It took Eddie Vedder and a Dixie Chick to seal the deal. That, and a little DNA. But last Friday, three men in Jonesboro, Arkansas walked out of prison after serving 18 years for a grisly crime they didn't commit.
Known as the West Memphis Three, the trio included Damien Echols, sentenced to death, Jason Baldwin, and Jessie Misskelley, both serving life sentences. In May, 1993 police found the bodies of three 8-year-old boys in a drainage ditch. They were naked, mutilated, and hogtied. The men were convicted based on a witness's testimony that he heard the three men talking about the murders, and on the confession of Jessie Miskelley, who also implicated the other two men. But DNA evidence collected at the crime scene and analyzed only in 2007 did not match any of the suspects. After four years of legal maneuvering, a documentary, and lots of celebrities taking up the cause, the West Memphis Three walked free.
On the face of it, it seems impossible that someone would make a false confession. Can you imagine any circumstances under which you would confess to a crime you didn't commit? Of course not! And yet the evidence shows that many people do just that. When the famous aviator Charles Lindberg's baby was kidnapped in 1932, more than 200 people confessed to the crime. According to the innocence project, of the 240 people exonerated so far by DNA evidence, about 1 in 4 were convicted based on a false confession. False confessions happen, and this post shows why.
Some false confessions are fairly easy to understand. In cases where suspects are tortured or coerced, they may say anything just to stop the abuse. In other cases, the confession is strategic. If the police convince a suspect that he will be convicted no matter what, then confessing to get a more lenient punishment can begin to make sense. The cases that are the hardest to understand -- and the most fascinating in their psychology -- are those in which the suspect comes to actually believe his own false confession.
What could make someone think that he committed a crime when he's innocent? According to psychologist Saul Kassin, this kind of outlandish belief can result from the simple and common error of confusing the sources of one's own thoughts under certain extreme conditions. To see how, let's look at a landmark study on the psychology of false confessions carried out by Kassin and Kathleen Kiechel. Undergraduate students volunteered for what they thought was a study on typing speed. The experimenter started by warning the subjects that they must not hit the ALT key because the computer program had a bug, and hitting this key would cause it to crash and the data would be lost. Volunteers were paired up with another student (who was actually an experimenter posing as a research subject). We'll call this experimenter/"subject" the reader because her job was to read a series of letters that the real subject (the typist) had to type using a computer keyboard. A minute into the typing, the computer was rigged to crash. At that point the experimenter got upset and accused the typist of hitting the forbidden key. In the control condition, that's all that happened. But in the critical condition, after the typist denied touching the key the experimenter turned to the reader and asked if she saw anything. The reader said yes, she had seen the typist hit the ALT key while typing.
Now, we are ready to see if the maligned typist will confess. To do this, the experimenter hand-wrote a confession note to the study's principle investigator which said, "I hit the ALT key and caused the program to crash. Data were lost." And asked the typist to sign it. In the control condition with no witness, 50% of subjects signed the confession. In the critical condition with a witness's false testimony, 95% signed. But did the subjects really believe their confessions?
To test this, the experimenter asked the typist to sit in a waiting room, where another new subjects was waiting to take the experiment (actually, yet another confederate working with the experimenter). This new confederate asked, "what happened?" The typist's reply was audiotaped to see if she admitted pressing the key or crashing the program. In the control condition only 6% said that they had caused the computer to crash, but 55% of those in the witness group "admitted" their "crime." Because these subjects believed they were simply talking to another student, it's a good bet that they actually believed that they had hit the key. Finally, the experimenter came back and asked the typist if she could reconstruct exactly how she had hit the key. None of the subjects in the control condition confabulated details, but 21% in the witness condition came up with specific details about how and when they had supposedly hit the key.
Of course, confessing to crashing a computer is a long way from confessing to murder. It would obviously be unethical for researchers to pressure people into false confessions about high-stakes events like crimes, so we have to make do with more mundane topics. But the steps involved in producing either kind of confession appear to be the same. According to Kassin, the four steps are:
(1) The police accuse the suspect and confront him with false but convincing evidence of his guilt, such as testimony from an eyewitness, results of a failed polygraph, or a fingerprint. In some cases the police deliberately deceive the suspect, and in others the evidence is simply a mistake. Either way, the suspect feels as if he has a major conflict between his own memory and the supposed facts.
(2) The suspect attempts to reconcile his memory with the evidence, often by presuming that he blacked out or repressed the event.
(3) The suspect makes a tentative admission, often expressing an inference rather than a belief, such as "I guess I must have done it."
(4) In the most critical step, police coax the suspect to repeatedly re-tell the story of how he committed the crime. Either knowingly or unknowingly, the police often use leading questions and supply details about the crime that "only the killer could know" as they help the suspect piece together a narrative that explains how he carried out the crime. After hours of this, the suspect's initial inference that he might have done it can morph into a belief that they really did it.
By the end of this process, the suspect has come to confuse the accusation of the police, "you did it" for his own belief, "I did it." And all of the imaginings and storytelling he engaged in along the way become vivid details that are mistaken for actual memories.
It's true that not all -- or even most -- suspects make this kind of memory error. It is most likely to happen under conditions of extreme stress and sleep deprivation. Memory becomes more vulnerable if the suspect is using alcohol or drugs at the time of the interrogation. Finally, suspects are especially vulnerable if they are young (adolescents' memories are more fragile than adults), mentally ill, or mentally retarded. If we think about all of these factors together, it becomes easier to see how someone might be pushed into a false confession even about a serious crime.
Which brings us back to the West Memphis Three. The confessor, Jessie Misskelley, has an IQ of 70. That's right on the border of mental retardation. His confession came after 12 hours of continuous interrogation. Although he later recanted, the damage was done with the jury. I suspect that the members of the jury, like most reasonable people, found the idea of a false confession impossibly unlikely. It just violates common sense that someone would confess to a horrible crime unless they were guilty. And yet, social psychology has been demonstrating for decades that common sense is not always a reliable guide to truth, and that the power of the situation is immense (think of Milgram's famous obedience studies, for instance).
If you ever find yourself on jury duty, the research suggests that you should treat a confession just like any other piece of evidence. Rather than iron-clad proof, it's just one more piece of information that might be true and might be false. There is no alternative but to weigh the credibility of both the confessor and the interrogators. Kassin advocates that interrogations should be videotaped in their entirety and the jury should watch the whole thing rather than just hearing about the confession. That way, the jury can see what the suspect went through before confessing.
So, can you imagine making a false confession? I still can't either. None of us thinks we would. But neither did those 60 people (and counting) exonerated by the innocence project... until they found themselves confessing. All of this means that it is next to impossible to know what we ourselves would do if we were under the pressure of an interrogation. Above all, it suggests a sort of humility in how we think about cases of false confessions. Rather than, "I would never do that!" maybe the better response is, "There but for the grace of God go I."
Sources referenced in post:
Kassin, S. M., & Kiechel, K. L. (1996). The social psychology of false confessions: Compliance,
internalization, and confabulation. Psychological Science, 7, 125-128.