Eyewitnesses Are Not Reliable. So Why Are We Still Relying On Them?

Few things impress a jury more than an eyewitness. Someone who saw the crime take place, and gets up on the stand to testify about it. Eyewitness testimony is compelling in its simplicity; anyone can witness a crime and tell the jury what happened. Scientific evidence is often snarled and confusing, not cleanly laid out like it is on CSI, even when attorneys question carefully to elicit information that’s as clear and useful as possible for members of the jury. Even with scant physical evidence, an eyewitness can tip the scale.

Eyewitnesses, however, are extremely unreliable. As it turns out, people are actually not so great at remembering events, and often overestimate the accuracy of their recollections. This is not new information (link via abby jean):

Harvard Professor Hugo Munsterberg made the point in his book “On the Witness Stand” a century ago. Forty years ago, the U.S. Supreme Court opined: “The vagaries of eyewitness identification are well-known; the annals of criminal law are rife with instances of mistaken identification.”

The problems with eyewitness testimony start at the scene, and carry on from there. Law enforcement officers often taint eyewitness evidence in their eagerness to collect information, and those biases carry through to the trial, where convictions on scant evidence send the wrong people to prison. Extensive research into this subject has been conducted by the psychology community, with little traction. Attorneys continue to call eyewitnesses, and jurors continue to rely upon them. Jurors also maintain a number of false beliefs, propagated by society:

…many respondents believed that a cross-racial identification (identifying a stranger of a different race) would be just as reliable as or even more reliable than a same-race identification…more than half agreed that ‘Human memory works like a video camera, accurately recording the events we see and hear so that we can review and inspect them later.’ (Ed. note—I added the link for people interested in more information about memory and eyewitness testimony.)

Slow reforms in the justice system are changing the way people use eyewitness testimony, but they come too late for many prisoners, and far too late for people like Troy Davis, who was executed in September after being sent to death row in a case with no physical evidence but ample eyewitness testimony. Seven of the nine witnesses later recanted, but that wasn’t compelling enough to commute the sentence, apparently.

Even though psychologists have been discussing the problems with eyewitness testimony for decades, these changes have been slow to come about. In no small part, this is undoubtedly because juries view eyewitnesses as compelling, which makes them appealing for law enforcement officers who want to resolve cases quickly and efficiently. The challenges to such testimony also involve scientific discussion and argument, which requires critical thinking on the part of people like judges handling appeals that question the use of eyewitnesses in a case, or the treatment of those witnesses by law enforcement and the attorneys who questioned them. It’s not that judges are not capable of critical thinking, but that testimony from psychologists is based in a different area of knowledge than that most judges specialise in.

The idea that human memory is infallible lies at the core of beliefs about eyewitness testimony. Jurors think back on their own experiences and think about the memories they believe are completely accurate, despite the fact that many are actually constructed and incomplete. Studies on human memory show that people can very quickly forget details, or that they never pick up details in the first place, and that the stress of being present at a crime scene, of witnessing a crime like murder, can also interfere with memory formation. As, of course, can suggestive questioning or discussion shortly after the crime, whether intentional or not.

This is an especially large problem when a crime occurred in the distant past. Eyewitnesses may make statements in the immediate wake of the crime, but if they aren’t called to a lineup for several weeks or months, their memories are already starting to slip. If the case takes years to go to court, as it may in some instances, they’re thinking back on memories from many, many years ago. Those memories are likely tainted and incomplete, and it is easy to create suggestibility in a witness on the stand to shift those memories even further from actuality.

Jurors not only believe that eyewitness testimony is sound evidence, they also often expect it. In cases where eyewitnesses are not available, it can be much more difficult to present convincing evidence to the jury, even if that evidence is more reliable than an eyewitness. Extensive physical evidence, for example, needs to be discussed and contextualised. Sometimes it involves the discussion of complex forensics to explain how and why it was linked to the crime, and how reliable that link is. Forensic technicians also face an uphill battle when it comes to accuracy measurements; 99% can be a very reliable estimate, but jurors go to the remaining one percent, and fixate on it as a source of doubt, thanks to the belief that evidence should be 100% accurate despite the fact that this is often not possible. DNA has a margin of error.

Mistaken identification continues, often with dreadful consequences, and juries continue to think that eyewitnesses are reliable, sometimes even when presented with evidence in court to show that an eyewitness may not be the best source of information about the case. Attorneys challenging things like the visibility at the scene may point out, for example, that it could be difficult to identify someone from 40 metres away on a dark night with poor lighting, but that person who saw something in an alley in the depths of winter will be considered more reliable that other sources of evidence.