1. For a brief overview of how memory works and a discussion of some of the factors that have been at play in the judicial attitudes toward the use of experts on the fallibility of some eyewitness identifications, read the following brief excerpt from the book Scientific Evidence in Civil and Criminal Cases, 4^th ed. (1995) by Moenssens, Starrs, Henderson & Inbau, specifically in Chapter 19, at pp. 1171-1180. It should be noted that much additional research has been done since the 1995 edition was published. The additional research, some of which is referenced in No. 2, infra, uniformly confirms the major premises of the excerpt, that (1) eyewitness identifications are fragile and can often result of grievous error, and (2) that the common beliefs whereupon "old law" excluding expert evidence on the reliability of eyewitness identifications was based have been shown to be popularly held misconceptions. [Some of the footnotes in the quoted text have been omitted; the others have been renumbered.]

§ 19.15 The Inherent Unreliability of Eyewitnesses—True or False?

The chronicles of legal history are filled with tragic cases of mistaken identity. The tale of Berson and Morales echoes that of hundreds of thousands of others. In the 1970s, Lawrence Berson was accused of several rapes and George Morales was accused of robbery. Both men were picked out of police lineups by victims of the respective crimes and both men were innocent. Berson was cleared when another man, Richard Carbone, was arrested and implicated in the rapes. Carbone was convicted on the rape charges and later confessed to the robbery, clearing Morales.

Regrettably, it is impossible to know how many innocent men and women have been sentenced to prison or death on the basis of faulty eyewitness testimony. However, it has been estimated that more than 4,250 Americans per year are wrongfully convicted due to sincere, yet woefully inaccurate eyewitness identifications. Over the past fifty years, scientific research has revealed that eyewitness testimony is often an incorrect account of what actually took place. Scientists now know that the human mind does not act as a video camera meticulously recording and replaying everything within its viewfinder. Rather, human memory is a complex process which is vulnerable at every stage. This knowledge, in turn, has led attorneys to invite psychologists who have done studies on these issues to testify to their conclusions when the critical evidence was that of occurrence witnesses. Most of the cases wherein this testimony has been offered have been criminal prosecutions involving eyewitness identifications of a defendant, though the principles of perception, memory, and recollection are applicable equally to event witnesses in civil litigation.

§ 19.16 Perception, Memory, and Recollection

The acquisition of information into memory involves a three-step process. At each stage of the process, errors are possible. During acquisition, the first step in the memory process, an event is perceived and information "bits" are initially stored in memory. In the second stage, information is held or retained in memory. In the final stage, memory is searched and pertinent information is retrieved and communicated. In the acquisition stage, information is "encoded" into a person's memory system. However, every detail of an experience is not encoded; the human mind can only process a fraction of the rapidly incoming physical stimuli. Both consciously and unconsciously, the observer determines which details are actually encoded according to where his or her attention is focused.

The physical aspects of an event are obviously compromised by the selective nature of the acquisition stage of memory. However, matters are further complicated by the fact that acquisition also involves a social component. Thus, a witness' ability to perceive accurately is affected by both event factors—those inherent to the event itself—and witness factors—those inherent to the witness.

1. INFLUENCE OF THE EVENT FACTORS

Event factors include duration of the event, complexity of the event and violence of the event witnessed. Common sense and science show that increased exposure time improves the accuracy of witness perception. However, witnesses generally overestimate time and have great difficulty gauging the duration of an event. For example, Professor Buckhout, a cognitive psychologist, staged an assault during one of his classes. The assault actually lasted 34 seconds, but the average estimate of time by 141 witnesses to the event was overestimated by a factor of almost two and one half to one (2.5:1).

Complexity of the event witnessed has a direct effect on the encoding of information. Psychological research has shown that complex events are more difficult to recognize and encode into memory than more basic events. For example, a street fight among multiple parties is less likely to be recollected with accuracy than a fight between two parties.

Violence also affects a witness' acquisition of information. In 1978, an experiment was conducted in which subjects viewed violent and non-violent tapes of an event. The subjects who viewed the violent version of the event had more difficulty perceiving and recalling the event than those subjects who had seen the non-violent version.

2. INFLUENCE OF PERSONAL FACTORS

In addition to event factors, perception is also affected by witness factors. Witness factors include stress, weapon focus, and expectation. Lay persons believe that stressful events accentuate memory, that what is seen or heard during periods of high stress is more accurately recorded and recalled. However, empirical data indicates that stress, anxiety and fear disrupt the normal perception process and distort subsequent memory. Thus, a witness' statement that he was "so frightened that his (the criminal's) face is etched in my memory forever" is a psychological oxymoron.

Psychologists term the relationship between stress and memory as the Yerkes–Dodson Law. The Yerkes–Dodson Law depicts the relationship between stress and memory as an inverted "U" curve. At low arousal, memory is low. Memory improves with increased stress levels to a point (optimal performance). After the point of optimal performance, memory decreases with increased stress levels. Thus, it is not uncommon for a witness to recall the details of crime less accurately than they recall routine events.

Psychological research has also shown that people under stress tend to concentrate on those aspects of an event or experience which they feel are most important. Weapon-focus is an example of this psychological phenomenon. An individual faced with a gun held to their head is likely to concentrate on the gun to the complete exclusion of the attacker's face or physical characteristics.

As early as 1909, it was noted that expectation has a profound effect upon perception: "observation is peculiarly influenced by expectation, ... we tend to see and hear what we expect to see and hear." Modern science can now account for the relationship between expectation and perception. As noted earlier, the human mind can only process a fragment of the physical stimuli digested by the senses. The mind compensates by integrating the stimuli with concepts based upon a fund of general knowledge acquired over time or, put more simply, with expectation. In this manner, individuals subconsciously reconstruct events from what they assume must have occurred.

Expectation includes personal biases, cultural biases, temporary biases, latent prejudices, and stereotypes each of which have the ability to influence acquisition. For example, a group of subjects were shown a photograph of several people standing in a subway train, including a white man holding a razor and apparently arguing with a black man. When asked to describe what they had seen, over half of the participants reported that the black man had been wielding the razor.

3. INFLUENCE OF STORAGE FACTORS

Memory is also susceptible to extraneous influences during storage, the second stage of the memory process. Memory of events decays over time. Although an individual may accurately perceive an event, its representation in the observer's memory system will not remain intact for very long. Considerable memory loss is likely to occur between an event and an eyewitness' actual identification of a suspect in a criminal case. The human mind actively fills the gaps created by long-term memory loss; this gap-filling process introduces inaccuracies into memory for the sake of a "complete" picture. Obviously, this process has great implications for the criminal defendant.

Lay persons tend to believe that there is a positive correlation between witness confidence and memory accuracy—that the more conviction a witness has the more likely his or her testimony is to be an accurate depiction of the event. However, psychological research indicates that there is in fact a negative correlation between a witness' confidence in his or her memory and the accuracy of that memory. Thus, the more confidence a witness has in his or her memory, the more likely the witness' memory of the event is inaccurate. Again, this counter-intuitive psychological principle has far reaching ramifications for litigants.

An individual's memory system may also be augmented or altered during the storage stage by intervening occurrences. For example, a witness may read or hear about an incident he or she observed. The mind tends to incorporate post-event information such as that conveyed by the article or discussion with that previously encoded from the incident itself.

In 1976, Dr. Elizabeth Loftus conducted an experiment to demonstrate how post-event information in the form of questioning can affect memory. In this experiment, subjects viewed a videotaped auto accident and were questioned about what they observed. The question about how fast the cars were going when they smashed into each other elicited significantly higher estimates of speed than those questions posed with words such as "collided," "bumped," "hit" and "contacted" instead of "smashed."

A week later the subjects were again questioned. They were asked "did you see any broken glass?" The videotaped accident in fact did not involve broken glass, but those subjects who had been questioned earlier with the verb "smashed" were more likely to assert that they had seen broken glass in the video tape. A memory was formed when the subjects viewed the tape, but it was later augmented and altered by the suggestive questions of the interviewers. The implications of this study on police questioning techniques and general investigative procedures is obvious.

4. INFLUENCE OF RECALL FACTORS

The final stage of memory, retrieval, is also vulnerable to outside influences. For example, social factors may influence the witness' recollection of events, especially in the criminal identification setting. Eyewitnesses, like other people, do not want to appear foolish. By arranging a line-up, police officials suggest to an eyewitness that they have apprehended the culprit. Consequently, the eyewitness may select an individual despite his or her lingering uncertainty, to avoid "letting the criminal go free" and to avoid looking foolish.

All of these factors have been stressed by psychologists who advocate admission of expert testimony on the unreliability of event witnesses as reasons for their advocacy. However, opponents have been no less vocal, and thus the legal literature abounds with writings on the issue.

Modern psychological science has revealed much about human memory and the fallibility of eyewitness testimony in the 200 years since T. Reid, an influential scholar of his day, wrote:

... and if a skeptical counsel should plead against the testimony of the witness, that they had no other evidence for what they declared than the testimony of their eyes and ears, and that we ought not to put so much faith in our senses as to deprive men of life or fortune upon their testimony, surely no upright judge would admit a plea of this kind. I believe no counsel, however skeptical, ever dared to offer such an argument; and, if it were offered, it would rejected with disdain[.]

Unfortunately, in this area of endeavor, the American legal system has not embraced the teachings of modern psychology wholeheartedly.

§ 19.17 What Can Experts Say?

In a jurisdiction where the expert is permitted to testify on the issue, the witness can explain the mechanism of perception, processing, memory, and recollection, and describe what empirical research has been conducted on the subject and what the results are. The expert will ordinarily not be permitted to testify on the accuracy of a particular witness. In a child sexual abuse case, in a jurisdiction that accepts opinion evidence in this area of the law, the expert typically can testify on the problems child victims of abuse have remembering and describing the incidents of which they complain, but cannot state whether a particular child was abused on the basis of psychological evidence, or that the child was truthful when complaining.

§ 19.18 Admissibility of Eyewitness Expert Testimony

Trial courts and appellate courts in America have taken decidedly different positions on the issue of expert testimony addressing eyewitness reliability. The majority of jurisdictions continues to exclude eyewitness expert testimony. However, a growing number of courts have upheld exercise of the discretionary power of trial judges who exercised discretion in favor of admitting expert opinion testimony on eyewitness reliability.

* * *

Despite these cases, in the majority of jurisdictions that have faced the issue, expert opinion evidence on the unreliability of event witnesses has been excluded. Most of the decisions to prohibit the testimony used as their rationale that the credibility of any witness is an issue for the jury to determine, and that to permit such testimony impermissibly invades the province of the jury. The reasons for exclusion are perhaps best reflected by People v. Enis, wherein the Illinois Supreme Court said:

"We caution against the overuse of expert testimony. Such testimony, in this case concerning the unreliability of eyewitness testimony, could well lead to the use of expert testimony concerning the reliability of other types of testimony and, eventually, to the use of experts to testify as to the unreliability of expert testimony. So-called experts can usually be obtained to support most any position. The determination of a lawsuit should not depend upon which side can present the most or the most convincing expert witnesses. We are concerned with the reliability of eyewitness expert testimony, whether and to what degree it can aid the jury, and if it is necessary in light of defendant's ability to cross-examine eyewitnesses. An expert's opinion concerning the unreliability of eyewitness testimony is based on statistical averages. The eyewitness in a particular case may well not fit within the spectrum of these averages. It would be [inappropriate] for a jury to conclude, based on expert testimony, that all eyewitness testimony is unreliable." [Citations omitted.]

§ 19.19 Path of the Future

Since the advent of the Daubert v. Merrell Dow Pharmaceuticals, which did away with the "general acceptance" rule for the admission of novel scientific or expert testimony and replaced it with a standard of evidentiary reliability, which must be determined by the trial judge prior to admission, courts, whether they admitted or denied admission to evidence on the factors affecting the reliability of event witnesses, may have to revisit the issue and explore whether the "reasoning or methodology underlying the testimony is scientifically valid and ... whether the reasoning or methodology properly can be applied to the facts in issue." That was in fact the reason for remanding the case to the trial court in U.S. v. Amador–Galvan, and the reason why the Supreme Court remanded U.S. v. Rincon for further consideration.

Though the American judicial system has been slow to accept the lessons of modern psychology, there does appear to be a growing group of judges willing to accept the logic of the premise of the unreliability of event witnesses and permit expert testimony on that topic. Whether these recent cases signal a trend which will grow, only time will tell. We can only hope that the legal community will, in this case, be as receptive to opinion testimony that has a sound basis in experimental science, as it has been in other novel fields. If so, perhaps one, two, or more innocent people who otherwise would have been convicted of crimes they did not commit will be set free; after all, it is better to acquit 10 guilty persons than to convict one innocent person.

---

Dr. Gary L. Wells, psychologist and a prominent researcher on eyewitness identification and the retention of data in one’s memory, who is a professor in the department of psychology at Iowa State University, and Dr. John Turtle, of the Department of Psychology and the School of Justice Studies at Ryerson Polytechnic University, recently cooperated in a study commissioned in 1998 by Attorney General Janet Reno, to examine what could be done to improve the quality of eyewitness identification and minimize the occurrence of good-faith but erroneous identifications.

A fairly concise overview of this work (approx. 9 pages) can be found at:

It contains recommendations for law enforcement on how to conduct lineups and show-ups in a manner that will avoid most misidentifcations.

Additional readings recommended, but not required

The United States Supreme Court first pointed to the many weaknesses of lineups conducted by the police in the case of United States v. Wade, 388 U.S. 218 (1967), as a result of which the Court concluded that a defendant who was going to be subjected to an in-person confrontation with an identification witness was entitled to the assistance of counsel for the purpose of observing the fairness, or lack of it, of the procedure.

In 1972, in Kirby v. Illinois, 406 U.S. 682 (1972), the Court’s majority held that the Wade right-to-counsel applied only to those confrontations which occur after the formal judicial adjudicatory process against a person has already commenced. Since most lineups occur fairly early in the process, soon after a person has been arrested, the Kirby case deprived the overwhelming majority of suspects of this Wade-created right of counsel.

In Manson v. Brathwaite, 432 U.S. 98 (1977) the Supreme Court held that "reliability" of an eyewitness identification is the linchpin in determining the admissibility of identification testimony." Factors to be considered in weighing the reliability were identified in Neil v. Biggers, 409 U.S. 188 (1972) and were said to include: the opportunity of the witness to view the criminal at the time of the crime; the degree of attention that the witness devoted to the criminal; the degree to which the description of the criminal as initially reported by the witness to the police matched the eventual in-person identification at the lineup; the level of certainty that the witness exhibited in making an identification; and the length of time that had elapsed between the crime and the identification of the suspect by the witness.

The Missouri Supreme Court has addressed the admissibility of expert testimony regarding eyewitness identification in State v. Lawhorn.  Are you persuaded by the Court's analysis?