University Place

Inside the Criminal Mind

– Michael Koenigs: Welcome everyone to the Neuroscience and Public Policy Seminar. I’m very pleased to introduce our speaker today, Dr. Arielle Baskin-Sommers, an associate professor of psychology at Yale University. Dr. Baskin-Sommers is no stranger to Madison. She’s a star alumna of our clinical psychology PhD program, during which time she worked with Professor Joe Newman on his longstanding program of research with state prison inmates. Following her time in Madison, she completed an internship and fellowship at McLean Hospital at Harvard Medical School before taking her current faculty position at Yale. She’s received numerous early career awards from professional organizations, including the Society for the Scientific Study of Psychopathy, the Association for Psychological Science, and the Society of Criminology. Her record’s been funded by both the National Institute of Health and the National Institute of Justice. What I find particularly impressive about her work is how her program of basic research on the psychological and biological mechanisms of psychopathy have been used to inform novel treatment strategies for this serious and costly disorder.

Dr. Baskin-Sommers is one of the world’s leading experts on psychopathy, particularly with regard to its manifestation and psychobiological mechanisms in the criminal offender population. So we’re very pleased to have her here. Please join me in welcoming Dr. Baskin-Sommers. [audience applauding]

– Well, thank you so much for having me and coming out here in the midst of all the crisis that is going on. I’m really excited to be back in Madison, my first time since receiving my PhD here, and I’m excited to share some of the work our lab has been doing along with other labs to identify mechanisms that contribute to antisocial behavior. And today I’ll be focusing on a prototypic individual who engages in antisocial behavior, that is the psychopath. And so, I’ll be starting by really defining what I mean by psychopathy and distinguishing it from other mental health disorders that we commonly see in the criminal justice system. Then talk about how generally psychopathy is viewed in the eyes of the law and policies that surround the handling of this type of disorder in the criminal justice system, and how some of our mechanistic work might challenge that view and assumption.

And then use that type of model to talk about treatment and management programs that might be more targeted mechanistic base. So to start in terms of psychopathy, generally when I’m talking about this disorder, I’m talking about individuals who display a combination of characteristic traits that fall into four domains. The first being individuals who are generally quite glib and superficially charming, and they can be very manipulative. You really actually enjoy engaging with them, enjoy having a conversation with them, despite sometimes knowing the things that they do or how they’ve just conned someone or committed an act that you might seem quite heinous or really difficult to understand. And so to exemplify this set of traits, I wanna play a brief clip from ESPN’s documentary about O. J. Simpson. In this clip, I would like you to hear how other people describe their engagement with O. J. and how, despite knowing what he possibly did to his ex-wife and others, and then certainly did following that trial, how they still really felt that they were drawn in by this individual.

– Woman: O. J. , I wanna say I don’t like you, I can’t stand you, I wanna call you names, I wanna throw you right out of here, but you know what?

– O. J. Your husband’s watching, better watch it. [laughs]

– You’ve done it to me. Can I invite you to a party?

– Yeah, sure, I’ll be there.

– Woman: His friends would call it the O. J. effect.

They would say, “Yeah, you got O. J. ‘d. ” Being O. J. ‘d is being charmed. The confusion that you feel after you’ve been in his presence, I don’t think he was not guilty, but I was in touch with the fact that I wanted to think he was not guilty. – So those are the types of characteristics that really fall within the interpersonal domain. Another domain is the affective traits of psychopathy, and so what’s meant by this is a general sense of shallow affect, either a lack of depth in emotion, so they might express or feel certain emotions, but very briefly in a fleeting way, or a lack of breadth in emotion, so they actually don’t feel a lot of emotion, but might feel certain ones like anger very strongly. You also tend to see with psychopathy, individuals displaying a lack of regret about the decisions that they’ve made and the implications for themselves, and remorse, the effect on other people.

And very characteristic is a general sense of a lack of empathy or callousness. So this next video, if it plays, is an interview with someone who is on death row for murdering a family. They’re gonna go through a few details that can be somewhat brutal, so I just wanna warn people. But I want you to listen, if you’re willing, to how this man describes his emotions. It’s almost like a drug, an adrenaline rush. You don’t hear the types of words that we would typically associate with fear, anger, or sadness. And then when challenged or confronted with difficult information, how the interviewee really shuts down and doesn’t even engage with the interviewer.

– Man: When you kill someone.

– Man: It’s just like that drug, I’m after that drug again. I don’t have a on and off switch, I’m just after that drug. I’m after that feeling. [tense music]

– Narrator: During the beating, Elaine Dardeen went into spontaneous labor, giving birth to a daughter who investigators believed to be alive at the time. The new baby was also beaten to death and the body of Dardeen’s husband turned up a day later, shot three times in the head.

– Okay, and so then the last set of characteristics that we associate with psychopathy are pretty common across a lot of individuals who find themselves in the criminal justice system, and these are impulsive and antisocial traits. So what makes this different in individuals with psychopathy is that they generally are showing a lack of decision-making that considers future orientation, so not considering the consequences that might result from their behavior for the next day or a following month. So they might quit jobs frequently and not have another one lined up. They might move on a whim and not think through these types of decisions. But they don’t always show kinda the classic impulsivity deficits that you would expect to see on a lot of neuropsych tests. So they’re not traditionally showing cognitive control problems in the way that you would see with other types of individuals who very classically have impulse control problems. In terms of antisocial behavior, you see this starting actually quite young in individuals with psychopathy, with an average age of onset of around five or six of beginning to engage in some sort of criminal or violent behavior.

I’ll never forget one of my first interviews was with someone who’s currently incarcerated at Dodge Correctional, and he said the first time he stabbed someone was when he was four, and he stabbed his preschool teacher with scissors. And so you begin to hear these stories that can be quite intense, and then they start very early in life. The other really interesting thing about individuals with psychopathy is they actually are quite versatile in terms of the crimes they commit. So they tend to be kinda the Jacks and Jills of all trades when it comes to criminal activity. You see on average individuals with psychopathy committing at least six different types of crime. So this might be a drug charge, an assault charge, arson, fraud, miscellaneous charges. And so they don’t tend to be specialists in terms of their crimes, and this is one of the common misconceptions that serial killers are most often psychopaths, and that’s not generally true because most serial killers are actually specialists. Their job is to kill people. And that’s not what you tend to see with individuals with psychopathy, who tend to just kind of commit crimes on a whim in almost the same way that we’ve talked about their impulsivity. So hopefully, fingers crossed, my favorite video clip is coming up, which comes from HBO’s documentary about Robert Durst.

So for those that are not familiar, Robert Durst comes from the Durst family who owns a lot of real estate in New York City. So he’s very well-off, very wealthy. At the point that this clip is going to start, he was on the run from the police for potentially killing his landlord.

– Narrator: Police searched his rental car outside [camera flashes] and found two loaded guns, [camera flashes] some marijuana, [camera flashes] $38,000 in cash, and an ID for one Morris Black.

– Man: Why would a guy with $520 in his pocket and $37,000 in the trunk walk into a Wegmans and steal a sandwich?

– So just for context, Wegmans’ hoagies are about $8. 50. They’re very good, but I’m not sure they’re worth potentially getting caught by the police. But that kinda last question that this documentary asks is really at the center of a lot of the work we do. Like how is it that if you have the means and resources to do something legally, that you might make a decision that we would all kinda consider shocking or defying any sort of logic? So when we talk about psychopathy, we’re really talking about individuals that are displaying these two types of traits, the Factor 1 traits being interpersonal and affective traits, and the Factor 2 traits being impulsive and antisocial. These traits combined is what we would consider psychopathy.

It’s represented about 1% of the general population, which is a similar rate to schizophrenia, but about 20-25% of currently incarcerated individuals. We usually, particularly within a criminal justice setting, will use Hare’s Psychopathy Checklist Revised, which is an interview that essentially covers all of those characteristics that I just described to you. We then score the 20 items and are able to get a continuous score anywhere from 0 to 40, and then also use a clinical cut score of 30 or above in the United States to be able to assess psychopathy. And keeping in mind these two sets of characteristics is really important because if we look at any one set alone, we actually might confuse it with other disorders. So for example, the interpersonal and affective characteristics overlap substantially with narcissistic personality disorder, particularly in terms of the callousness and lack of empathy. But generally, you don’t see in narcissistic personality disorder an increase in antisocial or certainly criminal behavior. Similarly, if we were to look at Factor 2 traits alone, the impulsive and antisocial traits, this substantially overlaps with antisocial personality disorder, which really is similar to psychopathy, but without the interpersonal and affective traits. And so you commonly would see these types of externalizing, impulsive, and antisocial behaviors in these two disorders. So if we look at just the behavior alone, it might seem like these disorders are similar, and it’s really looking at the mechanisms and considering psychopathy as really a combination of these traits that allow us to see it as a separate type of disorder that’s important to consider in its own right, particularly in the eyes of the criminal justice system. So now I’m gonna talk a little bit about how psychopathy, then, is viewed in the eyes of the law currently and how some of the science that we’ve been doing might challenge that view.

So psychopathy currently is not considered a form of serious mental illness in the eyes of the law. That is more commonly something like schizophrenia, a mood disorder, bipolar disorder, some anxiety disorders, or post-traumatic stress disorder. And because of this, it’s rarely considered in terms of competency or culpability hearings. So it would not be possible for an individual with psychopathy to be considered not guilty by reason of mental defect or insanity. Sometimes psychopathy is considered at a mitigation stage, so at the sentencing phase, particularly because we know that psychopathy is associated with genetic risk. You see a heritable risk factor of about 0. 7 for the callous emotional trait, so it seems to load quite heavily in terms of genetics. There’s also a lot of work that suggests that individuals with psychopathy experience early life traumas, and there’s a lot of work, and some I’ll talk about today, suggesting that there is no neurobiological aberrancies in individuals with psychopathy. So despite all of these characteristics, which commonly we would consider in a mitigation stage, the success with which psychopathy is considered a mitigating condition is inconsistent. So it’s sometimes tried, but the application of it and the outcome of whether it actually reduces a sentence is not totally consistent.

And so we might wanna consider, then, how their knowledge base about underlying mechanisms for psychopathy might challenge the position of this disorder in the eyes of the law. So some of the things we know about individuals with psychopathy is that they really show difficulties with emotion processing and emotion recognition. But there’s a nuance to their difficulty, so for example if you ask individuals with psychopathy to identify this emotion, which is showing fear, but it’s somewhat ambiguous, they tend to be pretty bad at that. But if you show them a classically very clear fear face, they are able to do that emotion identification. And so while psychopathic individuals certainly have difficulties processing emotion and have difficulties identifying emotions in others, there are some circumstances in which they’re able to do it, and it’s essentially when it is very clear, which emotion typically is not, or very obvious in the sense of the goal, the thing that they are trying to attend to, which again very rarely is emotion the thing that we’re walking around attending to, unless you’re basically in a therapy session. That’s the only time where I think focusing on your emotion is your goal. We also see that psychopathy is associated with deficiencies in the amygdala, so this is both structural and functional evidence, and ventromedial prefrontal cortex across a variety of contexts and different types of tasks. So there is really strong evidence that there’s something disrupted in emotion processing in individuals with psychopathy. We also in our lab recently completed a connectivity analysis using graph theory to look at hubs of emotion processing and whether there are certain regions of the brain that might act as essential communicator to other regions of the brain. So you might imaging saying, “Well, is this red dot here a hub of emotion processing “that might then communicate to other regions “that would be important for considering that information?” And essentially, we see that individuals with psychopathy are really less likely, so looking at this upper region, to have emotion networks as a central hub of neural communication.

This was just in resting state using graph theory. We’ve now begun to replicate this in a sample of nine to ten-year-olds who are showing callus, unemotional traits, and see that the lack of emotion processing as a hub of communication predicts, then, their deficiencies on the types of emotion identification tasks that I just described. So this is one set of difficulties we see in psychopathy. Another set is not just that they seem to be deficient in emotion processing, but they seem to be really good at goal-directed behavior. So individuals with psychopathy are amazing at focusing on their goals and essentially screening out any distractor. So an example of this is if your goal was to focus on the letter N, but the red box suggested that you might receive a shock, all of us would still tend to the red box ’cause we wanna know if we’re gonna get a shock or not, whereas individuals with psychopathy basically ignore the red box and just focus on the N. But if you ask them to focus on the red box, they can do that just fine. So again, suggesting that they can focus on things that might be threatening or emotional, it just depends on their goal. In a non-emotion context, if you are asking individuals with psychopathy to name the color of the box here, blue, they’re actually gonna show less interference with the word red. All of us might stop and pause a little bit because blue and red are not the same thing, we have to think about it, use a little cognitive control.

Individuals with psychopathy show markedly reduced interference. And in line with that, we actually see that individuals with psychopathy show increased activity in the prefrontal cortex, particularly during decision-making tasks, including moral decision making and more complex emotion tasks. In the same graph theory paper, we also recently were able to look at how efficiently information is communicated in individuals with psychopathy starting with resting state. And so you might imagine one way for these three regions of the brain to communicate is in a very linear type of way. So information goes from one hub to the next, to the next, and the advantage of this is that it’s quick communication with dots or hubs that are right next to one another, but that over time there is a loss of fidelity in information, and it’s actually quite inefficient if you end up wanting to communicate with the hub that’s all the way up here. So this is what we might call a line-like communication pattern. The alternative is something that’s more like this red or maroon color, which is a star-like communication. There’s one central communicator that then receives all the information and communicates any information, neurally or otherwise, out to all of the other hubs. So my grad student, when he presents this information, basically says it’s like how all the grad students come and talk to me, and then I just scream out at all the other grad students, which, you know, I’ll take it, but I’m not sure as he means it as nicely as I mean it in this instance. And so we can be able to quantify for each individual how well their neural communication is something that is line-like, the black and white dots, or more star-like, with the star-like being a more efficient pattern.

And what we see is that individuals with psychopathy, again, this upper region here, are actually showing more efficient organization of particularly dorsal attention networks. This was just resting state, but similarly we’re now replicating this to suggest that this efficient pattern, while on the surface might seem like a good idea, kinda too much of a good thing can be bad. And so it ends up being essentially hyper-efficiency that is then predicting the reduction in interference in some of the other types of task that I suggested. And while it, again, might be good to not be as distracted by other information, information outside of our goal can be very informative. It can cause us to stop and reflect. In Robert Durst’s case, if he had just taken a moment to think about the potential consequences of “What would happen if I steal a hoagie “in this high-security Wegmans, “and might that end me up in prison?” That is essentially hyper-efficiency, in the moment getting his goal, but leading to a long-term consequence. So a lot of the work that we had done up to this point really looks at basic processing and in some instances, just about simple resting state connectivity. And we were interested in seeing whether some of this dynamic of looking at what your goal is in the moment versus using that information for a future might translate into some other types of emotions, like regret, that are really important in the context of the criminal justice system. And so a more complex emotion like regret can be considered in actually two different ways based on decision and neuroscience. One way is commonly what researchers and clinicians will refer to as retrospective regret.

That is when you do something and you learn you could have done better or differently once you find out the outcome. And that is distinguished, again, conceptually and neurally from what we call prospective regret, which is the use of the experience of regret to then inform or change your decisions moving forward. And so to look at this particular type of, set of regret options, we ran a counterfactual task, which I will show briefly. Essentially, individuals just picked a circle and were told that randomly a ball will bounce around, allowing them to win or lose points. And then they were asked to rate how they felt about winning or losing points when they found out their outcome, rate how they felt when they found out what they could have gotten if they picked the other circle, and to do that repeatedly over time. So let’s see if this one works. So I’m selecting this circle. For some reason. . .

The video seems very strange, I’m not sure. It looked better in real life. [laughs] So I get 70 points. And I’m asked to do this repeatedly. So we’re able to actually measure a few things here. We can measure how people respond when they find out what they got. So basically in the instance where the circle shows that I got 70 points, how do I feel? And essentially, what we see is that individuals with psychopathy actually report more regret when they find out what they got and more pleasure when they find out that they got something positive, like 210 points instead of 70 points. Then we can also look at the regret score that happens when you find out what you could have gotten if you picked the other circle. So in this instance, I should be pretty pleased that I got 70 points ’cause I could have lost 70 points. And what we similarly see here in the red is that individuals higher on psychopathy are showing more emotional responses, in this instance regret, when they found out what they could have gotten if they picked the other circle.

And so kinda similar to the basic emotion threat processing information that I just described, when you ask individuals with psychopathy, “How are you feeling in response “to this information right here?”, they actually respond similarly, and in some instances even more intensely than individuals lower on psychopathy. But the question is, do they do anything with that experience? And that’s where we end up finding that individuals with psychopathy in the red here essentially show no use of prospective regret signals. So while in the moment they show retrospective regret, that doesn’t inform, then, future decision making on the next trials of the task. We were able to show that this particular type of phenomenon in individuals with psychopathy also predicted the number of incarcerations that they had. So again, if you continue to make decisions in the moment, you might know what you’re doing, but in the future it’s not being affected by that previous experience. So at this point, if someone’s brain tends to process information differently, if they’re actually seeing the world at a different speed, which would be a potential translation of the efficiency data, and they have a disconnect between what they did in a certain moment and what they’re doing, is it possible that actually in the moment, the time of the crime, the thing that the law cares about, that they actually don’t understand whether their behavior was appropriate or not, or that they actually have a disruption in a cognitive capacity that undermines them being of sound mind, which the other kind of legal phrase that we tend to think about? And so rather than being hyperemotional, they’re actually hyperrational. But it’s still an overexpression of a particular type of capacity that then undermines regulated behavior. And is it possible that showing these kind of fundamental, at-rest neural differences that occur both structurally and functionally is consistent with what the law would consider to be an extreme mental and emotional disturbance? And what we usually think about in the law is emotional disturbance that, again, is hyperemotional, that you were manic and you had so much energy and you had so much emotion that it led you to commit a crime. In this instance, we might actually be talking about something that’s the opposite, but equally detrimental. You didn’t experience enough of the emotion to be able to use that to affect your behavior.

So we’re kinda left at this point, certainly within translating science into the criminal justice system, with asking a question of whether we’re handling individuals with psychopathy appropriately. And that we might need to think about whether distinguishing kinda the knowing right from wrong that psychopathic individuals can often tell us they know what they did was wrong, how that might be different than what’s happening at any particular moment or how they’re processing information, while also in the context of understanding that these individuals are committing callous and heinous crimes at times, and so how do you most appropriately protect society, which is part of the goals of the criminal justice system. So in the next section I’m gonna talk about how we typically manage individuals with psychopathy in treatment and how some of our understanding of the underlying mechanisms of their behavior might inform new strategies. So as I’ve already mentioned, individuals with psychopathy are overrepresented in the criminal justice system, with about 25% of incarcerated individuals meeting a formal diagnosis of psychopathy. You tend to see that individuals with psychopathy are given longer sentences than other individuals, and that they disproportionately account for the days in segregation and the disciplinary infractions that they’re receiving while they’re being housed within a jail or prison. And so similar to some of the emotion and cognition work we’ve done, we were kind of left with the question of is it possible that how individuals with psychopathy are taking in information might affect how they are navigating the criminal justice system, and particularly during their incarceration? And so one of the phenomenon that you might be interested in is that do individuals with psychopathy have a capability of taking the perspective of other people? Can they understand that what they do might affect others, or that how they are being portrayed or they’re interacting with others affects some communication difficulties that they might be having? And so similar to the regret work because psychologists and neuroscientists like to make things complicated and parse things a lot, there are actually lots of different parts of theory of mind and perspective-taking that we can think about. And so one is called controlled theory of mind or controlled perspective-taking, and this is being asked to deliberately take the perspective of other people. And up until this point, this is where most of the work in psychopathy had been, it’s using kinda the classic Sally-Anne task where you see a narrative about this little girl who’s in a kitchen and she leaves the kitchen and her mother comes in and moves the cookies, and you’re asked “Does Sally-Anne know “where the cookies are?” And basically your job is to take Sally-Anne’s perspective. Is there any way she could have known that her mother moved the cookies when she left the kitchen? So it’s very deliberate, it’s very explicit. And what most of the work had shown at this point was that individuals with psychopathy can do this.

So it made it seem like, well, psychopathic individuals can take your perspective and they just don’t care, and so that’s why they’re doing all of the things that they’re doing that might con you out of your money or hurt you in some sort of way. But there’s this other part of perspective-taking called automatic perspective-taking, which is more effortlessly having the perspective of others affect your own. So for example, my job right now is actually not to take your perspective. My job is to deliver this talk in somewhat of a coherent fashion. But seeing people nod along or falling asleep or rolling their eyes might influence what I say next or what I do, and that more implicit process might then change my behavior. And we actually tend to see that this automatic tendency develops starting very early, including infancy, and begins to grow neurobiologically over time in different structures than what you might expect from a more controlled process, which typically we associate with the TPJ or other sort of kinda more general theory of mind regions or circuits. So we wanted to be able to test is it actually that individuals with psychopathy are totally capable of taking peoples’ perspectives, or is there something, again, more nuanced going on that might help us understand why it is that they seemingly ignore the thoughts and feelings of other people? And so to do this, we used an avatar task that had two conditions. One modeled a controlled type of setting where we told individuals that you are supposed to take his or he’s perspective. This was currently collected in a sample of all men who were incarcerated, so that’s why it’s always he. The avatar was wearing the exact same type of outfit or jumpsuit, whatever you wanna call it, as the people that we were testing, because there’s been a lot of work showing that theory of mind and perspective-taking can be influenced by in-group and out-group bias.

So you’re told take his perspective, and then you’re shown a number, two here, and you’re asked yes or no, can the avatar see two dots? And so the answer here would be yes, the avatar can see two dots. Then in the automatic condition, you are told to take your perspective, the number two, and the question yes or no, can you see two dots? And the correct answer here is actually no. So you can see three dots, but the avatar sees two dots here, unless the avatar had eyes in the back of his head, and this creates interference for people. Because the avatar sees two dots but I could see three, I tend to be a little slower. People are very accurate at this task, so it’s not like it translates into not being able to give the right answer, it just slows us down and causes us to stop and think a little bit more. So what we found, essentially, was that individuals with psychopathy were able to do the controlled processing. When told deliberately, “Take the perspective of the avatar,” they can do it. This replicates previous research. But when put in the automatic condition, “Take your perspective,” they essentially showed less interference from the avatar, which is this region right here. So while they have an ability to engage in perspective-taking, they have less of a tendency to automatically do so.

And this predicts their engagement in interpersonal crimes like assaults, but not crimes that don’t involve other people like a theft or some sort of destruction of property that’s not more face-to-face in terms of an interaction. And so what this might suggest from a management perspective is that correctional officers or other staff actually shouldn’t assume that individuals with psychopathy can see the environment in the same type of way that they do. Now, we could think about very nuanced communication strategies or changes in how you handle individuals that leverage our understanding of how they’re doing some basic process like perspective-taking. And so again the question becomes, are psychopathic individuals actually unmanageable, or is it just that we need to work with them differently? And so as of actually yesterday, we got approval and funding to start a project that actually trains COs, corrections officers, in communication strategies that kind of works off of some of this type of work of adjusting how they explain their decisions, particularly when they’re giving a disciplinary infraction to someone who’s currently incarcerated. And rather than asking something more general that requires some sort of automatic perspective-taking, like, “Do you know what you did?” Actually asking very deliberately, which we know psychopathic individuals can do, “What do you see from my perspective?” Or “What happened from my perspective?” So this would just be one example. It’s not like this is the whole training. But essentially, using a lot of this type of work to be able to think about alternatives for communication that would allow us to capitalize on what individuals with psychopathy can do and circumvent the things that they have difficulty doing. Similar to some of the management difficulties that we have with individuals with psychopathy, we also see that they are not as responsive to treatment as you would hope. So most of the treatment approaches in psychopathy tend to try and use what traditional therapies would be, contingency management, so rewarding more positive or pro-social behavior, cognitive behavioral therapy, asking individuals with psychopathy or others to think about their feelings, to think about their thoughts and challenge them. And what we generally see is that they’re minimally effective at best.

We tend to see that individuals with psychopathy drop out sooner if they have a choice, and that they’re generally noncompliant. There’s some work to suggest that if they do show positive gains, they never return to what we could consider kind of a normal baseline. So they might show some improvement, but it’s not going to return them to a level that we would consider more typical functioning. And there are a couple studies to suggest, actually, if you give psychopathic individuals the traditional treatments, they don’t get better. You actually make them better at being psychopaths. So they use all of the things like talking about emotions and challenging their thoughts and feelings to more effectively con and manipulate other people, which generally is not our goal of treatment, so is not a great outcome for the state of treatment. And so again, we’re kind of left with this question, are individuals with psychopathy actually untreatable or is it that they’re just not receiving the correct treatments? And a lot of what the traditional treatments require is some people to be, certainly, aware of and able to recognize their emotions to then be able to manage them. For individuals to be able to take other people’s perspectives or recognize how their behavior might affect others, and certainly for people to notice patterns in their behavior. But we know with psychopathy, all of those things are dysfunctional. They have difficulty identifying emotions, they have difficulty, certainly, automatically seeing how their behavior might affect others, or incorporating another’s perspective.

And they certainly have difficulty connecting all the dots of their life, essentially. It’s almost like each moment they come upon is essentially brand new. And so we’re asking them to engage in treatments that really are not going to be likely effective if we just take a moment and think about the neurobiological difficulties that they’re having. So what we wanted to do was try and say, “Okay, well, we have two things. “We know that individuals with psychopathy “are probably disproportionately put in treatment “when they’re in the criminal justice system, “’cause they are seen as severe and difficult to manage, “and so they get referred frequently “to treatment programs. “But we also don’t have the most effective treatments. “But what we do have is decades of amazing research “that allow us to understand what’s going on “with how they think and feel emotions. “And so can we kinda combine those two?” So we developed a computerized treatment to be able to try and train and target the neurobiological and cognitive-affective deficits that we see in psychopathy. So I’m just gonna show you the psychopathy arm of this training. Essentially, all of these tasks required that you learn to integrate and notice contextual cues.

So the one task, for example, that was given on the computer is this reversal learning task, where you see an elephant and you see a giraffe, you start winning points by selecting the elephant, and then all of a sudden, you start losing points when you select the elephant and you actually need to switch to select the giraffe ’cause that’s the one that’s gonna win you points. And usually, psychopathic individuals are not good at this task. So there’s a lot of work from James Blair’s group that suggests they have reversal learning deficits. And what it comes down to is that they’re really having difficulty tracking the contextual information. They’re just focused on whatever the goal is at the moment and not noticing the switch in the contingency. In the next task, we had people have to respond to whether the string in the middle was letters, numbers, or both, but they pressed one button if it was a green box and another it was a yellow box. So to do this task well, individuals with psychopathy have to attend to both the goal, the thing in the middle, and the contextual cue, the color of the box. Again, there had been work prior to this stage to suggest that psychopathic individuals do pretty poorly on this task because they have difficultly integrating the two pieces of information. Then in the last game that we had them play, we showed them emotion faces and then eyes looking left or right. And on most of the trials, you only had to tell us are the eyes looking left or right? So if their eyes are looking left, you press the left button, and if the eyes are looking right, you press the right button.

But each day you came to play this task, you were told about your target emotion. So today’s target emotion is fear. If the face is showing fear, you actually press the button opposite of the eye gaze. So if the eyes are looking left, you really press the right button, and if the eyes are looking right, you really press the left button. But only for the fear faces, not for any other. So to do this task well, you have to notice both the eye gaze and the emotion on the face. In a previous study, we had shown, again, psychopathic individuals don’t do very well on this task because they had a lot of difficulty, certainly quickly, integrating the emotion on the face and what their goal is, which is really just focus on the eyes. So we had people come in, this was collected in individuals who were referred for treatment who were currently incarcerated but hadn’t received it yet. All of the participants had a substance use disorder diagnosis, and so they were generally being referred for treatment related to substance use problems. We were able to do a clinical interview and evaluate them for psychopathy, and then give this training, which we call the training to attend to contextual information, once a week for six weeks.

They completed a series of tasks before the training, the training took an hour, and then they came back after the six weeks and completed the same tasks again. And then some of the individuals with psychopathy were randomized into another treatment that had nothing to do with contextual processing. So it was just how do you manage your emotions and not overreact to them? All computer-based, all pretty simple. No real connection or other therapeutic discussion. They were shown graphs at the end of each day of their progress, which actually tends to be very motivating when you’re not doing very well. Most people wanna try and improve and earn more points. When you are doing well, you wanna keep up with it, and when you’re kind of being stagnant, you’re told it sometimes takes a lot of effort to learn something new. So really, the only motivational feedback were these graphs that were shown at the end. So, first looking at the training tasks, again, six weeks once a week for an hour, we see that individuals with psychopathy in the white bar here, who received the attention to context training, showed significant improvement over the six weeks. But if they received the training that was not meant for them, the affective cognitive control one, they did not improve.

And this was interesting, but it really could’ve just been a practice effect because individuals with psychopathy started out so poorly on all of the tasks. When you kind of suck at something, the only way to go is up. And so while this was exciting, this wasn’t like a, “I could retire” moment. So what we did was look at the pre-post task, the tasks we gave right before they started the training and then that they completed right after. And these tasks included collecting startle potentiation, which might be an indirect measure of amygdala responsivity, different ERP or event-related potential responses that were markers of how well they attended to certain information, and other tasks that other researchers had shown might be deficient in individuals with psychopathy. And we essentially show, again, that the individuals with psychopathy who received the attention to context training showed small but still significant improvement pre and post. So while before they might have shown a deficit in startle potentiation in response to threat, they now no longer were showing that deficit. But only if they received the deficit-matched training, not if they received the one about controlling their emotions. And again, this is really interesting. The issue is these are all laboratory-based tasks, and while we are showing change, what does that mean for real behavior? So the other thing that we were able to do because everyone was currently incarcerated was collect information about disciplinary reports that they might have gotten while incarcerated.

And so while you’re in prison, you could get a disciplinary infraction or a ticket for a variety of things. It could be for not having your shirt tucked in, it could be for standing out of line or not showing up, or it could be for like shanking and stabbing someone. So we have a number of infractions that we can examine, and we started by looking at the severity of infraction because whether you get a ticket or not in prison is somewhat dependent on what unit you’re on and what correction officer you’re dealing with. But the level or the severity is a little bit harder to personalize. It’s supposed to be more objective, supposed to be. So what this is showing you is an absolute value of improvement two months post the training on the severity of conduct reports, and what we see is that individuals who got the attention to context training who had psychopathy showed less severe, so more positive means less severe, conduct reports two months after receiving this training. And again, if they got the one that was more about just generally controlling emotions, they really didn’t show any change. So what we’re able to see here, really, is that individuals with psychopathy are capable of processing contextual cues like emotion, like punishment, like interference, which is actually hard to change. We basically wanted to make something they’re really good at a little bit worse so that they could behave in a way that might be more controlled and considerate of information. And that we’re able to leverage our understanding of the basic cognitive-affective mechanisms to develop treatments that more directly target the types of deficits they have.

Now I wanna be clear, I don’t think giving every incarcerated individual an iPad is going to solve our problems. I think what this type of treatment likely might do is allow for some practice and development, neurobiologically, of individuals with psychopathy to kinda normalize how they process information, or certainly build compensatory strategies for taking in information in a more typical way, and then refer them to the traditional treatments, which we actually know in the criminal justice system work. Like cognitive behavioral therapy, contingency management, if given the way that they should be given, work for individuals who are incarcerated, just not individuals with psychopathy. And so we might need to give them a little bit of pretreatment before we put them in the typical types of treatments. And then ultimately if we could do this, it really raises questions about the appropriate facility for individuals with psychopathy, and this is something that Europe does much better than the United States, where actually more frequently, individuals with psychopathy are put in a forensic mental health center, not in a traditional jail or prison that we might examine here. So overall, what I’ve mentioned to you is that psychopathy is certainly disproportionally represented in the criminal justice system, and this one syndrome accounts for a lot of the cost of crime because of their repeated acts, repeated incarcerations, and referrals to treatment, even though the treatment is not effective for these individuals. And that we have seen a lot of work, again, from our lab and others that suggests that these individuals are not seeing or interacting with the world in the same way that other people are. But that they can experience emotion and they can take other people’s perspectives, and by having this more nuanced understanding, we might be able to actually consider different ways for communicating or treating these individuals. And that by bringing all of that work together, we could understand that their behavior is changeable and is fact is not something that we should just give up on and not consider as a disorder that is worth effort and worth consideration for improvement. So I wanna thank you all for listening, and I also wanna thank Mike, certainly, for inviting me here.

It’s so fun to be back in Madison. For my lab, who are amazing and do a lot of this work. For Wisconsin, who started my work in corrections, and then Connecticut who has allowed us to build a lab inside of the prisons there. And then all of the funding agencies who support this work. So thank you. [audience applauding]