'Patients with the disorder deserve to be helped'

Chief of the Unit on Affective Cognitive Neuroscience at the National Institute of Mental Health in the US – that's quite a job! How did it come about?

In around 2001 the NIMH decided to develop a programme focused on understanding mood and anxiety disorders with scientists working at the molecular, systems neuroscience and behavioural levels. I was fortunate enough to be hired by Dr Dennis Charney, its then director, as an investigator within that programme.

I note that a lot of your research concerns the amygdala. That pesky amygdala – so much to answer for!

There appear to be three main systems that show dysfunction in individuals with psychopathy/callous-unemotional traits: the amygdala, ventromedial prefrontal cortex and the caudate.

The amygdala is important for learning about the goodness and badness of objects and this capacity definitely appears to be disrupted in individuals with psychopathy/callous-unemotional traits. One of the ways that we learn about actions is through the distress of others – their sadness and fear. The processing of both these emotions appears to rely on the amygdala and is disrupted in individuals with psychopathy/callous-unemotional traits.

Ventromedial prefrontal cortex is important for representing the value of objects and actions. It allows us to make good decisions – to decide what would be the best thing to do given the probabilities of the possible rewards and punishments available. The function of this system also appears to be disrupted in individuals with psychopathy/callous-unemotional traits.

The caudate is important for detecting prediction errors – times when you were expecting to gain a reward and actually didn't or times when you gained an unexpected reward. Detected prediction errors lead to fast learning – the brain knows that there's something important out there to learn about with respect to reward/ punishment. Individuals with psychopathy/callous-unemotional traits have difficulties here also.

Notably, the amygdala, ventromedial prefrontal cortex and the caudate are highly interconnected.

I attended an event recently which discussed your work in a school context, and I was shocked by the number of 'callous-unemotional' children there apparently are out there – some as young as four.

I'm not sure what numbers you heard but I would not expect the proportion of youth to be that high. There have been estimates of conduct disorder of 3–5 per cent of the male population, and it appears that those with significant callous-unemotional traits are only around 30–40 per cent of that population so the diagnostic rate might be around 1–2 per cent in males. Of course, these are guestimates so must be taken with a degree of caution.

If we say that such children have 'psychopathic tendencies', does this give a false sense of harmlessness? At what age should we simply say they are psychopathic?

That is a tricky one. We don't work with anyone younger than the age of 10 years. However, this is primarily because of the difficult of obtaining good fMRI data with children younger than 10 – they tend to find staying still for the requisite length of time very difficult. I think it would be difficult to reliably classify anyone at four years old though. Certainly, there are currently no longitudinal data to suggest such a classification would be reliable.

When we first met, I was looking at childhood conduct problems from quite a social perspective, and I'm sure there was more room for that in your theorising than there appears to be now. Are you solely focused on the neurological now in terms of contributory factors?

I focus on neuro-biological factors because I believe that is where most of the greatest amount of traction with respect to understanding this disorder is available currently. But that's different from my theorising. I am relatively confident that how callous-unemotional traits present, whether they are associated with significant antisocial behaviour and what forms of antisocial behaviour, is primarily dependent on social factors.

At the very least, an affluent youth is less likely to mug someone than a less affluent youth, irrespective of their callous-unemotional traits, because they already have financial resources available. To what extent social factors contribute to the underlying callous-unemotional traits is, however, currently somewhat less clear.

So given your focus, do you still consider yourself a psychologist?

Probably. Over the last year, I've certainly been introduced as a psychologist but I've also been introduced as a neuroscientist and sometimes as a cognitive neuroscientist – that's the term I'm probably most comfortable with.

Is it true that psychopaths don't learn when punishment is used?

Not exactly. First off, the brain does not learn from punishment in a single way. We use punishment to learn that some things are bad and to be avoided. We also use punishment to stop doing whatever we've just been doing because it got us hurt. Individuals with psychopathy/callous-unemotional traits have difficulties learning the value of objects and actions – learning that some objects and actions are bad and should be avoided (and also having difficulties learning that some objects and actions are good and should be approached). But it is not that they are unable to perform this learning, just that they are poorer at it than typically developing individuals. However, with respect to using punishment to stop doing actions immediately and try something else, this use of punishment information appears to be intact in individuals with psychopathy/callous-unemotional traits.

Why doesn't such a failure result in widespread learning difficulties?

Being poorer at learning the value, the goodness/badness, of objects and actions definitely leads to poorer decision making and difficulties in socialisation. However, it should not interfere with episodic or semantic memory, and it will not interfere with general intellectual ability. Moreover, executive functions that are not emotion-based are also intact. In short, individuals with psychopathy/callous-unemotional traits have many strengths despite their difficulties with learning about reward and punishment values.

Previous research has suggested that perspective-taking skills are amongst those strengths. If psychopaths have good perspective-taking skills, shouldn't they also subscribe to fairness?

I'm not sure that I agree with your logic here. I'm going to leave aside fairness because I'm not sure we're really sure that there is a good computational story as to what fairness might be. But take something like the ultimatum game. Two players interact to decide how to divide a sum of money that is given to them. The first player proposes how to divide the sum between the two players, and the second player can either accept or reject this proposal. If the second player rejects, neither player receives anything. If the second player accepts, the money is split according to the proposal. Performance on the ultimatum game is known to require the role of ventromedial frontal cortex in the representation of value information. This is one of the functional impairments shown by individuals with psychopathy/callous unemotional traits. Recent work by Joe Newman and his collaborators have shown that individuals with psychopathy do show impairment on the this game.

Despite this, or perhaps because of it, we still tend to hear a lot about so-called 'white collar' psychopathy. In his book The Psychopath Test, writer Jon Ronson claims that psychologists have described psychopathy as 'the madness that makes the world go round'. Are there really significant numbers of psychopaths oiling the wheels of industry and government?

I can't comment on the numbers, it's not work that I do. Paul Babiak would be the person to ask. But I certainly consider it to be a theoretical possibility that some individuals with psychopathy achieve positions of importance. Given the difficulties in decision making seen in this population, though, the inadequate representation of the rewards and punishments of decisions, it is unlikely that they are successful in the long term.

Where do you stand on the Skeem and Cooke view, that 'the process of understanding psychopathy must be separated from the enterprise of predicting violence'?

I think I agree with them. Certainly, my work is totally focused on understanding the bases of this emotional disorder so that we can provide effective treatments.

And what's your view on current treatment regimes for psychopathy – are they any more humane than the Inuit approach to a 'kunlangeta'?

I certainly believe that current psychotherapeutic practices are more humane that pushing someone off on an ice floe!

What are your hopes for psychopathy in DSM-5?

I would like to see a qualifier of the presence or absence of callous-unemotional for diagnoses of conduct disorder and antisocial personality disorder, as I hope this would be useful.

I'm sure many psychologists will have had a go at scoring themselves on the PCL-R. It seems rather easy to self-diagnose psychopathic traits!

It's true that the terms used suggest that it is rather easier to score higher than it actually is. At least, when the PCL-R is administered properly. Sure, you may have engaged in a little lying here and there, but pathologically? I don't think so. Shallow affect? Assuming that you're still in touch with your parents/siblings and you have some form of significant other, you should score a 0. Many of the other items – for example parasitic lifestyle, impulsivity, irresponsibility, juvenile delinquency, early behaviour problems, revocation of conditional release – are all typically scored on the basis of a real history of antisocial behaviour. So assuming you don't have a secret criminal life and you're not as frisky as a jackrabbit in spring, it's actually pretty hard to get anything other than a low score.

These are thoroughly unsympathetic characters and treatment – of the core deficits at least – seems pretty ineffective. So can you still find it a rewarding area to work in?

The current lack of very effective treatment means that there is something very important to strive for. I think it's this above everything that makes this area a rewarding area to work in. If we can help youth and adults with the disorder, we can really do some good.

Since you've had children, don't you find it harder to work in such an area? I can't even watch the news any more.

Again, I think patients with the disorder deserve to be helped. My belief is that psychopathy/callous-unemotional traits is an emotional disorder that we will be able to help in the same way that we can help patients with other emotional disorders – not with the same treatments but with the same approach to developing treatments.

Can you imagine returning to the UK?

I think it highly unlikely. I could imagine a return to Europe perhaps though.

Two of your mentors, John Morton and Uta Frith, are simply the most generous and bright I have come across in my own career. It must have been amazing to have them as mentors in yours.

I was extraordinarily lucky. Both were incredible. I would probably not have stayed in their business if I had lacked their mentorship. I certainly would have achieved much less.

• James Blair is Chief of the Unit on Affective Cognitive Neuroscience at the National Institute of Mental Health in the USA