Recollections of crimes often tainted

In this five-part series, the National Post explores the mysteries of memory, how it works and how it fails. This is the third instalment, on memory and the law.

---

The 1982 police lineup in which Ivan Henry was restrained in a police headlock was so obviously unfair that it would appear amusing were it not for the fact it was part of the evidence used to convict him of a series of rapes in Vancouver. One victim testified that she was "pretty sure" that he was the person she remembered as her attacker.

He spent more than 26 years in prison before he was released on bail this year, following the release of information that pointed to another suspect as the actual perpetrator.

While his tainted lineup was an extreme, false identifications by well-meaning witnesses are not isolated occurrences. More than 75% of convictions in the United States later overturned through DNA testing were a result of faulty eyewitness identification, according to data compiled by The Innocence Project, at the Cardozo School of Law in New York.

Just last week, James Bain was freed after 35 years in prison in Florida, convicted of a rape that he did not commit. The conviction was based on a mistaken identification by the victim and Bain spent from age 19 to 54 in prison, until he was cleared by DNA evidence.

In Canada, one of the best-known instances of eyewitness evidence leading to a wrongful conviction was the case of Thomas Sophonow, who spent four years in prison before he was released and later cleared. The Sophonow Inquiry, conducted by Justice Peter Cory, included a number of recommendations in 2001 about how to conduct police lineups to reduce the chance of eyewitness error.

Gradually, police departments across Canada are implementing those recommendations and starting to embrace more than two decades of research about witnesses and the frailties of memory.

Widely accepted research shows that the memory of witnesses to a crime is never like a video camera and there are many ways the already faulty recollections can be further tainted. Suggestions by police, or simply a desire to be a good citizen, can lead to an identification of someone who looks like the suspect, rather than the actual perpetrator.

The use of hypnosis or recovered memory therapies, which not so long ago were widely used in court, have now largely been discredited, after research found that the suggestive techniques were just as likely to create false recollections as to enhance a person's ability to remember a past event.

Memory is a "rough code" in someone's head of a past event, said John Turtle, a psychology professor at Ryerson University in Toronto, who frequently advises police on eyewitness identification issues. "There is no good research to say memory varies that much. Instead, it is situational. It depends on the circumstances. It is almost a crapshoot."

What psychologists like Prof. Turtle and his colleagues have attempted to do for the past several years is come up with techniques to reduce eyewitness error.

When assessing the accuracy of eyewitness testimony, it is important "to look for core details that do not seem to be changing," he said.

Presenting witnesses with mug shot photos sequentially instead of an array of six or 12, is now more common, to reduce the chance of a witness picking the person who looks most like the suspect they remember.

Police are also now more aware that people are better at picking out suspects of their own race, that people considered attractive or unattractive tend to be more memorable and that both very young children and the elderly are more prone to mistaken identification.

Without proper procedures, the potential damage to the criminal justice system is significant, said Rod Lindsay, a psychology professor at Queen's University in Kingston, and prominent researcher in this area.

There are on average about 8,000 photo lineups conducted by police in Canada each year. If even one per cent resulted in false identifications, that would affect 80 criminal cases annually, he noted.

The confidence of a witness, rather than the accuracy of the identification, was found to have a greater impact on mock juries in a 1988 study conducted by Prof. Lindsay and other researchers. The seminal study also suggested that the level of experience of the prosecutor and defence lawyer in a case did not counter the impact of a confident witness.

This tool in a prosecutor's arsenal may be why Prof. Lindsay said he has encountered more resistance from Crown attorneys than police when advocating for changes in the way eyewitness information is collected and how it is used in court. "I tell them, if you have other evidence, why introduce garbage evidence," he said.

Even eyewitness testimony from police officers is often no more detailed or, ultimately, reliable, than that of civilians.

"You are human first. Your brain does not change," said Prof. Turtle.

In the Boxing Day 2005 shootout in downtown Toronto that claimed the life of Jane Creba, an off-duty officer was only a few metres from the gunfight; neither the officer nor anyone else on the crowded shopping thoroughfare that day was able to positively identify any of the suspects.

Similarly, none of the 17 customers who were packed into the tiny Just Desserts restaurant in Toronto in 1994 when a customer was fatally shot were able to positively identify any of the three men who were tried for the crime.

In both cases, the stress of the

incidents had a huge impact on the observations and memory of the witnesses.

"Stress is very complicated. It narrows attention. People will remember certain things very well," said Prof. Lindsay. Other things have much less clarity; a concept called "weapon focus," refers to the fact that witnesses often give more detailed descriptions of weapons than suspects when caught in the middle of a violent crime.

Despite the frailties of memory and eyewitness observations, Prof. Lindsay said it is still a valuable source of evidence in a lot of cases.

"Memory is fallible, so you can only reduce error, you can't eliminate it," he said.

In attempting to reduce these errors, the Vancouver Police Department is a world away from the days of the tainted lineup that helped convicted Ivan Henry, and is now a leader in eyewitness identification procedures, said Deputy Chief Doug LePard.

Vancouver police have embraced the recommendations of the Sophonow Inquiry. As a result, sequential lineups, a script for the person conducting the lineup so there are no improper suggestions and having an officer who does not know the identity of the suspect, are all standard procedure.

The memory of a witness, without corroborating evidence, will almost never be enough to charge someone with a crime, stressed Deputy Chief LePard.

Since the new practices were implemented in 2005, there seems to be fewer identifications overall by witnesses, according to the information the deputy chief is receiving from his detectives. "That is a price we pay. But we know the price in being wrong."

---------

HYPNOTICALLY REFRESHED MEMORY

One of the best known cases where hypnosis played a key role in a Canadian trial was the original prosecution of Robert Baltovich.

The Toronto man was convicted in 1992 of second-degree murder in the death of his girlfriend, Elizabeth Bain. Four Crown witnesses were hypnotized and the supposedly improved memory of sightings of Mr. Baltovich by three of them were essential to the prosecution's case.

Mr. Baltovich spent nearly nine years in prison and the Ontario Court of Appeal overturned his conviction in 2004 and ordered a new trial.

By the time of his re-trial, early in 2008, however, the legal landscape had changed dramatically when it came to the use of hypnosis as a tool for refreshing the memory of witnesses. The Supreme Court of Canada ruled in February 2007 that post-hypnosis evidence was presumptively inadmissible, because of concerns that it was essentially junk science. The ruling came in ordering a new trial for Stephen Trochym, a Canada Post employee convicted of killing his girlfriend. Questions about the accuracy of hypnosis were raised as far back as the early 1980s. As a result of a 1984 ruling in Alberta, courts tended to admit post-hypnosis evidence as long as police followed certain guidelines aimed at ensuring the hypnotist was independent and did not put improper suggestions to the witness.

In its 2007 ruling, the Supreme Court looked to ongoing research about post-hypnosis evidence and the increasing skepticism about the practice, especially in the United States. "There is a general consensus that most individuals are more suggestible under hypnosis, that any increase in accurate memories during hypnosis is accompanied by an increase in inaccurate memories, that hypnosis may compromise the subject's ability to distinguish memory from imagination, and that subjects frequently report being more certain of the content of post-hypnosis memories, regardless of their accuracy," wrote Justice Marie Deschamps for the majority. "While it is not generally accepted that hypnosis always produces unreliable memories, neither is it clear when hypnosis results in pseudo-memories or how a witness, scientist or trier of fact might distinguish between fabricated and accurate memories," she added. "This technique and its impact on human memory are not understood well enough for post-hypnosis testimony to be sufficiently reliable to be used in a court of law." After several weeks of pre-trial legal arguments in the Baltovich re-trial, the Crown ultimately decided to call no evidence and Mr. Baltovich was acquitted of murder in April 2008.

Shannon Kari,

---------

THIS WEEK

More On Memory

Our series continues with Memory in Music and Literature and the Destruction of Memory.

---------

REPRESSED MEMORY

The use of so-called "recovered memory therapy" to unearth repressed memories of traumatic childhood events gained much favour in the 1980s and 1990s. It led to dozens of cases across North America where people were charged in cases of alleged historical sexual assault, usually of children. In some cases, the allegations included bizarre claims of satanic rituals. Along with the use of hypnosis to prompt someone's memory, this practice is now treated with extreme caution by the courts in Canada. "I think it has pretty much been discredited," said Stephen Lindsay, a psychology professor at the University of Victoria, who has conducted research in this area. The fundamental criticism of the technique was that therapists made too many suggestions that induced patients into believing certain traumatic events had occurred in their past. "We don't remember anything without cues," Prof. Lindsay said.

"I don't buy the notion of repression. We simply forget a lot. I think we forget most of our lives."

While an individual may be more likely to remember a traumatic event in their past, they may also be "motivated to forget bad things" and simply not think about it, he explained. In testing the accuracy of these kinds of memories, he said it is crucial to examine "how did the person come to remember this."

---------

GUIDELINES FOR PHOTO LINEUPS

Sequential lineups Photos of suspects should be presented one at a time, instead of in array of 12 presented as a group. Otherwise, witnesses are more likely to compare all 12 at once to see who looks most like the suspect.

Videotaping lineup identifications This ensures that there is an accurate record of what takes place when a witness is asked by police to see if a suspect can be identified.

Pre-lineup instructions Witnesses should be told the suspect may not be in the lineup. This lowers the chance of a witness feeling obligated to pick someone out of the lineup who most looks like the person they remember.

Use of fillers Only one suspect should be included in a photo lineup. The rest should be "fillers," people known to be innocent, yet who fit the general description of the suspect.