Not long ago, I had a meeting with an undergraduate from a large lecture course. It was the first time I had spoken with her outside of class, and she asked me what type of research I conduct. When I told her, she responded by saying that my work reminded her of a presentation she sat through during freshman orientation–a presentation concerning psychological perspectives on diversity conducted by two faculty members. From that point forward, our conversation took a surprising turn:
Me: "Right, I was one of those presenters."Student: "No, there was a Black professor and a White professor, but the White guy was someone else."Me: "Actually, that was me. Professor Maddox and I conduct diversity workshops, including that one at freshman orientation."Student: "It was Professor Maddox, but the White professor was another guy."
I gave up at that point. I refrained from pulling out my daily planner to show her the date in question with "DIVERSITY WORKSHOP" written in big, capital letters. Of course, even if I had, she probably would've suggested that my calendar referred to a different workshop. Or perhaps that it was a blatant forgery.
It seems like such an easy thing to do, to recall a person you saw weeks, days, or even hours earlier. Indeed, in the legal domain there are few types of evidence that a jury finds more persuasive than the eyewitness who can take the stand and point directly at the defendant while stating, "absolutely, that's the man I saw." But I just returned from the annual convention of the American Psychology-Law Society, where research presentation after research presentation demonstrated how surprisingly inaccurate and malleable eyewitness evidence can be. And the consequences for crime suspects as well as victims are far greater than the marginal ego blow suffered by the unrecognized professor who learns that he apparently nothing more than an average-looking, generic White guy.
Consider, for example, the recent spate of cases in which falsely convicted individuals have been freed after years in prison because DNA analysis has now demonstrated their innocence. The legal non-profit Innocence Project has documented and assisted with more than 200 such exonerations in the past 15 years. In over 75% of these cases, at least one eyewitness had testified against the defendant in the original trial. That is, in more than three-quarters of these mistaken convictions, an eyewitness mistakenly fingered the wrong suspect and helped send him to prison.
A striking example of the limitations of eyewitness memory is told in the new new book Picking Cotton. Years ago, Jennifer Thompson was raped in her apartment at knifepoint. During the assault, she somehow had the presence of mind to study her assailant's face. She would later help police come up with a composite sketch of the suspect, and when a tip led to the arrest of a man named Ronald Cotton, Thompson positively identified him. She was "100% sure" that this was the man who raped her. After serving 11 years in prison for the crime, a DNA test proved that Cotton was absolutely not that man.
Picking Cottonserves as but one example of the limitations of eyewitness memory, but the end of this particular tale is particularly amazing. You see, the book is written by Thompson and Cotton (left). Together. Now co-authors, the mistaken eyewitness and the falsely convicted man travel around the country giving talks to spread the word about wrongful convictions and seeking reforms to eyewitness identification procedures to prevent future miscarriages of justice.
Why is being an eyewitness so deceptively difficult? Researchers have identified two sets of factors that contribute to mistaken identifications. The first are estimator variables. These are characteristics of the actual scene that impair an eyewitness' memory encoding. Like how far away the culprit is from the witness, how poor the lighting is at the scene, the fear and other emotional reactions experienced by witnesses as a crime is ongoing... these and other estimator factors contribute to eyewitness difficulties.
But perhaps more surprising to many laypeople is the important role played by system variables in the performance of eyewitnesses. These are factors under the control of the legal system, of the police who investigate crimes. For example, whether a lineup is conducted using photos or actual people, how "fillers" are chosen for this lineup, what instructions eyewitnesses are given during the lineup, and many, many other considerations have a powerful effect on witness accuracy, as demonstrated by decades of research by psychologists like Gary Wells at Iowa State University.
What to do, then? Psychologists can offer said system many research-based suggestions for reform. As just one example, a leading cause of false identification is that eyewitnesses often react to a lineup the same way students do to a multiple choice test–they pick the best answer among the available alternatives. This process of relative judgment may pay dividends on the SAT, but it leads to serious problems in the legal setting, especially when the true culprit isn't part of the lineup. This can be avoided by showing photos to the eyewitness one at a time as opposed to all at once. With such sequential presentation, witnesses can't use the same strategy of process of elimination that they do with simultaneous lineups. Another effective safeguard is a simple instruction to a witness that the culprit "may or may not" be in the lineup.
I've just scratched the surface of this issue. There is a wide range of factors that contribute to mistaken eyewitness identifications, and a similarly broad catalog of potential reforms to consider in the effort to curtail the phenomenon. But education plays an important role as well–after all, every single citizen is a potential future juror. Now, by no means am I suggesting that eyewitness memories should be dismissed or discounted; to do so would render some crimes essentially unprosecutable. But neither is it reasonable to treat eyewitness identifications as flawless, irreftuable evidence, as many of us tend to do when seated on juries. So it's worth bearing in mind that the challenge of remembering the faces of those around us is not as easy as we often think it is, as my office experiences with my former student made quite clear.
Of course, you might suggest that it's one thing to try to remember where you've seen a professor before, but it's another altogether to remember the face of someone from a crime scene. I agree wholeheartedly. If we can make memory mistakes for people we see during daily interactions, then just imagine the potential for error when you add into the mix the limited exposure times, adrenaline, fear for safety, and surprise that burden many an eyewitness to actual crimes.