In 1996, a 22-year-old Louisiana man confessed to raping, beating and murdering his 14-year-old step-cousin and leaving her body under a Mississippi River bridge. He was convicted of the crime and sentenced to death.
But Damon Thibodeaux was not a murderer. The details of the crime that he confessed didn't match what had actually happened, according to The Innocence Project, a legal group that works to exonerate the wrongfully convicted. Thibodeaux told police that he'd strangled his cousin with a gray or white wire from his car, for example. She'd actually been killed with a red wire, which was found near her body.
Thibodeaux is no isolated example; as many as 25 percent of people wrongfully sentenced to death in the United States (an estimated 4 percent of those on death row) falsely confessed to the crime, research suggests. Eighty-eight of the first 325 DNA exonerations by The Innocent Project involved a case where the suspect had confessed despite being innocent. But why would anyone confess to a crime they didn't commit?
A new study finds that one answer may be sleep deprivation. And it doesn't take much. After one night of lost sleep, almost 70 percent of people participating in a lab experiment made a false confession — urged on only by a couple of stern computer warnings.
University of California cognitive psychologist Elizabeth Loftus and her colleagues used a well-established laboratory method to evoke false confessions from study participants. They brought volunteers into the lab and had them do some computer questionnaires and tasks, all while warning them not to press the "escape" key, lest they erase data and wreck the experiment.
A week later, the participants — 88 in this study — returned and did more computer work after seeing another warning against pressing the "escape" key. Then they all spent the night in a sleep laboratory. Half of them got to sleep, while the other half were randomly assigned to stay up all night.
The next morning, participants filled out more questionnaires. At the end of the task, each person saw a screen stating that a research assistant had witnessed them pressing the forbidden "escape" key and asking them to confirm this account and sign the confession. If they refused, they saw the screen a second time, and were again urged to confess. In reality, none of the participants were guilty.
Despite this, the sleep-deprived had a hard time maintaining their innocence. While 82 percent of the well-rested participants resisted signing the first confession, only half of the sleep-deprived participants refused to sign. When urged a second time, 68.2 percent of the sleep-deprived participants falsely admitted to pressing the key, compared with a 38.6 percent of the well-rested participants. The findings appear this week in the journal Proceedings of the National Academy of Sciences (PNAS).
"People don't want to do this very much if they're rested, because they didn't hit the key," Loftus told Brain Decoder. "But if they're not rested, the data sort of speak for themselves."
Participants were more likely to make a false confession if they scored high on tests of impulsive decision-making, the researchers found. And regardless of whether they'd stayed up all night or gotten shut-eye, self-reported sleepiness was linked to making a false confession. Those who said they were the sleepiest were 4.5 times more likely to confess than those who said they were not sleepy or only somewhat sleepy.
In the real world, suspects of crimes are often interrogated into the wee hours of the night, Loftus said. Prisoners at Guantanamo Bay were also subject to long hours of sleep deprivation, according to reports by the International Committee of the Red Cross and accounts by detainees.
Previous work by Loftus and her colleagues found that sleep deprivation makes eyewitnesses more susceptible to false memories. Suspects falsely confess for multiple reasons, Loftus said. Some are mentally unstable and want to be associated with a famous crime. Others see confession as the lesser of two evils. Someone accused of child abuse might confess if the police tell them that coming clean is the only way to see their kid again, Loftus said.
"There's maybe a little of that happening here," Loftus said of the latest research. "Maybe some of these sleep deprived people just want to go home and just don't feel like having a confrontation."
There's another, more insidious way that sleep deprivation can work, though. When pressured enough, people may actually believe that they are guilty. They start "remembering" the crime and even provide details about what they did and how they did it. Sleep deprivation might have made participants less confident in their own memories, Loftus said.
"Certainly sleep deprivation affects cognitive functioning, your speed of processing, your judgment," she said.
For eyewitnesses, experts advise interrogation by someone who doesn't know who the suspect is in the crime, in order to avoid memory contamination. Something like that would be harder to do when questioning suspects themselves, Loftus said, but there might be other ways to combat the suggestive effects of interrogation. One way to combat false confessions would be to videotape all interrogations, she said. The footage might be able to show if a suspect was exhausted or not, or if they were being badgered or otherwise pressured by the interrogators.
In the 1996 murder case, Thibodeaux recanted his confession the next day, and no physical evidence linked him to the crime. But confessions are powerful. As a study published in the journal Law and Human Behavior in 1997 (the year Thibodeaux was convicted) revealed, juries are more likely to convict a suspect who falsely confessed, even if they know the confession was untrue. Thibodeaux spent 16 years behind bars before DNA testing exonerated him. He was released in 2012.