Being sleep-deprived makes people much more likely to give false confessions

Why in the world would someone confess to a crime he didn't commit?

It's a baffling question that has long confounded the criminal justice system. Confessions are powerfully convincing evidence for juries — but false confessions are also relatively common. According to the Innocence Project, one in four people who have been exonerated for crimes they didn't commit confessed to that crime.

Psychologists have documented several reasons this might occur. The big one is that interrogating police officers can impose their suggestions on suspects: "We have evidence proving you were there!" "Your fingerprints were found!"

But there may be another reason people will confess even when they're innocent: They're exhausted.

Law enforcement "really needs to be super careful when a person is being interrogated after they have been up a long time," says Elizabeth Loftus, a co-author on a new study on sleep deprivation and false confessions in the Proceedings of the National Academy of Sciences.

How sleep deprivation can lead to false confessions

According to Loftus's study, the majority of false confessions occur when interrogations last more than 12 hours. That fact made her and her colleagues wonder: How much of a role does sleepiness play? Sleep deprivation, after all, is awful for the body and mind, decreasing our abilities of reason and judgment.

So they designed an experiment meant to elicit false confessions.

In Loftus's experiment, participants were told to do assignments in a computer lab and were warned never to press the "escape" key on the computers. Doing so, they were told, would ruin the study. Then, after a few assignments, half the participants got to go to sleep overnight. The other half were kept up all night.

The next day, the participants were all given a form to sign — a form that basically said they were seen pressing the escape key. Anyone who signed this form was basically admitting to guilt and to ruining the study.

Only about 18 percent of the well-rested participants signed the form (such is the baseline power of an authority figure demanding guilt). But the results were more dramatic in the sleep-deprived condition. "That 18 percent now has risen to 50 percent," Loftus says.

Even those who refused at first seemed pliable. When they were told a second time to sign the form and admit their guilt, 68 percent of sleep-deprived participants gave in. (On the second request, 38 percent of the rested participants signed.)

To be sure, pressing the escape key is not the highest of stakes, but other studies have found that participants are indeed mortified when they believe they've ruined a whole experiment. "It would probably be scientifically prudent to go out and demonstrate it again with a more serious paradigm," Loftus admits. But there are also ethical limits to how far researchers can manipulate participants into thinking they've done something horrible.

The idea here is that in the interrogation room, feeling tired may wear down people's capacity to fight back against the suggestions of the police officers. Perhaps that's why the study found that people who are more prone to making rash decisions — i.e., less thoughtful — were even more susceptible to this tactic.

PNAS

Another question is what sleep deprivation does to the guilty, Loftus says. It could be that police officers use long interrogations because they often do succeed in getting confessions from people who actually commit the crime. In other words, sleep deprivation could wear down the motivation to lie just as it wears down the motivation to insist on innocence.

"It’s not something that we looked at," Loftus says. "But it is something someone should look at next."

Loftus is arguably the most influential psychological researcher when it comes to criminal justice in the United States. She's famous for demonstrating how memories can be easily manipulated or even falsified. She's found the mere suggestion by an interviewer that cars "collided" instead of "hit," will lead people to recall a car accident as being more severe than it was. She's also found that through subtle suggestions, people can be made to recall childhood memories that never happened.

Both the New Jersey and Massachusetts Supreme Courts have ruled that juries must be informed of the imperfection of human recall before testimony because of her research.

Overall, Loftus says research from her lab and others is finding that the harsh, "Law & Order" (her term) style of interrogations may not be the ideal way to get to the truth of matters.

It may be "better to have a milder, gentler information-gathering approach," she says. "You want to develop a rapport with the person and get them talking, because the more you can get them comfortable and talking, the more likely it is they will provide you with clues and details that you can use to exonerate or convict."

Will you support Vox’s explanatory journalism?

Most news outlets make their money through advertising or subscriptions. But when it comes to what we’re trying to do at Vox, there are a couple of big issues with relying on ads and subscriptions to keep the lights on:

First, advertising dollars go up and down with the economy. We often only know a few months out what our advertising revenue will be, which makes it hard to plan ahead.

Second, we’re not in the subscriptions business. Vox is here to help everyone understand the complex issues shaping the world — not just the people who can afford to pay for a subscription. We believe that’s an important part of building a more equal society. And we can’t do that if we have a paywall.

So even though advertising is still our biggest source of revenue, we also seek grants and reader support. (And no matter how our work is funded, we have strict guidelines on editorial independence.)

If you also believe that everyone deserves access to trusted high-quality information, will you make a gift to Vox today? Any amount helps.

One-Time Monthly Annual

$5/month

$10/month

$25/month

$50/month

Other

$

Yes, I'll give $5/month

Yes, I'll give $5/month

We accept credit card, Apple Pay, and Google Pay. You can also contribute via