interrogration

The Psychology of False Confessions

Like a whole bunch of people, I recently got caught up in the new Netflix series, Making a Murderer. It was a fascinating–and frustrating–glimpse into police dynamics, judicial procedure, and human psychology. I’ve written about psychology and the law elsewhere on this blog, and I’m not here to make any particular claims about the series’ subject, Steven Avery’s, guilt or innocence, but one plotline speaks to an interesting psychological phenomenon: the psychology of false confessions.

In the series, Brendan Dassey endures hours of interrogation, without his lawyer present, before finally confessing to rape and murder along with his uncle. The filmmakers compellingly argue that there are serious reasons to believe that Dassey’s confession was coerced (i.e., he never did anything but said he did). In the courtroom proceedings, however, the prosecutor argues that “innocent people don’t confess.” Things don’t seem to have changed much since 1923 when John Henry Wigmore wrote that false confessions were “scarcely conceivable.”

But could that be true? Do innocent people ever confess–and why? It might seem unreasonable to believe that an innocent person would confess to a crime that he or she didn’t commit, but there are plenty of cases in which people did just that.

The Reality of False Confessions

In his excellent book, Unfair: The New Science of Criminal Injustice, Adam Benforado writes, “[f]alse confessions and incriminating statements are the leading contributors to wrongful homicide convictions, present in over 60 percent of the known DNA murder-exoneration cases in the United States.” 

crime1Of course, it’s hard to collect any reasonable statistics about the rates of false confessions that haven’t been later proven false. How could we possibly know they happened? That aside, we can at least say that they do sometimes happen. Let’s take it all the way back to the Salem Witch Trials of 1672 in which about 50 women confessed to being witches, according to historical accounts.[1]  I’m hoping we agree that this is a pretty clear-cut case of false confessions. Otherwise, we’d be saying that at least some of these women really were the witches they were accused of being.

Since then, there have been many provocative cases of people who have confessed to crimes they did not commit, only to be found false later through other evidence (e.g., DNA). There’s Damon Thibodeaux who, after a 9-hour interrogation, gave a recorded confession to murdering his cousin. There’s Juan Rivera, who signed a written confession to the rape and murder of an 11-year-old girl. There’s Phillip Bivens who spent 30 years in prison, having confessed to rape and murder, before being exonerated with DNA evidence in 2010. All three of these people were eventually found innocent, making their initial confessions pretty perplexing. For more examples, see the comprehensive directory at The Innocence Project.

Why it Matters: Confession are Powerful

It’s not just that innocent people confess and the error is quickly discovered. As you can see in the above cases, people end up going to jail–often for a very long time–before their confession is found faulty (if it ever is). The reason is simply that juries are convinced by confessions. After all, if someone admits to murdering someone, surely he understands that this puts him in jail for a very long time, and why would someone do that unless he was guilty?

We tend to have a hard time believing that innocent people confess to crimes, and in fact, classic social psychological research has documented why we’re resistant to the idea that confessions can be coerced out of innocent people.

Maybe you’ve heard of the fundamental attribution errorThe gist of it is that we’re pretty hesitant to say that external forces influence other people’s behaviors. If someone is rude to us, for example, we’re more likely to chalk it up to the person’s salty personality than we are to consider that the person is just having a bad day and would otherwise be gracious and polite.

In one classic study, researchers gave people essays that argued in favor of or against Fidel Castro.[2]  Note that this was in the 1960s. After reading the essays, everyone made a guess about the author’s true feelings about Castro. Naturally, people who read a pro-Castro essay figured the author was a Castro supporter, and people who read an anti-Castro essay figured the author was anti-Castro. So far, this isn’t that unreasonable.

But sometimes, when the researchers gave people that article, they told them that the author had no choice about which position he took. He was assigned to take a particular position and had no control over it. Rationally, this should mean that the readers wouldn’t attribute the essay’s position to the author’s true opinion. However, even when they knew that the author was forced to write an essay with a particular conclusion, they still assumed that the essay’s position matched the author’s true opinion.

So relating this back to false confessions, it might be the case that even when there’s considerable doubt about the truth of a person’s confession, a jury may still find truth in that confession via the same attribution error.

Why Would an Innocent Person Confess?

Although it’s clear it happens and that it can have important consequences, you might still be wondering, “but why!?” After all, when you think about it, do you think you would confess to a murder you didn’t commit? Of course not! So how could it happen?

One reason is that by the end of a lengthy interrogation, people just want to be done with it. Whereas ordinary police interrogations usually last less than 2 hours, research on documented false confessions has found that interrogations that produced false confessions lasted 16.3 hours, on average. That’s a long time to be questioned as an innocent person! I’ll talk more in a bit about why these interrogations might go on so long.

crime3On the one hand, these lengthy interrogations obviously produce fatigue and sleep deprivation, which have been shown to impair decision-making and increase suggestibility. In addition, other experimental research has found that innocent people who resist falsely confessing show greater sympathetic nervous system activation–signs of physiological stress–than innocent or guilty confessors.

On the other hand, after such a long process, suspects are often reminded that if they just confessed, they could end the whole unpleasant experience and go home. In this way, maintaining your statement of innocence is an act of self-control! In general, people often prefer things that will provide immediate benefits, compared to having to suffer now and reap rewards later. It’s like the conflict between eating a delicious slice of pie right now or abstaining from dessert to lose weight in the future.

In an interrogation room, it’s a conflict between getting relief now (confessing) and avoiding jail in the future (maintaining innocence). This still might seem crazy, but after more than 10 hours of badgering, the ability to make it all stop could be too tempting to resist.

There’s another conflict at play, too. Going to court can be a long, drawn-out process, and it’s one that we likely wouldn’t look forward to (especially when it means people will be pointing the finger at us the whole time and that our punishment would be steeper if found guilty). Sometimes, it can just seem easier to admit to a crime than to endure the legal proceedings and risk a worse fate.

In fact, one study simulated these very conditions. Researchers took unsuspecting students and gave them a test of logic puzzles to complete in the same room as another student. Although students diligently did their own work, the researchers later accused these innocent participants of cheating on the test. The students then had a choice: admit to cheating and lose the study participation reward they would have gotten or have the issue brought before the university’s “Academic Review Board” where students are found guilty 90% of the time. The latter option also came with mandatory ethics training if found guilty.

The results are revealing: 56% of innocent students chose to admit guilt for a smaller punishment instead of risk a drawn out ethics case review.[3]  Just to note–the study actually varied whether the results of being found guilty through academic review were relatively harsh or lenient. It didn’t matter. Innocent students were likely  to avoid the risk regardless. These are objectively innocent students who are choosing to admit guilt for fear of a risky trial.[4]  Also, the researchers had a separate group of students who they set up to actually be guilty of cheating. Among these truly guilty students, 89% chose to admit guilt and be done with it, so it’s interesting to point out that at least innocent people are less likely to confess than guilty people.

Some People are More Likely to Falsely Confess

It’s worth noting that there are two groups that are disproportionately likely to give false confessions: young people and cognitively impaired people. Brendan Dassey belonged to both of these groups.

Some studies have simply asked people of various ages whether they would confess to a crime that they did not commit. Researchers tell them a story about someone faced with such a decision and then ask what the correct choice would be. In one study, even after controlling for differences in measured intelligence, 13 – 15 year-old boys were more likely to choose false confession than 16 – 18 year-olds.

There is also a disproportionate amount of false confessions that come from cognitively impaired individuals. There could be many reasons for this, one of which is that this population tends to respond “yes” to a range of questions to which an affirmative response is not appropriate.

In both of these populations, false confessions might also be especially likely because of a desire to gain the approval of adults and people of authority. Indeed, overall, people who falsely confess tend to be more compliant people in general.[5]  With the caveat that these are people who claim to have given false confessions with no ways to independently prove it.

Are the Police to be Blamed?

crime2It would be bold of me to say that police want to get innocent people to confess to crimes. However, I think that a social psychological perspective reveals some definite flaws in common interrogation practices.[6]  There are other concerns relating to common interrogation practice and the often-used “Reid Technique.” These include directly lying to the suspect about other evidence or making empty promises to elicit a confession. This is a clear case of pure deception and dishonesty, and while it does reveal plenty about the interrogation process, it’s a bit outside the scope our discussion of social psychology.

The first is that investigators often enter an interrogation with the expectation that the person is guilty. The goal, then, is not to uncover truth–it’s to extract a confession. Social psychology research is filled with warnings against strong expectations. When we believe something is true, we’re biased to find confirmation of that. We ask leading questions, we interpret ambiguous behavior in ways that align with our expectation, we adopt mannerisms that help elicit the response we expect from others, etc.

In his book, Benforado reviews an experiment with mock interrogators which found that “participants who started off with a guilt mindset were more than 20 percent more likely to confirm guilt than those who started off with an innocence mindset.

Similarly, some of the techniques used by interrogators have been shown to produce false confessions, just as they produce true confessions. These techniques of “minimization” and “maximization” involve trivializing the offense (“You didn’t mean to hurt anyone!”) and offering a deal (“If you give us what we want, we’ll make sure things get settled quickly”). One study that recreated the conditions of interrogating truly guilty and truly innocent participants found that these techniques were successful in producing confessions from innocent participants.

The Psychology of False Confessions and Implanted Memories

Implanting memories is not just the stuff of Inception. I want to wrap up by noting that even after considering all of these forces that elicit false confessions, one of the scary outcomes is that the suspect may begin to believe the confession him or herself.

crimememoryFalse memories have been a common topic of research in psychology and the law. Simply changing the way a question is asked can succeed in influencing a person’s memory of an actual event. There are also issues pertaining to eyewitness identification and faulty memory. For now, though, we’ll just focus on false confessions and their effect on memory.

The real problem is that not only do innocent people sometimes confess due to external pressures, but they can also come to believe that they committed the crime. As one example, a 14-year-old boy was accused of stabbing his sister. Although he strongly denied these claims (and was later actually found innocent), by the end of his extensive questioning he transformed, saying “I’m not sure how I did it. All I know is I did it.”

Some laboratory research has shown how this can happen to unsuspecting members of the public. In one study, researchers accused student participants of breaking a computer by using the wrong keys. In truth, all of the participants were innocent and initially denied the accusation. By using common interrogation tactics, however, the procedure ended with 69% of innocent participants signing a confession. Furthermore, 28% had actually come to believe that they broke the computer, and 9% had even come up with additional details (that didn’t happen) to support their new belief.

So, Do Innocent People Confess?

I think the answer is a pretty clear “yes.” The experience of being interrogated by people who are trained to extract confessions and who believe the suspect to be guilty creates a host of powerful psychological influences that can break down a truly innocent person. Moreover, another set of psychological states makes false confessions believable enough to send people to prison.

Now I’m not saying that every confession is false, nor am I necessarily saying that Brendan Dassey’s confession was false[7]  But come on…it totally was, right?. The real point here is that we really ought to think carefully about how a confession was obtained, and we shouldn’t be so quick to assume that false confessions never happen. I’d encourage you to check out the website of The Innocence Project for more on how these individuals finally find justice and what changes could be implemented to reduce the rate of coerced confessions.

Also, I haven’t covered every detail of the psychology of false confessions, so join in on the discussion and point out additional considerations. Oh, and if you haven’t watched Making a Murderer, it’s about time! Especially if you’re as interested in human psychology as I am.


Feature Image: “Interrogation” by Paul Kehrer –  Licensed under Creative Commons CC BY 2.0 via Flickr, cropped

Footnotes   [ + ]

1.  I’m hoping we agree that this is a pretty clear-cut case of false confessions. Otherwise, we’d be saying that at least some of these women really were the witches they were accused of being.
2.  Note that this was in the 1960s.
3.  Just to note–the study actually varied whether the results of being found guilty through academic review were relatively harsh or lenient. It didn’t matter. Innocent students were likely  to avoid the risk regardless.
4.  Also, the researchers had a separate group of students who they set up to actually be guilty of cheating. Among these truly guilty students, 89% chose to admit guilt and be done with it, so it’s interesting to point out that at least innocent people are less likely to confess than guilty people.
5.  With the caveat that these are people who claim to have given false confessions with no ways to independently prove it.
6.  There are other concerns relating to common interrogation practice and the often-used “Reid Technique.” These include directly lying to the suspect about other evidence or making empty promises to elicit a confession. This is a clear case of pure deception and dishonesty, and while it does reveal plenty about the interrogation process, it’s a bit outside the scope our discussion of social psychology.
7.  But come on…it totally was, right?

2 thoughts on “The Psychology of False Confessions

Leave a Comment