wrongful conviction. However, not all evidence is discouraging. Survey data on judges’ knowledge and beliefs about eyewitness factors revealed that while judges may agree correctly with statements on eyewitness issues, these same individuals report that the average juror would not be likely to respond correctly. A modest correlation (r = .30) was reported between a judge’s knowledge of eyewitness factors and the willingness of the judge to admit expert testimony.

Agreement between judges and experts was observed on less than half (40%) of 30 eyewitness factors, which included (but were not limited to) the role of attitudes and expectations, the cross-race effect, and the impact of exposure time. Judges were in agreement with experts on less than half (37%) of the listed system variables, including the malleability of an eyewitness’s confidence, the biasing effects of showup lineups, and what constitutes a fair lineup. Judges were not in agreement with experts regarding the phenomenon of hypnotic suggestion. When data collected from a 2004 survey of judges were compared with the results of experts in a previous survey, judges and experts agreed on 7 of 8 items.

The Opinions of Jurors

If a case is tried by a jury, the jurors serve as the ultimate arbiter of fact in the courtroom, and they must decide not only on whether case-relevant eyewitness factors should be taken into consideration during deliberation but also on what weight should be given when considering a verdict. The testimony of experts may serve to allay juror concerns about eyewitness phenomena. Nearly three quarters of respondents in one survey replied that their primary aim as an expert witness was to educate the jury. Thankfully, few researchers reported that they would be willing to testify in court regarding an eyewitness factor on which the published literature was “inconclusive.” In contrast, approximately three quarters of those experts who regarded the evidence as “generally reliable” and a plurality of those who saw it as “very reliable” were willing to testify about these factors. Ninety-five percent of these surveyed experts believed that expert testimony on eyewitness issues had a positive impact on juries.

There was correspondence between experts and jurors on only 4 of the 30 survey statements (13%). As expected, significant differences in the rates of agreement emerged between experts and jurors on the statements focusing on factors classified as system vari-ables, such that jurors were less in agreement about the eyewitness factor than the experts. The four eyewitness factors that experts and jurors did agree on were statements regarding the effects that violence, alcohol, and stress have on an eyewitness and the fact that trained observers are not more accurate witnesses than untrained people. The largest discrepancies observed between experts and jurors were found for statements regarding lineup instructions and hypnotic suggestibility, with jurors expressing significantly less agreement about those eyewitness factors than experts. Juror accuracy (50.7%) differed significantly from the level of accuracy seen among judges and law enforcement personnel when accuracy was defined as agreement with those statements to which at least 75% of experts agree.

Other Evidence

Understanding the general acceptance of eyewitness factors among law enforcement personnel is critical in that members of this population draw their knowledge on the subject both from empirical literature and their experiences in the field. Law enforcement personnel were in agreement with the expert community on only 12 of the 30 statements (40%), among them the role of attitudes and expectancies, the suggestibility of the child eyewitness, and the cross-race effect. Notably, they perceived the influence of only two of the eight (25%) system variables in the same manner as the experts. Experts and law enforcement personnel did not differ in their judgment of the biasing effect of showups and the importance of members of a lineup resembling the description of the suspect. Law enforcement personnel, however, had significantly lower agreement rates than experts with respect to all other system variables (e.g., the malleability of eyewitness confidence, the impact of question wording). Of interest is the fact that the most significant differences between the agreement rates for law enforcement personnel and experts were observed for statements concerning the presentation format of the lineup and the instructions administered during the lineup. When agreeing with statements endorsed by 75% of experts, judges and law enforcement personnel were equally accurate (65.9% and 63.8%, respectively).

General acceptance can be indexed not only in terms of the degree of correspondence among opinions across various participants in trial proceedings but also in terms of the decisions made by policymakers with respect to the implementation of applications derived from empirical psycholegal research. For example, a panel of legal scholars, law enforcement practitioners, and psycholegal experts made recommendations as to procedures that should be adopted by the police when they obtain eyewitness evidence. One example of this is the recommendation that witnesses and those law enforcement officers who conduct lineups both be unaware of who is a suspect and who is not (double-blind procedures) when lineups are conducted. Some states (e.g., New Jersey, North Carolina, and Wisconsin) have implemented such recommendations at the time this research paper was written, and additional states are considering this and other reforms as well.


  1. Benton, T. R., Ross, D. F., Bradshaw, E., Thomas, W. N., & Bradshaw, G. S. (2005). Eyewitness memory is still not common sense: Comparing jurors, judges, and law enforcement to eyewitness experts. Applied Cognitive Psychology, 20, 115-129.
  2. Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579 (1993).
  3. Kassin, S. M., Ellsworth, C., & Smith, V. L. (1989). On the general acceptance of eyewitness testimony research: A survey of the experts. American Psychologist, 44, 1089-1098.
  4. Kassin, S. M., Tubb, A., Hosch, H. M., & Memon, A. (2001). On the general acceptance of eyewitness testimony research: A new survey of the experts. American Psychologist, 56, 405-H6.
  5. Seltzer, R., Lopes, G. M., & Vanuti, M. (1990). Juror ability to recognize the limitations of eyewitness identifications. Forensic Reports, 3, 121-137.
  6. United States v. Amaral, 488 F.2d 1148 (9th Cir. 1973).
  7. Wise, R. A., & Safer, M. A. (2004). What US judges know and believe about eyewitness testimony. Applied Cognitive Psychology, 18, 427-443.

See also: