Eyewitness testimony: a study of perception and memory

Eyewitness Testimony, etc.

In a Psychology Today article in 2001, Elizabeth Loftus, Ph.D. And William Calvin, Ph.D. discussed what was then known about memory, and what was yet to be discovered. Loftus has written 18 books, one of which is titled Eyewitness Testimony. Loftus and Calvin noted that concern with memory and its imperfections dates to long before the period of modern jurisprudence, although -- in the aftermath of overturned murder convictions that had been based on eyewitness testimony -- memory has again become a hot issue. "Medieval and modern philosophical accounts of human cognition stressed the role of imagination. The 18th-century philosopher Immanuel Kant talked about imagination as the faculty for putting together various mental representations such as sense percepts, images, and concepts" (Loftus, 2002, March, p. 41+). As long as 1885, the first experiments concerning memory were published, by German psychologist Hermann Ebbinghaus. Since then, much has been learned.

Among the surprising bits of knowledge, probably unfamiliar to the layman, is that the human brain is cluttered, unlike the "pigeonhole" data docks in a computer's RAM. Calvin noted that when a person loses a thought or memory, it often pops up an hour later; it had been lurking just below consciousness all the time. "The serious difference between computer and human memory is that we don't pop out a pristine copy of the original event, the way a computer does. Instead, we reconstruct things as best we can from all the clutter. We guess. Often that isn't good enough, especially for a fair judicial process" (Memory's Future, 2001, p. 55).

The judicial process, about twenty to twenty-five years ago, was dependent on "eyewitness" testimony in a number of high-profile cases, many of them involving supposed child abuse by caretakers of every sort. Loftus notes that then, eyewitness fallibility and the inaccuracy of "recovered" memories was not known (Memory's Future, 2001). Indeed, the effect of this attitude toward memory was dramatic as " 'Repressed memory therapists' went out and prospected for early childhood memories of trauma. 'Are you sure you weren't abused?'" (Memory's Future, 2001, p. 55).

Loftus and Calvin assert that those pro-active seekers for 'truth' "inadvertently created false memories of the worse sort in some of their clients" (Memory's Future, 2001, p. 55) and that the aggressive conduct of these professionals led patients to false memories of child molestation, and many of those led to arrest, conviction and incarceration of innocent people.

Loftus and Calvin reach the conclusion that in these days of inexpensive electronics, it would be far better for justice to virtually dispense with eyewitness testimony in favor of video. "A publicly supported surveillance trend began in Britain nearly 20 years ago. They installed 60 remote-controlled video cameras at various 'trouble spots.' Where they put them, crime dropped by half" (Memory's Future, 2001, p. 55).

In recent weeks, those cameras have also been instrumental in identifying those responsible for the London bombings.

In a more academic paper less than a year later, Loftus suggested, regarding eyewitness testimony, that there might be a better concept to use when dealing with it. She wrote:

In light of what I have learned about human memory, I propose a more realistic alternative: 'Do you swear to tell the truth, the whole truth, or whatever it is you think you remember?' (Loftus, 2002, March, p. 41+).

Indeed, other research seems to support that unique viewpoint.

Photographic accuracy

Loftus and Calvin both asserted that memory is at least as much invention as recording, making it, therefore, notoriously unreliable as an 'expert' device in any criminal case.

To prove it, psychologist Craig Barclay conducted a test with graduate students at the University of Michigan. "Barclay's major thesis is that people do not reproduce the past, they reconstruct it in accord with 'self theories' of how they are likely to act" (Rubin, 1985, September, p. 38+). The experiment involved the students writing down three events each day for four months and recalling them later; however, their own accounts were interspersed with accounts by others. What Barclay concluded was that "We convey in precise and honest terms a plausible and consistent record of our own intentions and actions, but this record need not be, and in fact cannot be, complete and accurate. Our autobiographical memories reflect not only our past but also our personalities and beliefs about ourselves" (Rubin, 1985, September, p. 38+). He notes that this is useful for psychotherapists, and a lot less useful for jurisprudence. Moreover, Rubin concluded from Barclay's work, "People are more likely to accept a false description as the resemblance between the false and original descriptions increases and as time passes" (Rubin, 1985, September, p. 38+).

Other cogent findings vis-a-vis eyewitness testimony is that an individual's state when in the process of remembering can alter the memory itself; likewise the memory can change the individual; clearly, this could be a spiraling process that brings the 'rememberer' farther and farther away from actual events, an unfortunate possibility in terms of eyewitness testimony.

Another sobering possibility is that of undiagnosed disorders that can impair memory even beyond the normal range of alteration. Two researchers in England, Alan Baddeley in Cambridge and Barbara Wilson at Oxford, studied this and found that some amnesiacs who were relatively normal in functioning day-to-day also, if they had sustained frontal lobe damage, "falsely remembered events" suggesting that they could not properly use memory retrieval mechanisms available to normal individuals (Rubin, 1985, September, p. 38+). So stunning is this possibility that Baddeley and Wilson reported the case of one person who reported clear memories concerning the death of a brother who had never existed (Rubin, 1985, September).

False convictions

The year 2002 saw "the release of the 100th person nationwide to be freed from prison after genetic testing" (Loftus, 2002, March, p. 41+). She also noted that:

Larry Mayes of Indiana, now 52 years old, spent 21 years in prison for a rape of a gas station cashier. The victim had failed to identify him in two separate lineups and picked him out only after she was hypnotized by police. Mayes' story is a common one; analyses of these DNA exoneration cases reveal that faulty eyewitness memory is the major cause of wrongful convictions (Loftus, 2002, March, p. 41+).

The 1990s were, in fact, rife with cases that captured public attention because of their subject matter, often "impossible memory claims" including intergenerational satanic ritual abuse or memories of having been molested at the age of six months (Loftus, 2002, March, p. 41+). From the standpoint of psychology, the cases were suspect, possibly even then. From the standpoint of jurisprudence, such cases lead to the possibility that prosecutors simply had a lust for convictions at any cost. "The cases proceeded under the belief that when people are repeatedly brutalized, their memories can be completely repressed into the unconscious and later reliably recovered with hypnosis, dream interpretation, sodium amytal, or other therapeutic 'memory work.' In fact, no credible scientific support has been found for such claims" (Loftus, 2002, March, p. 41+).

Another high-profile eyewitness standard from the 1990s was the "repressed" memory. One famous case involved an Arizona pediatrician, John Danforth, whose former patient, Kim Logerquist, remembered after an interlude lasting two decades that the doctor had repeatedly molested her for a period of two years, from the time Logerquist was eight until she was 10. Logerquist wanted $3 to $5 million in damages, claiming that the flashbacks that gave rise to her claim were triggered by watching a children's aspirin ad. During the period leading up to her accusation, Logerquist has spent 57 sessions of analysis during which she was "urged to try to remember abuse that might explain her problems such as self-mutilation, depression, suicide attempts, obesity and bulimia" (Loftus, 2002, March, p. 41+).

It took ten years for Danforth to clear his name, although it took the jury only 40 minutes to find against Logerquist, who had also spent time considering which other men she had known might also have abused her.

Loftus calls this a "landmark repressed-memory case" and notes that it might have helped take the wind out of the sails of such eyewitness testimony (Loftus, 2002, March, p. 41+).

The problem for those who would profit from such eyewitness testimony, however, seemed to be, after the Danforth exoneration, that the accused was alive and would therefore defend himself. Clever accusers then began claiming abuse took case in a "McDonald's bathroom or on a Royal Caribbean cruise or in the high school art room" (Loftus, 2002, March, p. 41+) -- and that since that event, the accused had since died. The accuser then sued the estate with a fair degree of success. "Even a well-funded corporation has a difficult time defending against supposedly repressed memories about events that purportedly happened 30, 40, or 50 years ago" (Loftus, 2002, March, p. 41+).

Worse, yet, the issue of planted suggestion is inherent in all repressed memory 'eyewitness' accounts. "Psychological studies have shown that it is virtually impossible to tell the difference between a real memory and one that is a product of imagination or some other process" (Loftus, 2002, March, p. 41+). Loftus notes that science has found "post-event information" is integrated into what most people have actually experienced because, "when people experience some actual event -- say a crime or an accident -- they often later acquire new information about the event. This new information can contaminate the memory" (Loftus, 2002, March, p. 41+).

In addition, many false memories are created, deliberately or by accident, in response to leading questioning by therapists or aggressive lawyers. "Subtle cues can be inadvertently conveyed and social reinforcements provided by interrogators operating with a biased set of expectations. But here, too, therapists, interrogators, lawyers, or worried parents may be innocent of any conscious intent to produce false testimony" (Callahan, 1993, p. 6+).

Callahan noted that once a false memory has been established, it becomes, for all intents and purposes, a true account to the one remembering it. "It is important to realize that what psychologists have called 'confident confabulations' are not the same thing as malicious lies which are consciously devised to deceive or harm others. Of course people do lie and knowingly bear false witness, but persons can also be sincerely mistaken in their memories" (Callahan, 1993, p. 6+).

Confabulation may also be a factor when hypnosis is used in a forensic context, in the belief that it confers enhanced recall, according to Webert (2003).

Vested interests

Sanchirico dealt with the appearance of eyewitnesses in certain crimes, noting that "The prevalence of witnesses with evident interests is no coincidence. To be useful in illuminating the underlying event or condition, a witness must have been both present and paying attention" (2004, p. 291+). He also noted that it followed, both theoretically and experimentally, that those who were present and paying attention probably had a personally meaningful reason to do so, an explanation of the odd occasion that an eyewitness is also found to have been in some way complicit in a crime.

Legal ramifications

Loftus noted that about 200 people a day are added to the roster of criminal defendants after being picked out of a lineup or even a photo spread. At the same time, wrongfully convicted individuals are being exonerated at a rapid pace because of DNA evidence, which "has given the world a real appreciation of the problem of faulty eyewitness memory, which is the major cause of wrongful conviction" (Loftus, 2002, March, p. 41+). In fact, as long ago as 1996, the U.S. Department of Justice reported that 80% of the innocent people convicted who were included in the department's study had been convicted "because f faulty eyewitness memory" (Loftus, 2002, March, p. 41+).

This is not a peculiarly American problem. A major case in Canada involved a man named Thomas Sophonow who had been wrongfully convicted of murdering a young waitress. After Sophonow had spent nearly four years in prison, an officially inquiry was held and corrections Commissioner Peter Cory, in discussing the wrongful conviction finding that resulted, noted that Sophonow was "psychologically scarred for life. He will always suffer from the core symptoms of post-traumatic stress disorder. As well, he will always suffer from paranoia, depression, and the obsessive desire to clear his name. His reputation as a murderer has affected him in every aspect of his life, from work to family relations" (Loftus, 2002, March, p. 41+). The case brought the issue to the attention of Canadian law enforcement, and even led to specific procedural changes in lineups and also generally guidance "such as encouraging judges to emphasize to juries the frailties of memory, to recount the tragedies of wrongful convictions, and to readily admit expert testimony on the subject of memory" (Loftus, 2002, March, p. 41+).

The United States also found itself rocked by a spectacular eyewitness-based wrongful conviction. In March 2000, Illinois instituted a moratorium on execution, the impetus for which had been the release of 13 men from death row during the previous ten years.

One had been sentenced to death on the "dubious testimony of a single eyewitness" and another because of two eyewitnesses who later recanted "and another man subsequently confessed and is now in prison" (Loftus, 2002, March, p. 41+). In addition, an Illinois commission made 85 recommendations, including "training in the science of memory for police, prosecutors, and defense lawyers and the development of jury instructions to educate the jurors about factors that can affect eyewitness memory" (Loftus, 2002, March, p. 41+).

Wells and Olson note, however, that sometimes eyewitness testimony is the only evidence available concerning some crimes, and they add it might never have come into question until DNA evidence became available and literally overturned hundreds of convictions based on eyewitness accounts. At that point, researchers attempted not only to define the parameters of memory, but also to define whose memories might be more reliable than those of others might.

Researchers asked whether there were characteristics tahta could indicate greater witness reliability, or lesser. They found no clear evidence that males or females are markedly different in their accuracy, at least as far as choosing criminals from lineups. One analysis indicated females might be "slightly more likely to make accurate identifications but also slightly more likely to make mistaken identifications than are males (due to females being more likely to attempt an identification), thereby yielding an overall equivalent diagnosticity for males and females" (Wells and Olson, 2003, p. 277+).

What did demonstrate a greater difference was the age of the eyewitness, with very young children and the elderly performing much less accurately than younger adults. In fact, research has found that"

The eyewitness identification errors of young children and the elderly are highly patterned: When the lineup contains the actual culprit, young children and the elderly perform nearly as well as young adults in identifying the culprit, but when the lineup does not contain the culprit the young children and the elderly commit mistaken identifications at a higher rate than do young adults....(Wells and Olson, 2003, p. 277+).

Gordon and Follmer reported similar findings. Accuracy of younger children in response to specific questions in their study, was less than that of older children, and "only 7-year-olds responded correctly at levels significantly above chance" (1994, p. 283-294).

The possible effects or race was also examined, and showed a better ability for individuals to accurately recognize members of their own race; this finding has held true for more than 25 years of continuing research into it (Wells and Olson, 2003, p. 277+).

Personal characteristics do apparently have an effect on eyewitness accuracy. Self-monitors, individuals who adapt their behavior to cues regarding what is appropriate are more likely to be inaccurate.

Individuals with high chronic anxiety (a general attitude of apprehension) were more likely to be accurate than those with low anxiety (Wells and Olson, 2003, p. 277+).

Characteristics of the event were also investigated. It was found that distinctive faces are more likely to be accurately recognized than non-distinctive ones. Those that are very attractive or very unattractive are easier to recognize. Simple disguises, as simple as a hat, also reduce accuracy. Changes in appearance due to the aging of the suspect also affect accuracy. The length of time a culprit's face was viewed during an event also positively affected the likelihood of an accurate eyewitness identification (Wells and Olson, 2003, p. 277+). The presence of a weapon being visibly used during a crime reduces the likelihood that a witness will accurately recall the criminal's face (Wells and Olson, 2003, p. 277+).

The certainty of an eyewitness that his or her information is reliable may be slightly correlated with accuracy.

Eyewitnesses who make an identification from a lineup relatively quickly are more likely to be accurate than those who take a long time. "In an impressive set of results, Dunning & Perretta found that those who made their decision in less than 10-12 seconds were nearly 90% accurate in their identifications from a lineup whereas those taking longer were approximately 50% correct" (Wells and Olson, 2003, p. 277+).

Wells and Olson compare conducting a lineup to conducting scientific research. "All the types of things that can go wrong with an experiment to cause misleading results can also go wrong with a lineup. For instance, the instructions might bias the witness, the hypothesis might be prematurely leaked, the design might be flawed, the behavior might be misinterpreted, confirmation biases might be operating, and so on" (Wells and Olson, 2003, p. 277+).