Editor's note: This is an excerpt from the latest episode of the Invisibilia podcast and program, which is broadcast on participating public radio stations. This story contains language that some may find offensive.

This is the story of a prisoner who committed a horrible crime and says he's no longer the same person who did it. It's also the story of why it's so hard for us to believe him.

In the early 1960s, a young psychologist at Harvard University was assigned to teach a class on personality. Though Walter Mischel was excited to prove himself as a teacher, there was one small problem: He didn't happen to know very much about personality.

"So, realizing I had to teach this stuff, I decided to look at the literature," says Mischel, who now works at Columbia University. "And I found myself enormously puzzled."

Mischel, like pretty much every other psychologist at the time, had some basic assumptions about personality. The first was that people had different personalities, and that those personalities could be defined by certain traits, such as extroversion, conscientiousness, sociability.

At the time, personality researchers liked to argue about which traits were most important. But they never argued about the underlying premise of their field — that whatever traits you had were stable throughout your life and consistent across different situations.

"For example, a friendly person is someone who should be friendly over time," Mischel says. "So if he's friendly at 20, he should be friendly at 25. And if he's friendly, he should be friendly across most situations in which friendliness is a reasonable and accepted possible way of being."

Thus an honest person would behave like an honest person no matter where they went or how much time passed, and a criminal would remain a criminal.

But when Walter Mischel sat down to do his literature review, he didn't find much support for the idea that personality is stable. "I expected to find that the assumptions would be justified," he says, "and then I started reading study after study that found that actually the data didn't support it."

One enormous study on honesty in children was done way back in 1928. The researchers, Hugh Hartshorne and Mark May, had put thousands of children in different settings where they had the opportunity to cheat or steal.

"And it came out with results that were shocking at the time," Mischel says. The same child who cheated in math class could be honorable in a different class — no cheating. "They were not consistently anything," he says. "They were inconsistent in their honesty."

The studies Mischel was reviewing were all looking for consistency in personality across situations — and none of them were finding it.

And researchers seemed to be ignoring this, dismissing the fact that study after study was finding no consistency in personality.

Mischel ended up writing a book called Personality and Assessment in 1968 that challenged some of the most basic ideas we have about the role personality plays in our lives. He said that the idea that our personality traits are consistent is pretty much a mirage.

But that idea was so hard for people to wrap their heads around. Mischel tried in many ways to make it stick, but never did. In fact, the irony of Walter Mischel's career is that he himself is remembered as proving the very opposite of what he actually believes.

It has to do with Mischel's most famous experiment, called the marshmallow test, which he first conducted in 1960. You can still find videos of it on YouTube. Mischel would give a small child a marshmallow, a cookie or a pretzel, telling her or him that they could eat it now — or if they could wait for a few minutes, they'd get two marshmallows or cookies. Then he left the room. Given that the children in the study were 4 to 6 years old, the marshmallow often got gobbled up.

But sometimes Mischel told the child ahead of time that she could just pretend that the marshmallow was not really there. Then "the same child waits 15 minutes," he says now. "It's a very small change that's been made in how the child is representing the object — is it real or is it a picture? And by changing the representation, you dramatically change her behavior."

The vast majority of children in Mischel's study were able to delay gratification when they reframed their interpretations of the situation in front of them.

The point of the marshmallow test was to show how flexible people are — how easily changed if they simply reinterpret the way they frame the situation around them. But that's not the moral that our culture drew from the marshmallow study. We decided that those traits in the preschoolers were fixed — that their self-control at age 4 determined their success throughout life. They're happier, have better relationships, do better at school and at work.

The marshmallow test became the poster child for the idea that there are specific personality traits that are stable and consistent. And this drives Walter Mischel crazy.

"That iconic story is upside-down wrong," Mischel says. "That your future is in a marshmallow. Because it isn't."

So how did we get it so wrong? Psychologists have come up with all sorts of theories. One is that the consistency we see in people's personalities is an illusion that we create. No matter how people behave, we shoehorn them into the idea we already have of them.

Lee Ross, a psychologist at Stanford University, has another intriguing idea. He had read Mischel's book on personality when it came out in the 1960s and immediately understood the profound puzzle it presented. He thinks we actually are seeing consistency in human behavior, but we're getting the reason for it wrong. "We see consistency in everyday life because of the power of the situation," he says.

Most of us are usually living in situations that are pretty much the same from day to day, Ross says. And since the circumstances are consistent, our behavior is, too.

But sometimes the dynamics at work and home ask us to be different people. The violent gangster at work may be the kind father at home. In the 1960s and 1970s, a number of experiments were done where the researchers put people in an extreme situation to see if it would change their behavior.

One of the most infamous is the obedience experiment done in the 1960s, by Stanley Milgram, a social psychologist at Yale University who was intrigued by the concepts of conformity and authority. In the experiment, a "learner" was wired with electrodes, and a "teacher" was told to give the learner an electric shock every time they got an answer wrong. The learner wasn't actually getting shocked; the part was played by actors who pretended to be hurt. But the teachers didn't know that, and they kept administering what they thought were stronger and stronger shocks, even as it made them very uncomfortable, because they were in a situation that required them to do it.

The point, Ross says, is that ultimately it's the situation, not the person, that determines things. "People are predictable, that's true," he says. "But they're predictable because we see them in situations where their behavior is constrained by that situation and the roles they're occupying and the relationships they have with us."

Even though these experiments were done almost 50 years ago, we're still struggling with the notion that human personality and behavior isn't a constant. Consider the story of Delia Cohen. Like most of us, she believed in a core consistency in humans; she'd seen it in people her whole life. The good people were good; the bad, bad.

Cohen had been doing some work for the TED organization, the one that does TED talks, and had heard about a TEDx event that was put on in a prison, by prisoners. So she went to Marion Correctional Institution in Marion, Ohio, to see it.

One of the first prisoners she met was a man named Dan. He had the word "hatred" tattooed on the back of his neck.

But as soon as she started talking to Dan, Cohen completely forgot about the tattoo. His personality didn't seem to have anything to do with hatred.

The prisoners' presentation included poetry and music. Cohen liked it so much that she told Dan she wanted to work with him to re-create it in prisons across the country. Then a friend of hers Googled Dan and started reading descriptions of his crime to Delia. It turned out that Dan was a sexual predator who had raped a woman at knife point. Cohen says: "I was absolutely horrified."

She went home and tried to figure out how she should think about Dan (the prison asked us not to use his full name so it would be less difficult for his victim). Was he the kind, creative, competent person she had met, or the brutal, sadistic personality who had committed the crime?

To find out, she decided to ask him. She sent him an email.

"I've had a lot of people contact me," Dan says. "I don't know if they see some of my poetry online or they see something about me and they write me." Then at some point those correspondents Google Dan like Cohen's friend did, and they vanish. "So I'm kind of used to people having adverse reactions."

Dan emailed Cohen back, saying he was willing to talk with her about his crime. But the conversation didn't go well. "It was an incredibly awkward conversation," Cohen says. "Basically he said he was a horrible, horrible person."

After talking for almost an hour, Cohen didn't feel like she was any closer to answering the question that plagued her: Was there something in Dan's personality that caused the crime, and did that thing still exist?

We decided to ask Dan ourselves.

When he committed the crime, Dan says, "I was a real live piece of s***." But he says that person has ceased to exist.

It happened six or seven years into prison, Dan says. His best friend in prison had stolen something from another inmate, and Dan, who had a reputation for being violent, told the other inmate to back off. Then he went back to his best friend and beat him up.

"I remember while I was doing it, he was asking me to stop, and I was like, 'How could you be so stupid? I'm going to beat the stupid out of you.' "

Dan remembers looking down at the man he was beating, and thinking, "Whoa, what am I doing? This is someone I say I love. This is someone I care about. This is someone I say I treat like family. And this is how you treat family?"

Dan says that's the last time he was physically violent. He decided to quit the group of guys he hung out with in prison. He physically isolated himself. It was hard. "It was easier to be a no-good m*****-f***** than it was to be alone."

Dan says it took him about two years to reconfigure his personality. He wanted to be less aggressive, less impulsive, more conscientious. He says he's now a different person. But he knows most people won't see him that way.

"I'm forever going to be a criminal," he says, "which I'm not. I've become a completely different human being at this point."

Delia Cohen had a hard time accepting that Dan had changed; you hear those words so often from people, and they're often not true. But she decided to suspend her disbelief and work with him on the TEDx project. They started exchanging emails.

Cohen knew that there are people whose horrible crimes really do emanate from their personalities, such as psychopaths. But after more than a year of working with Dan, Cohen felt sure he wasn't a psychopath.

As she tried to figure out if something rotten remained in Dan and the other prisoners she worked with, she decided to reject the frame of reference she used to think about people. "All these people I've met that have done really horrible things are not horrible people," Cohen says. "They're not bad people. Which was shocking to me."

She's still working with prisoners on the TEDx project, and she says she no longer even thinks about their crimes. "I'm more curious about who they are now."

Even though Dan says he's no longer the man who committed the crime, he knows why he's in prison. "I have to atone for my crime. But I realize now I'm just paying for someone else's debt. The person who committed the crime no longer exists."

There's something more than a little disturbing about that sentence — being in prison now for someone else's crime. But just because it's disturbing doesn't mean it can't be true.

"Maybe we're not thinking right about who we are and what we could be," says Walter Mischel.

It's no wonder, Mischel says, that we're drawn to this idea that personality is important and stable. It makes us feel better. "I mean, how can you marry anybody unless you believe that they are basically going to stay the way that you have them pictured now?" he asks. "We like to feel like we're living in a stable world."

We realize the outside world can change in a heartbeat, "but when it comes to human beings, we really don't have tolerance for realizing that there is an enormous amount of instability."

Still, we're not slaves to that instability. Traits and life situations both affect our behavior, Mischel says. But so do our minds.

The beliefs, assumptions, expectations that you've gotten from your friends, family, culture — those things, Mischel explains, are the filter through which you see the world. Your mind stands between who you are, your personality and whatever situation you are in. It interprets the world around it, and how it feels about what it sees. And so when the stuff inside the mind changes, the person changes.

Copyright 2016 NPR. To see more, visit http://www.npr.org/.