Memory: Recognition, Recollection, and Justice

The way in which our memories work is poorly understood.  Some people still believe that our brains are capable of recording every image we view in great detail, but personal experience and scientific evidence renders that hypothesis absurd.  Scientists believe that our brains use our periods of sleep to sort through experiences and decide which events to record more permanently and which to let slip away.  We can clearly affect that process by repetition and assigning a sense of emotional priority to certain things, but there is no guarantee that the brain will respond the way we wish.  Memory records events, but it also alters them, lets them dissipate over time, and even can create records of events that never occurred.  It works subconsciously according to rules evolved over millions of years of natural selection, and we know little of those processes and have even less control over them.

Given this state of affairs, it is somewhat unsettling to consider that our legal system depends critically on memory as a factor in deciding innocence or guilt—and in some cases, life or death.  A recent article in The New Yorker, Recognition, by Paul Kix reminds us that depending on memory is fraught with uncertainty and can produce tragic consequences.

Kix describes the experiences of Michelle Murray who was a college student in Texas in 1985.  At that time she was approached by a black man as she was seated in her car on campus near her dorm.  The man forced his way into the car, drove her to a remote location, and raped her.  Murray came to the police with the information that the man smoked, wore blue jeans and a yellow shirt, and that she had tried her best to remember the man’s face so that she would recognize him in the future. 

There had been similar instances where a black man had approached a white woman in or near her car and raped her.  The police decided to make a suspect of a black man named Timothy Cole who had been seen driving in the area in which the Murray had been parked when the assailant approached her.  Cole stopped and talked to a white woman and drove away.  This aroused suspicion.  The police obtained a picture of Cole and showed it to Murray along with pictures of five other men.  The five men were viewed in standard mug shots, while Cole’s was in a different format.  The details of this process are important.  From Kix:

“Five of the six photos were standard mug shots: men in police custody, holding up placards and gazing away from the camera. Murray’s eyes were immediately drawn to the Polaroid shot of Cole, in front of wood panelling, looking directly at the lens.”

“Murray deliberated for a few seconds. Then she pointed to Cole and said, ‘I think that’s him’.”

“‘Are you positive?’ Nevarez asked. He seemed pleased with her choice.”

Murray decided she was positive.  Cole was found to have a ring and a yellow shirt in his apartment along with an empty pack of cigarettes of the same brand as the rapist smoked in his car.  He was brought before Murray in a lineup.

“Murray entered the dark room and noticed the man she’d identified in the photo spread the day before. She was suddenly certain that he had raped her; she pinpointed him immediately after stepping into the room. As the officers around her celebrated, she signed an affidavit that read, in part, ‘I am positive of my identification of this man and there is no doubt in my mind’.”

It turned out that the yellow shirt and the ring found in Cole’s apartment had nothing to do with Murray’s assault, and Cole was asthmatic and unlikely to be smoking.  It was claimed that the empty cigarette pack belonged to a friend of Cole’s brother.

Prosecutors understand, if they have any professional standards at all, that eyewitness testimony is highly unreliable, yet they continue to prosecute cases and get convictions based upon it.

“In a dissenting opinion in 1981, Supreme Court Justice William Brennan wrote that ‘eyewitness identification evidence is notoriously unreliable.’ Dozens of scientific studies support this claim. Nevertheless, eyewitness testimony continues to be used widely, and many criminal cases hinge on it almost exclusively. Since 1989, two hundred and eighty people have been exonerated of sexual-assault charges in the U.S. Nearly three-quarters of those wrongful convictions relied, in whole or in part, on a mistaken identification by an eyewitness.”

Eyewitness identification when the witness and the accused are of different races is particularly inaccurate.  There is something called the “cross-race effect.”  From Wikipedia:

“A study was made which examined 271 real court cases. The results from this study showed that witnesses correctly identified 65% of the defendants who were of the same race as them. On the other hand, 45% of the defendants were identified who belonged to a different race than the witnesses.”

The prosecutor offered Cole an incredible plea deal.  If he acknowledged guilt, he would be given “five years of probation, with no prison time.”  Remember, this is Texas, ground zero for judicial executions.  Cole refused the bargain as he continued to profess his innocence.

The jury ignored all other evidence and convicted Cole based on the adamant assertion by Murray that Cole was her assailant.  He was sentenced to 25 years in prison. 

In 1999, Cole, while still claiming his innocence, suffered an asthma attack and died in prison.  Meanwhile, Murray had moved on with her life.  She was certain that she had acted properly, and her main concern was that Cole would be released from prison and seek revenge.

In 2008 Murray, now going by her married name of Mallin, received a phone call from George White, one of the investigators in her case.  Someone else had admitted to raping her and his confession had been corroborated by DNA analysis.

“On the phone, White told her that Tim Cole had died nine years earlier. She was relieved, but also confused. ‘Why are you telling me this now?’ she said”

.

“’Tim Cole did not rape you,’ White said. Mallin was stunned into silence. ‘You shouldn’t feel bad about this,’ he went on. ‘He let himself be in that lineup’.”

Mallin went from someone who had been congratulated for asserting her confidence in her memory, to one who was being blamed for making a tragic mistake.  How do well-intentioned people like Mallin make mistakes?

Alison Winter provides an excellent study of the various changes in beliefs about memory and how it works in her book Memory:Fragments of a Modern History.  She tells us that two contending views of memory have battled for supremacy over the years.  One view has it that memory is a faithful recording device and with the right assistance all memories can be recalled with veracity maintained.  The alternate view holds that memory is an uncertain source of information and it is subject to change by conscious or subconscious processes. Significantly, it is also easily influenced by the suggestions of others.  Courtrooms have often been the battlegrounds for these competing views.  Legal decisions and many studies have supported the view that it is risky to base legal decisions solely on the memory of a witness.

Two studies described by Winter are relevant here. 

The first is an experiment performed in 1974 by Robert Buckhout a professor studying memory.  He staged a scene on a television show where viewers were provided with a scene where a man was situated off to the side of a corridor waiting for a woman to come by so he could snatch her purse.  When she arrived, he grabbed her purse and ran straight towards the camera.  Viewers were provided with six images of men and invited to call in and identify the thief or claim it was none of the six.

The result was that few of the viewers made the correct identification.  The results were essentially consistent with the viewers making random guesses.

So, startled viewers subject to an unexpected event have trouble recalling what they actually saw, but what about someone like Murray/Mallin who had time to burn an image into her memory during a traumatic event?  Winter claimed that for a long time people believed that events with high emotional content would be stored in a different mode than others—they would be more accurately recalled.  Memory and photographic recording have always been conflated (usually improperly).  Consequently, these hypothetical types of memory were referred to as “flashbulb memories.”

When the Challenger space shuttle exploded in 1986 it produced a significant emotional impact on viewers of the event.  Researchers moved in within days to survey recollections of the event from a group of people.

“They asked detailed questions about respondents’ memories, then repeated the questions nine months later to see how memories might have changed over time.  They found over time that memories became less detailed, and sometimes inaccurate, and that claims were introduced that had not been there in the original narratives.  Their conclusion was that although people were unusually confident in these memories, they were not more reliable than ordinary memories.  There was no special mechanism at work.”

It is the power of suggestion that is most insidious in altering, elaborating, or falsely confirming a given recollection.  The first legal case provided by Winter involving memory and legal proceedings involved a man who was a suspect in a murder in 1906.  Under police questioning he confessed to the crime and signed a statement to that effect.  Several weeks later he recanted claiming the police had somehow “hypnotized” him while in custody.  The man also had eleven people who provided him an alibi that proved he could not have committed the crime.  Little was known about human memory then and the concept of an innocent man making a false confession was incredible to the jury.  The man was convicted based that confession.  We now know that false confessions are common.  About a quarter of those who are released from prison based on DNA evidence have been convicted based on a false confession.  Some are coerced into confessing through plea bargaining, but others were merely confused at the time of confessing.

Isolating a person and surrounding him with authoritative people who can fabricate a scenario in which the man commits a crime, and then claim to have evidence that the man committed the crime, will lead some people to doubt their own memories.  It works, and police have been taught how to confuse people and take advantage of this process.

Let us go back and consider Murray/Fallin’s experience now with the influence that suggestion can exert in view.  First of all, she was presented images of six suspects with Cole’s image in a different format than the others.  The other five were typical mug shots.  It was a clear suggestion that something was special about Cole and the others might have been pulled randomly from police files.  Her initial response was “I think that’s him.”  She expressed certainty only after the officer expressed pleasure with her selection.  When she identified Cole in lineup the officers “celebrated.”  It might take little more than the suggestion that Murray/Fallin had “done a good job” to convert uncertainty into certainty.

Kix brings in Gary Wells, a psychology professor at Iowa State University to explain how police should seek an identification and thus eliminate a lot of false identifications.

“In 1998, Wells was the principal author of a paper suggesting a few simple reforms. Lineups should be ‘blind,’ a standard borrowed from scientific experiments: the officer administering the lineup should know nothing about the case, so as to avoid unconsciously influencing the proceedings. In photographic lineups, images should be presented sequentially rather than simultaneously, allowing witnesses to compare each image against their memory instead of choosing from among a group. And witnesses should be instructed that a lineup might not include the perpetrator. In 1999, Wells was part of a federal panel that published a report elaborating on these findings. The report was non-binding, but a few jurisdictions—New Jersey, Tucson, Minneapolis—took up its recommendations.”

Kix then provides some specific data related to Texas.

“In 2001, Texas passed a law allowing inmates to request DNA testing of evidence relevant to their convictions. Rick Perry, then the governor of Texas, was cautious about criminal-justice reform, but he signed the bill. Since 2001, DNA testing in Texas has led to more than forty exonerations, a greater number than in any other state. Misleading or inaccurate eyewitness identification was a factor in a vast majority of overturned convictions in those cases.”

The ultimate purpose of Kix’s article was to demonstrate that a bit of good can emerge from a tragic mistake.  Lessons can be learned.  

Mallin was devastated by her mistake.  It was suggested that she make contact with Cole’s family who had long been trying to exonerate him.  She did, and was surprised to discover that they welcomed her as a fellow victim.  She became close with the family and assisted them and others in promoting legal reforms and negotiating a pardon for Tim Cole.  They helped accomplish quite a bit.

“While Tim Cole’s family was trying to clear his name, twenty-nine state legislators, both Republicans and Democrats, had joined forces as the authors of three related bills: House Bill 498, which would establish the Timothy Cole Advisory Panel on the prevention of wrongful convictions; Senate Bill 117, which would require state police to be trained in more advanced lineup practices; and the Tim Cole Act, which would increase the amount of money the state owed to exonerated ex-convicts. Texas’s policy was to pay fifty thousand dollars for each year of wrongful incarceration—already one of the highest rates in the country. The bill would increase the amount to eighty thousand dollars per year spent behind bars, plus funds for college tuition.”

Eventually all bills were passed and Tim Cole was granted a posthumous pardon, the first in Texas history.

“Research carried out in Austin between 2008 and 2011 shows that the use of sequential lineups substantially ‘reduces mistaken identifications with little or no reduction in accurate identifications’.”

This led one expert to conclude:

 “In terms of conducting eyewitness identification, Texas is doing better than any state in the Union.”

Now that Texas has learned how uncertain the legal justice system is and how many innocent people get convicted, perhaps they could consider eliminating the death penalty.  Why kill a person when you know the probabilities suggest he might be innocent?