Thursday, September 9, 2004
Session 2: Neuroscience, Brain, and Behavior VII:
Understanding Aggressive Behavior Through Neuroscience
Emil Coccaro, M.D, Director, Clinical Neuroscience and Psychopharmacology
Research Unit, Department of Psychiatry, The University of Chicago
CHAIRMAN KASS: We turn from a session on neuroscience
and the law to a session on understanding aggressive behavior
Without further ado, Dr. Coccaro. Thank you. Very much.
DR. COCCARO: Thank you, Dr. Kass. It's an honor
to be here.
It was interesting sitting in this morning's session, because
I agreed with almost everything Dr. Morse was saying. And I'm
going to present a fair amount of data about the features and
neurobiology of aggressions. Also, some data about treating impulsive
But I really come from the point of view of trying to understand
these behaviors biologically and phenomenologically, in a way
so we can come up with strategies for intervention. Hopefully,
those interventions would be voluntary interventions, not, you
know, state-mandated interventions.
So let me start first by disclosing that I have been a consultant
for a number of drug companies, have had research grants and other
kinds of remunerations. This is just sort of a general disclosure
which people in my field have decided we should be making at the
beginning of every presentation.
I'm not sure that many people here have connections to drug
companies, but it is important to let you know that. Most of
the work that I have done has, however, been funded by either
the VA Merit Review System or the National Institutes of Mental
Health. That's where most of my research funding comes from.
So in order to start this, I think we really need to talk about
what aggression is, because it's complicated, and this is
one definition of aggression — I will try not to spear anybody
with this laser pointer. But it's behavior by one individual
directed at another person or object, in which either verbal force
or physical force is used to injure, coerce, or express anger.
And, of course, this is, as I said, very, very broad.
But it's important to realize that aggression is part of
a complex triad of behavior, emotion, and cognition. So aggression
is the behavior, anger is the emotion, and hostility is the cognition.
These things all interact in ways to lead to the aggressive event.
And the order of them is up for grabs. It may very well be
that emotion is what starts this off. That then leads to hostility
and then to the behavior, but it could be the reverse. It could
be that people have hostile attributions about what other people
are doing. That then makes them angry and makes them act in an
Now, there are different types of aggression, and there's
various ways you can split the pie. This is one way that I do.
One would be socially-sanctioned aggression, and, of course, this
is — anybody know who this is? This is Tom Hanks playing
in Saving Private Ryan, and this is an individual who obviously
had to be aggressive in the context of World War II, but was not
an aggressive person in general. And he had to be aggressive;
that was his job.
Another way you can get aggressive is have a medical problem,
a real medical problem. And this is King George. If you're
wondering why I'm using movie posters it's because it's
not good to use real people in these slide presentations. So
he had porphyria episodes of severe agitation and screaming and
shouting and slamming doors and those kinds of things. But it
was really due to the porphyrian. And when those episodes abated,
the aggressiveness abated.
Now we get to the more interesting types of aggression, and
we have premeditated aggression. I think the classic example
would be a psychopath. This is Robert DeNiro playing a character,
a very hardened criminal, in a movie called Cape Fear. And definitely
a very premeditative individual.
I'm not going to speak that much about premeditated aggression
in terms of the biology, because we know very little about that.
But this is impulsive aggression. Anybody know who this is?
Bobby Knight. And he could be your neighbor, actually. And this
is the kind of people that I study. I don't study Bobby Knight,
although when I was in — when I got to Chicago he still
was in Indiana. And, unfortunately, he didn't — Indiana
University didn't call me up as a consultant on his case;
they just fired him. So I might have helped him.
Now, what about the types of aggression. Premeditated aggression
is goal-directed aggression. So when it comes to the issue of
culpability, there's no question here. I mean, somebody is
making a decision to do something aggressive. It gets a little
trickier when you talk about impulsive aggression. But even I
believe that impulsive aggression, which is threat-induced or
frustration-induced, there will still is that split second time
to form the cognition of whether this is right or wrong.
So I do not believe in general terms that impulsive aggression
is something that really forgives you of the responsibility of
your acts. What I'm interested in is: why do people act
this way? And what can we do to make them not act this way, or
decrease the frequency at which they act this way?
Well, there's another way you can break this down. It's
just primary aggression and secondary aggression. And I talked
about that a little bit before, but here's just a slide to
say what could be secondary aggression. Well, as in the King
George example, you have systemic and metabolic disorders. You
can also have primary diseases of the brain.
Anything that affects the brain can affect — can increase
irritability, increase the risk of aggressive behavior, and several
psychiatric disorders can do that. However, what needs to be
remembered is that while aggression can run across various psychiatric
disorders, and a lot of people in my field, you know, want to
think aggression is the same in every disorder, you can't
really do it. There's a real caution here, and that is the
phenotype of aggression may not have the same pathophysiology
And the reason I feel very strongly about this is that even
if you were to say that there's a relationship between some
biological parameter and aggressiveness, the brains of individuals
who are schizophrenic, manic depressive, drug — whatever,
are different from each other, and the milieu is different.
And you can have a situation where, let's say in mood disorders,
depression for example, where people can make suicide attempts
and be aggressive, their brains are different. There's something
different about their brains. It might be that their catecholamine
system isn't working so well. And if that system doesn't
work well, all bets are off in terms of how other systems interact
So this is an important caveat. And while I'm a dimensionalist
in terms of looking dimensionally at aggressive behavior, rather
than necessarily looking at it from, you know, disease entities,
you have to be cognizant of the fact that these categories, you
know, do have relevance.
So if you want to talk about primary aggression, that is, to
our best way of thinking, what you could call intermittent explosive
disorder. So let me define what that is. In DSM-IV, this is
what the criteria — they're not very good criteria,
but this is what they are.
Several discrete episodes of failure to resist aggressive impulses
that result in serious assaultive acts, destruction of property
— one. Two, the degree of aggressiveness is grossly out
of proportion to precipitating psychosocial stressors. And, three,
it's not better accounted for by something else, some other
mental disorder, some physiological effects of substance or a
general medical condition. But the problems are really substantial.
What constitutes a serious assaultive act or destruction of
property? DSM doesn't do such a good job on that. How many
aggressive acts do you need to have to make the disorder? What
timeframe? And what's the nature of the aggressive act?
Which I think it's really critical.
And we really focus on impulsive aggression, even in IED, because:
a) from my point of view, the premeditated aggression really is
more relevant to criminal justice rather than to mental health
systems. Explosivity really suggests impulsivity, and the impulsive
forms of aggression are the ones that have clear psychobiological
That doesn't mean there isn't a psychobiology to premeditated
aggression, but, rather, the tendency to be premeditatively aggressive.
But it's really the impulsive forms that really have most
of the science behind them at this point.
We had to develop research criteria for IED to really bring
this into line with the biological work that I'll be telling
you about in a few minutes. And what we've done is we've
said, "Look, you know, the way DSM sort of defines it, you
have to be almost like Bruce Banner, who gets" — that's
the alter ego of the Incredible Hulk — "who, you know,
gets angry and he becomes this horrible monster."
That's not what most aggressive people are like. Most aggressive
people don't have huge outbursts and then in between they're
perfectly fine. Most people have big outbursts and little outbursts,
and you need to be able to account for that. So we allow for
frequent or low intensity aggressive acts, as well as infrequent
but high intensity aggressive acts.
We also require those aggressive acts to be impulsive in nature,
which means not a whole lot of forethought going on. That doesn't
mean they can't commit any premeditated acts of aggression,
but most of the acts of aggression need to be impulsive. And,
critically, there must be some distress or impairment due to the
aggressive acts. Otherwise, it's not a disorder, especially
if you allow for these low frequency attacks, into the diagnostic
Am I speaking not loudly enough or too fast? Oh, okay. Okay.
I guess I was making the wrong attribution of your face.
Okay. I'm going to skip through the research criteria,
but that basically incorporates those ideas. And you might ask
how much IED is out there. How much primary aggression is out
there? Because nobody really knows, and you can't actually
determine it until you have these kinds of criteria.
And, fortunately, I've been involved in some studies that
have looked at this. Is something wrong? Too much feedback?
I did a small study that just got published indicating that four
percent lifetime have this problem. And two unpublished studies
that will be published sometime in the next year or two, both
large studies, showed the same exact number.
Four percent lifetime have this problem of being impulsively
aggressive and getting into trouble because of it, not necessarily
criminal aggression, but aggression, nevertheless, that's
certainly disruptive to their lives and the people around them.
And the age of onset is early. It begins in the pre-teen years.
It peaks in the teen years, and then drops over time. And the
age of onset date is interesting, because a lot of the times this
problem occurs before the onset of other disorders that people
think probably are related to the aggression.
Now let's get to the meat of the matter. What are the underpinnings
of aggression or impulsive aggression as we know them? Well,
you can look at it from a familial and a genetic environmental
and biological perspective, and that's what I'm now going
So if you're looking at the familiarity of IED, you're
looking at how does it run in families. And this is a study that
we did a few years ago, which we are in the process of writing
up. And we looked at people who met research criteria for IED,
and we looked at people who were controlled — did not have
And we found that more than 25 percent of the relatives, first-degree
relatives, had this problem. Whereas you only had about eight
percent of the controls. We know there's lots of co-morbidity,
so we wanted to see, did having histories of other kinds of problems
in these IED patients relate to the family risk of aggressiveness.
And it actually didn't.
So individuals who had IED, if they had a history of suicide
attempt or mood disorder or alcohol or drug, did that change the
family risk? It did not. And having other kinds of co-morbidities
in the relatives did not. So what it meant was that if the relative
had IED, the chance that they had some other disorder was something
in the range of 60 to 70 percent, and that was pretty much what
it was if they didn't have IED.
So what this really suggested was this is a signal that runs
in families, not due to other kinds of things. That doesn't
mean other things don't interact, but there really is a signal
Now, what family studies tell you — this actually was
a family history study, meaning we asked relatives about other
relatives rather than interview the relatives themselves. We're
currently doing a family study with NIH support to actually interview
all the relatives that we can. It gets to genetics but not completely,
because what runs in families could be genetic and could be simply
cultural or simply environmental.
If you want to do genetics, you've got to get twin studies,
and we have done some twin studies, which we are continuing to
do. This is a little twin study we did looking at the Buss-Durkee
aggression scale. We did this because the twin sample was a male
And irritability on this scale is the tendency to blow up.
Verbal assault is the tendency to scream and shout. Indirect
assault is a tendency to slam doors and throw things. And direct
assault is they hit people. And what's important to point
out here is not that environment is more important than genetics,
but that genetics has a role to play here.
For irritability, it's almost 40 percent But a nice point
is that as you go from the least severe forms of aggression to
the more severe forms of aggression, the genetic influence increases
over time. So there's more environmental factors going on
in screaming and shouting and less in indirect assault, and about
equal maybe in actually physically hitting people. Again, what
we're talking about is ultimately going to be an interaction
between the two.
Now, what's interesting what the Buss-Durkee scales is they
are a tendency to behave a certain way. There's another assessment
of aggression called the life history of aggression, where you
actually ask people how many times they've been aggressive.
The heritability of how many times they've been aggressive
is actually much lower than the heritability of the tendency to
And the reason for that is really the interaction between the
tendency to be aggressive and actually have things in your environment
provoke you to be aggressive. So that's sort of a fine point
on the genetics of aggression, which is critical I think to raise
Now, what about environmental factors, because, you see, these
— these greenish bars here account for most of the variance.
What's going on with environment? If you look at all the
studies, the big things that come out are experience of aggression
as a child, meaning being aggressed upon, witnessing aggression
as a child, and parental dysfunction.
And there are a lot of people in the field who feel very strongly
that the affective environment, which would be the aggression
against childhood, leading to aggressive behavior, is actually
mediated by a series of processes called social information processing.
And this has to do with assessing what's going on in the environment,
in the immediate environment.
And the big things that seem to be coming down the pike is that
there's a reduced encoding of relevant social information,
meaning that if you ask people what they're picking up from
the interaction they will not give you as — if they're
aggressive people, they won't give you as many cues or many
clues as to what's happening, and they will also tend to be
— have an increased tendency for hostile attribution.
The way these kinds of studies are done is they present kids
with these socially ambiguous kinds of vignettes where it's
not clear that somebody did something to the person — the
other person in the vignette on purpose or not, and they ask the
kid, "Why did this person, Johnny, you know, hit this other
kid in the back with a ball?" And it's ambiguous.
And the kids who have the impulsive aggressive problems, who
have the history of aggression in childhood, tend to be the ones
saying, "He was trying to hurt this other kid." So
there's this hostile attribution problem.
Now, one way you can assess hostile attribution in the laboratory
— and we can do it in two ways, one with sort of the laboratory
measure as well as a paper and pencil measure, which I'll
tell you more about in a minute, is you can look at, let's
say, a neutral face in the lab and ask people — they say,
"What emotion is on this face?" We don't tell them
that they can pick neutral. They have to pick something. They
have to pick one of these faces.
And what happens is, when you look at these IEDs and you contrast
it or look at the history of trauma, childhood trauma they have,
there is a positive relationship between the number of times they'll
call a neutral face angry and the amount of childhood trauma they
have. And this is one of the first pieces of data, and it's
not a lot of data — pieces of data to suggest that this
finding that has been reported in kids is also true in adults.
We've done some more work with this, and we developed direct
aggression vignettes, for example, for adults, which has not been
done before. And this is an example of one of those vignettes.
Imagine that you're in a karate class competition, and you
have to demonstrate your abilities to your instructor.
You're matched up to fight with someone in the class who
you do not know well. And while you're being evaluated, your
karate classmate hits you in a way other than you were taught
and you are hurt. That's pretty ambiguous. You don't
really know, you know, what that means. And then when ask people
— and, for example, we might show them this sort of, you
know, physical — this sort of illustration of what happens.
You don't really know what's going on. A person has to
then tell you, you know, what really went on.
And we can ask people, well, what did they do? How likely is
it that the classmate wanted to physically hurt them or make them
look bad? And that's hostile. How much did the classmate
want to win the match? And whether they hurt them or not —
and did he do it by accident?
What we find in the IEDs, these impulsive aggressive folks,
not surprisingly is that they're more likely to say that,
"The person tried to hurt me or make me look bad."
A little less likely, but more likely than normal is to say, "They're
just trying to win, and less likely benign."
So these folks have a hostile attributional bias that is —
in adulthood that we — that other investigators have seen
in children as well. We'll get back to more the neuroscience
of this or how we're trying to approach that in a few minutes.
Now, are these environmental effects mediated by biogenetic
factors? And the answer to that is probably so, because we really
live in a world of environmental gene interactions. So it's
not a matter of determinism; it's a matter of probabilistic
kinds of things.
Here's a study that was published in Science by Caspi, a
very exciting study, where they looked at the MAOA genotype in
— these actually weren't male children at the time they
got the genotype. It was later on in early adulthood.
But what they had way back when when these individuals were
kids was they had this assessment of whether they had been maltreated,
probably maltreated, or severely maltreated, and then they got
the genotype data, and then they correlated it with their disposition
toward violence, which is what's on the left, and whether
they were convicted for a violent offense.
And what you see here is this. You have low MAOA activity in
both of these graphs on the left, and what this really meant was
that these people had a genotype that prevented this enzyme in
the brain from breaking down norepinephrine, serotonin, and dopamine,
and on the other side, other people who had high MAOA, meaning
that they were able to break these neurotransmitters down appropriately.
You really want to be in this category here. You want to have,
you know, good functioning genes. And when you look at the people
who have good functioning genes, you have this sort of linear
relationship between aggressiveness and maltreatment, as we might
expect, and it's not very pronounced. You really see it in
the folks with the low monoamine oxidase enzyme activity. So
there is this gene environment interaction that is extremely important,
and more studies will come out to show this same kind of thing.
Now, what about specific neurobiological factors? The big neurobiology
factor where most of the data comes from is dysfunction of the
central serotonin system. And as I was saying to some people
before, I started out as a serotonin kind of guy, and I've
gotten a little more Catholic since, because this is all not serotonin,
which, of course, was heresy back in the late 1980s.
But there's also a dysfunction of other central systems,
and there are also issues about social and emotional information
Now, why is serotonin important? Serotonin is important because
those neurons, which originate in the nuclei in the membrane,
project to diverse areas of the brain, including the cortex, hypothalamus,
all over the place. They even go down the spinal cord.
And so you have a system that casts a very wide net of influence
over the brain, number one. Number two, those neurons fire in
a very consistent fashion, a very tonic fashion. They pretty
much fire at the rate they fire all the time. And it's very
hard to get them to change their rate. If you put an electrode
in a cat brain and you looked at the firing of a serotonin neuron
and paraded a mouse in front of the cat, serotonin neurons wouldn't
do anything. If you put that in the norepinephrine part of the
brain, it would fire like crazy, because norepinephrine responds
But serotonin really are like the breaks. They're the modulating
influence in the brain, and it kind of makes sense that if they
— if it doesn't work well, if it's dysfunctional
in some way, it's going to release the breaks. People are
going to have bad breaks, and that seems to be what's going
In the late 1970s, an interesting study from the NIMH came out
looking at the correlation between 5-HIAA, which is the major
metabolite or breakdown product of serotonin in the brain, and
a life history of aggressive behavior.
This study was interesting because this is not what they went
out to find, and the story is kind of funny because what you had
was Brown being a junior investigator at the NIH was asked by
his bosses to go out and get controls. And so he went to the
Naval Hospital and recruited people who he thought were controls.
They were people who weren't drug abusers, they weren't
schizophrenic, they weren't bipolar. They weren't anything.
And he figured he had normals. And the data comes back and his
boss goes, "What's up with this data? These values are
all over the place." And what they discovered was these
really weren't normals. They were severely personality disordered,
and one of the big dimensions was aggressiveness.
And so they went back and got a way to measure aggressiveness,
and sure enough they got this very nice inverse relationship between
this measure of serotonin and a history of aggressive behavior.
And this is a history of a behavior, not the tendency to behave,
but the actual history of acting aggressively.
They also got the relationship with suicidality, although that
had been shown before. And they postulated that there was a trivariate
relationship. So if serotonin was the coin — was the coin,
then this side is serotonin, is aggressive behavior directed outward,
and this side is aggressive behavior directed inward, which would
be suicidal behavior.
What happened right after was an important study coming from
Finland showing that it was impulsive aggressive individuals that
had the low 5-HIAA finding, not the non-impulsive aggressives.
And this has been replicated in a number of other stories.
At the same time, work was looking at specific receptors in
the brain, and violent suicide victims, finding that certain receptors
were down and certain receptors were up. Specifically, serotonin
transporter receptors were down, and serotonin-2 receptors over
here were up. And the suggestion was this system doesn't
work so well, so this system goes to compensate.
Well, there are problems actually with that interpretation,
but ultimately you don't really know what the final result
is, because if you've got pure compensation nothing really
changes. And that's when I entered the scene. I started
to do work looking at pharmacologic challenges, and what we did
was we gave agents that specifically simulated the serotonin system
in a physiologic fashion.
You could look at a variety of outcome measures. We looked
at proactive response, because it was the best measure to use.
And when you gave this drug, you had a nice proactive response.
If you gave a placebo, you got nothing. And how much this went
up was an index of how much you activated that system. And what
we found was a very nice correlation between how much you activated
the serotonin system and how aggressive, impulsive, and irritable
these people tended to be.
These are personality disordered individuals — males,
in particular. And this was assaultiveness and irritability.
And this kind of finding can also be seen in the brain. This
is work done by Larry Siever, who was my mentor at the time, who
has gone on and done some other work in this area, looking at
a PET scan, giving this same drug — fenfluramine —
and seeing where in the brain is there differential activation?
And it's really this frontal part of the brain, as I'll
talk about later, that seems to be differentially subactivated
in these impulsive aggressive kinds of individuals.
Now, what's interesting about the serotonin system, it's
all over the place and the findings are all over the place. What
I mean is you can find them anywhere. Here's a study looking
at platelets. This had nothing to do with the brain, but the
platelets are a model for what goes on in the brain.
And, in fact, if you look at the serotonin transporter on the
platelets, you will see the same inverse relationship between
a measured serotonin function and aggressiveness. Almost anything
you look at with the serotonin system you will get this finding.
Now, okay, so serotonin is out of whack. These are all correlational
studies. In animal studies, we can do things to manipulate serotonin
and see if these things are true. Can we do this in humans?
Yes, kind of we can do this in humans. This is a study by a friend
of mine, Don Dougherty, who took individual — relatively
healthy individuals, although some of them were more aggressive
than others, and gave them a thing called a tryptophan depletion
Tryptophan depletion is an interesting kind of paradigm, because
what happens is you give people these amino acids, without tryptophan,
their livers go, "Ooh, amino acids, it's time for me
to make protein." So they go to make protein, but there's
no tryptophan that you've given them. So they've got
to pull it from the bloodstream, and they wipe out their blood
And why is that important? Because the brain serotonin cells
need tryptophan to make serotonin. If they're deprived that,
they won't make it. So you give the tryptophan depletion,
and five hours later they are hypo-serotonergic. And in this
study what they did was they put people through a laboratory measure
to sort of piss them off and see how aggressive these people would
be. And it was only the aggressive folks who actually became
aggressive, more aggressive, on tryptophan depletion.
So if any of you are worried that someone is going to slip you
a tryptophan depletion mickey, and you're going to do something
awful like commit a crime, don't worry about it, because it
only happens in individuals who are already low in their serotonin
systems to begin with.
There are other neurotransmitter dysfunctions. The catecholamine
system, the vasopressin system, and the GABA system may positively
predispose individuals to aggression. Catecholamines are a little
complicated. Some measures will show positive relationships.
Some measures will show inverse relationships. There's probably
some general dysfunction there that actually will make more sense
the more research we do.
Vasopressins are a very interesting peptide that in animal studies
clearly shows a relationship between facilitating aggressiveness.
Oxytocin, which often is opposite, and many people — including
Tom Insel, who is the head of the NIMH, feel oxytocin is very
important in affiliation, which, of course, is the reverse of
aggression. We see reverse — a reverse kind of finding,
an inverse relationship. The less oxytocin, the more aggressive.
And so let's just come back to social and emotional information
processing, because what the biology does — it sets up the
threshold. There's really this balance between bad brakes
and having the foot on the pedal. And that's the sort of
situation you walk into — into the interreaction with another
individual. But how you appraise that situation gets into social
and emotional information processing and leads one to whether
they actually commit an aggressive act at that moment in time.
So I showed you this before, but I just want to bring it back
to you. When we looked at IEDs, and we showed them this karate
vignette, and we have, you know, nine other vignettes, they're
more likely to say, "This person is trying to hurt me."
The normals don't say that for the most part. I mean, they're
less likely to say that.
But something else goes on with the IEDs. Not surprisingly,
they're more likely to be upset by this, because the other
question we asked was, if this happened to you, how angry would
you be, and how upset would you be? And they clearly are angry
and upset, which makes sense. And how much the hostile attributions
go up is how much anger — how angry they are. There's
this positive correlation.
When we do fancy statistical — actually, not so fancy
statistical analysis, it actually looks like it's the emotional
piece. It looks like how angry they get is really what targets
or really modulates their hostile attributional bias. So it looks
like emotion may be the critical thing.
So usually there's this correlation between the hostile
attributions and the negative emotional response, but, like I
said, emotion is probably even more important, because emotions
really set up how you interpret what's coming at you. And,
of course, those things are bound up in neurocircuitry.
Now, the neurocircuitry that is important, areas that are important
in emotional regulation and impulsive aggression are the ones
that are lit up in colors. The big ones really are green, which
is orbital-frontal, and red, which is ventral-medial. Dorsal
lateral has a lot to do with working memory, but the real circuit
are A, C, and D.
So A is the pre-frontal cortex, D is the anterior cingulate,
which is sort of the way station between the cortex and the —
and this is really sort of the circuit. These are all involved
to some degree, but the big circuit is really this one. And there's
probably something wrong in that circuit. We don't exactly
know where we can find that out. We're doing very specific
studies with fMRI.
But one of the first studies that was done, at least in these
IEDs, was a study done by Mary Best, who was one of my graduate
students when I was in Philadelphia. This was a study published
in PNS a few years ago, and we looked at these IED individuals.
And one thing that she did was she gave them a smell identification
test, and why we would do that — because the neurons that
are involved with identification of smell run right through the
orbital frontal cortex.
And if there was something wrong with that frontal cortex —
cortical area, perhaps they would have impairment in smell identification.
And as it turns out, they do. The IEDs cannot identify odors
as well as the normal controls. Even if you looked at only the
non-smokers, you saw the same thing. So this wasn't confounded
More interestingly, really, was this business of facial recognition
and recognition of emotions, which gets into emotional information
processing. So we showed them these pictures, and we asked them,
what are these pictures of? And the IEDs are more likely to make
mistakes. They're more likely to say a face of anger is something
else; it isn't really anger. Same thing with the disgust,
and same thing with surprise.
There's something amiss. They can't identify the faces
properly. Not in every case, but in many cases. And like I said,
they're more likely to say a neutral face is disgust or anger
or something or else.
Where in the brain are these abnormalities specifically? Well,
a very important study was done by Hannah Damasio looking at the
case of Phineas Gage. This was someone who had a normal personality
until a tamping iron bore through the skull and brain. Physically
recovered but later became irreverent and impulsive, a poor display
of judgment. When they reconstructed the lesion of the skull,
it suggested damage to the ventral-medial and orbital prefrontal
cortex. So basically the breaking mechanism was knocked down
in the system.
Now, other studies in terms of neural imaging have also looked
at these kinds of individuals. There was a study done by Adrian
Raine looking at predatory and affective murderers. Predatory
murderers are really the premeditated folks. Affective are really
the reactive, impulsive aggressive folks.
And this is a relatively crude PET study. In a PET study, you're
really looking at glucose utilization in fairly big areas of the
brain, and pretty much at rest. And what they found was that
in both sides of the brain — the prefrontal cortex —
the affective murderers seemed to have a deficit much more so
than the affective murderers would certainly be in the controls.
What was interesting, though, however, was when you looked at
the subcortical system, which would include amygdala and those
kinds of systems, both of them were a little hyperactive. So
it looked like in the affective murderers their inhibitory systems
weren't working so well, but their drive systems are working
pretty well, whereas in the premeditated murderers their inhibitory
systems were okay, but their drive systems are a little high.
Nevertheless, this doesn't really take away responsibility.
This is just interesting data to look at.
This is a copy of some scans. This was a healthy volunteer,
showing actually very nice activation in the frontal areas of
the brain. And the murderer pretty much — blue is very
low activation. And these are very pretty pictures and they've
very exciting, but, you know, it's — you have to take
it with a grain of salt to be honest with you, and you can't
really use it on an individual basis in court or to decide what
you're going to do with patients individually.
I showed you this before, this work by Larry Siever —
just to bring it back — that we're seeing the same sorts
of things, same areas of the brain affected, when you target the
This is another study by — in Larry's group —
Antonio New published an article in the Journal of Psychiatry
a couple of years ago looking at the anterior cingulate. And
what happens here is with the normal controls, they've got
very nice activations.
You go from the posterior part of the cingulate to the frontal,
but you get the reverse sort of pattern in impulsive aggressives,
suggesting there's something wrong in the anterior frontal
part of the cingulate. And that's the part of the anterior
cingulate involved in emotional information processing, so there's
something off with these folks.
Another very interesting study is a different kind of a PET
study, and this is looking at normal controls. And what they
did was they asked people to imagine themselves in an elevator
going up in a building, and they're in the elevator with their
mother and two other people. And in the first scenario they just
get in the elevator and then go up, and they get off on the floor.
In the next scenario that they imagine, they're in the elevator
and one of the guys starts bothering the mother, and they don't
do anything. In the next scenario, that you guys start to bother
the mother and they are — they don't do anything. They're
held back from doing something.
In the last scenario, they start messing with the mother, and
the guy goes nuts and beats the hell out of them. That's
the unrestrained aggression part. And what they found was, when
they looked at the unrestrained aggression part versus the neutral
condition, there was — there's a very large deactivation
in this part of the brain, and also the anterior cingulate, in
the front part of the brain, sort of decreasing inhibition, and,
therefore, allowing the subcortical structures, which are the
drive mechanisms perhaps, to come to the fore.
There's also activation, however, in the cingulate, which
is sort of this decisionmaker of what they're going to do.
So there certainly is some — you know, some imaging data
that sort of suggests that something is going on there. But we
are really in the very early stages of this kind of thing.
I think what's also important is since we've identified
impulsive aggression as a problem, we've identified certain
biological substrates. How do we treat it? There are a variety
of agents that can have some effect on impulsive aggressive individuals.
Lithium was the first drug that was shown to be — to do
this. Other drugs include SSRIs and anti-convulsants and some
other agents. I'm not going to be, you know, comprehensive
This is the first study done looking at lithium on aggression
in prison inmates, and what — this was done a very long
time ago — 1976. You can't do these studies anymore.
And they took these prison inmates and they put them on placebo
for a while, then they put them on randomized lithium or placebo.
The lithium folks' aggressive acts, not the tendency to
be aggressive, the actual aggressive acts dropped to zero after
three months of lithium, compared to the placebo where it didn't
do much of anything. But, importantly, when they put back on
placebo, went back up to where they were.
So there's two important findings there. One is that lithium
did something to impulsive aggressive behavior. Two, it was impulsive
aggressive behavior and impulsive behavior that responded to the
drug. And, three, all you're doing is suppressing the tendency
to behave a certain way. None of the drugs are going to cure
this. There are some neurobiological abnormalities that can be
potentially treated, maybe not in everybody, but in some folks.
And another important finding was that some people liked being
on lithium, some people didn't like being on lithium. The
folks that like being on lithium were the ones who felt that the
aggressive behavior was not a desired trait, that it was ego dystonic
for them. And they like lithium, they like being treated.
Other people who had incorporated aggressive behavior as part
of who they were hated lithium, which means it would be very difficult
for them to be treated, because they wouldn't stay on the
Well, the work that I did really pointed to looking at the serotonin
system, and so we did studies using fluoxetine, which is a serotonin
uptake inhibitor. We looked at impulsive aggressive folks, mostly
personality disordered individuals with this impulsive aggressive
IED situation, and looked to see if increasing serotonin in an
experimental fashion would make people less aggressive, and it
did. It didn't make everybody less aggressive. It made a
lot of people less aggressive.
But not everybody was responsive. And if you look at this at
endpoint, there's still some aggressiveness going on in some
of those folks.
What's important is in these kinds of studies we also looked
at the pretreatment biological status, and we had some data to
look at — pretreatment serotonin system function, how did
they respond to fluoxetine. And we found, much to our surprise,
a positive relationship.
So folks who had really bad serotonin function just did not
respond to the fluoxetine, and that actually makes sense, because
if their serotonin system substrate is not intact the drug can't
really do its thing. What that also suggests is that the more
aggressive somebody is the less likely they are to respond to
an SSRI, and that gets us into differential psychopharmacology,
which is very important.
We went back and looked at this, and we divided our IEDs into
people who have — were highly aggressive lifetime and moderately
aggressive lifetime. And the drug effect was only in these moderately
aggressive folks — people who were slamming doors, screaming,
shouting, and that sort of thing, maybe occasionally hitting people.
People who are very aggressive — and I don't mean
necessarily criminally aggressive — have no effect. It
just doesn't do anything. The placebo does just as much good
as the drug does. What do you do for these folks?
Well, Rich Kavoussi, who was a partner at the time, took not
these specific folks, but people like them who did not respond
to these SSRIs, gave them a totally different drug — Divalproex,
a mood stabilizer, and found that it did work. It made them less
aggressive, suggesting that there may be a way to sort of subdivide
these folks pharmacologically, not simply on the response to the
drug but on the state of their biological system.
And, in fact, a study published last year that I was involved
in — this was a study funded by Abbott, who makes depakote,
found that if you looked at the cluster B, being the impulsive
aggressive personality disorder folks with IED, they, in fact,
did respond to Divalproex, compared to placebo. And these are
folks that actually were more aggressive than other sorts of folks.
Now, do these mood stabilizers or anti-convulsants work through
something going on in terms of epilepsy? Probably not. There's
a study from Reeves that came out last year showing that it didn't
matter if the EEG was normal or abnormal. The anti-aggressive
response to these anti-convulsant mood stabilizers is really pretty
much the same.
So the method of action is unclear, but probably not by treating
some irritable focus in the brain.
Now, not to leave out psychological intervention, because in
point of fact psychological intervention probably ought to be
the first thing that you do — there are many types of this
— but it's often referred to as anger management. The
best is a system of treatment called cognitive relaxation, coping
training — coping skills training, or CRCST. Mike McCloskey
in my group is taking the lead on this and now doing studies in
IED. And, in fact, this works.
We compared wait-list individuals in groups, and you see nice,
big effects of CRCST and decreasing aggressive behavior in these
IEDs. What's really happening in CRCST is you are first teaching
them to relax and identify when they start to get angry, so they
can sort of de-escalate. And you're teaching them about cognitive
distortions, have to do the hostile attributional bias.
Well, maybe this person in the karate match wasn't trying
to hurt you or make you look bad. Maybe they were just trying
to win the fight. That's not so bad. You're trying to
win the fight. Or maybe it was an accident. If you give people
options of how other people are thinking, it diffuses it. And
then, also you have coping skills. How are they going to cope
with the anger that they have, so they don't reach the point
where they act?
And so, in conclusion, I think what some of this work shows
is that impulsive aggression, while it is a dimension, you can
look at it as a categorical expression. And IED would be the
way to do that. IED is far more prevalent than anybody ever thought.
The DSM thinks it's extremely rare. It's not rare —
four percent lifetime, one to two percent at any moment in time.
It runs in families, has a substantial genetic component. There
are neurochemical correlates, there are neuroscience correlates,
and it does respond to treatment. Not everybody responds to treatment,
not everybody responds the way we'd like to do or as quickly
as we'd like to, but they can respond. My interest in neuroscience
is understanding the behavior, not simply from a mechanistic standpoint
of biology, but the interaction between biology and environment
and genetics, and really look at the whole picture.
Sort of like — actually, what's exciting for me to
be at this meeting is to be opposite Dr. McHugh, who was thinking
this way back, what, 30, 40 years ago? Your whole foundation
is based on this whole concept of the whole person and biology
and temperament and behavior interacting.
CHAIRMAN KASS: Thank you very much.
Could we have lights and the machine off?
Before people get started with their own questions, could I
see if I can get from you something that might look like a kind
of conceptual model of what we are talking about with this particular
disorder? You talk about general — some kind of general
problem likely, say, in the serotonergic system, and also certain
suggestions that there might be difficulties in the circuits in
particular anatomical regions.
Could you sort of translate, I guess into relatively layman's
language, what does this mean in terms of your sort of understanding?
I mean, is this a problem of lack of inhibition? Is this a problem
of hyperactivity of a certain kind of impulse? Can you put together
some kind of a model that would put together these interesting
but at least to me still somewhat bewildering findings into a
kind of theory of what's going on? Or are we too early for
DR. COCCARO: Oh, no, no, we're not too early for
that. When we started out, we really had just the serotonin hypothesis.
And I even knew back then that couldn't be all of it, because
you really have response — you have a stimulus response
kind of thing.
What you really have is predispositional issues. So people
with impulsive aggressive problems are predisposed to act aggressively
in a certain way, and they can be predisposed: a) by genetic
factors which come in the genes, they can be predisposed in addition
by environmental factors, which could be prenatal, perinatal,
postnatal, as well as other kinds of environmental insults that
So that sort of sets the stage. What's going on in the
brain is a balancing act. So if your serotonin is low, that neurotransmitter
which tends to mediate behavioral inhibition in many systems will
be — won't function as well. So that tends to disinhibit.
But there are other systems that can act in — either in
response to that or can be primary.
So you can have, you know, decreased inhibition and increased
activation or any combination of those. And it's really a
balancing act. So there's really this threshold at which
anyone here is going to get angry and be aggressive. For many
of us here, that threshold is very high, extremely high.
But if you have the right — the wrong genetics, the wrong
environmental insults or influences impacting upon neurotransmitters,
that threshold will drop. So that's the first thing.
So people walk into a situation, as young people or adults,
with a predisposition with a threshold. And then, depending on
what's going on at that moment, whether they go over that
threshold depends on what's happening.
So, for example, if somebody has a low.ish threshold and then
just finished being on call and is sleep deprived, and they've
had lots of coffee, their basal arousal is going to be up here,
their threshold is here. It's not going to take a whole lot
for them to reach that threshold and explode.
But they still don't have to explode unless something is
stimulating them to respond. And then that gets into the social/emotional
information processing idea. So they're in a social interaction,
because these kinds of things happen in a social interaction.
So they're in an interaction with somebody, and something
happens to them and they have to make a decision about what's
going on. Did that person try to hurt them? Is that a threat
or a frustration? And how they respond to that, then, further
puts them towards, are they going to, you know, reach over that
threshold and be aggressive or not?
So it's a complex model. It's not a simple linear model.
And it's a probabilistic model. You can't predict —
I mean, I can predict if somebody is, you know, given the rest
of their lives going to be pretty aggressive or not. But I cannot
predict if, you know, in five hours they're going to do something.
I could tell you, okay, somebody has got low serotonin, they've
got high norepinephrine, these sorts of these — they've
got bad genes, that if somebody challenges them in a bar, what's
the probability they're going to hurt them? Pretty high.
But they could have an incident where they go a bar, somebody
challenges them, and they don't do it. It's not determined
by that. Everything is probabilistic, and that's how I feel
about these sort of things.
What we're really trying to do is, I think as clinical scientists,
is understand what those probabilities are, understand what those
mechanisms are, so we can come up with strategies to intervene,
to make them — if giving SSRIs is one of them, then we want
to do that. If giving them psychotherapy, cognitive behavior
work is part of that, we want to do that.
Is that clear enough? I mean —
CHAIRMAN KASS: Yes. Thank you very much.
The floor is open to questions. Mary Ann, then Carson. Since
the questions left over from last time I think were perhaps directed
to Professor Morse, what I thought we would do is we'd have
the questions on this presentation, and then at a certain point
try to integrate the two presentations and see if we can get the
general comments at that point, if that's okay.
PROF. GLENDON: Some of us work in environments where
it's almost taboo to consider that there might be biological
differences between men and women. So one can't help —
it's true, it's —
DR. COCCARO: That there is or there isn't?
PROF. GLENDON: So I can't help that this is going
to be probably a question which has a very simple and obvious
answer, but I can't help being struck by the fact that the
studies seem to be confined to male subjects where impulsive aggression
is concerned. And as I read your paper, I was wondering, I mean,
you did list testosterone as a factor.
But I was wondering — that can't be the whole story,
right? There are familial situations where boys and girls grow
up in violent households, but you don't have the same incidents
of impulsive aggressive behavior. I wonder if the genetic patterns
— probably you haven't studied this, but I'd just
be interested to hear you talk about why — how you see the
differences between men and women here.
DR. COCCARO: Well, it's very interesting. I mean,
certainly women can be aggressive. And, in fact, they are often
as aggressive as men. They're just less effective in being
aggressive there. They're not quite as strong, and so they
don't — they don't often hurt the other person as
Women are aggressive, and I can tell you that in some of the
studies that have been done the findings are in the men and they're
not in the women. But the biology between men and women is probably
going to turn out to be that the biological relations we see in
men more often than maybe not are also true for women. So this
isn't just in women.
In fact, the studies that I did — the study I showed you,
the fenfluramine challenge, was in men, that as in New York.
When I went to Philadelphia and was working there for 10 years,
we also did women. And we see the findings with some kinds of
aggression with women, so suicidal behavior, for example, in women
does go along with low serotonin function, but not — and
some kinds of aggression seems to go with it. It's not as
clean a story in the females than in the males, but it's definitely
In the treatment study that I told you about with fluoxetine,
they were men and women, and there was no difference in the response
between men and women in treating aggression. And the difference
between aggression in men and women is actually not that big,
you know, so there is certainly room to explore gender differences.
And, in fact, I was one of the first people to suggest that we
need to do that.
CHAIRMAN KASS: Ben Carson and then Jim and Bill.
DR. CARSON: That was certainly very fascinating, as
I was looking at the PET scans that you were displayed. We certainly
use those a lot for our seizure surgery in looking at the metabolic
changes. But I was curious as to whether in fact you see in the
anterior cingulate and the rest of the limbic system actual changes
when you have a subject on the table who is calm and then becomes
angry? Are you able to actually pick up a change?
DR. COCCARO: Well, actually, I don't do that kind
of work just yet. We're starting to do fMRI work, but our
studies are probably not going to be anger induction in the lab.
But I suspect that we'll — that you can see, you know,
some of those changes. They are functional changes. I'm
not sure there are going to be physical changes.
There might be some issues where maybe it will make it a little
bit smaller or bigger, and that kind of thing. But these things
are going to be matters of degree, and really it's a function
that's probably the most important.
CHAIRMAN KASS: Jim?
PROF. WILSON: Thank you very much. It was quite informative.
I've read some of your literature. You know vastly more than
I, but what I have read suggests that your presentation is correct.
The task of social science is to explain behavior. The task of
neuroscience and medicine generally is not only to explain, but
ideally to help it.
The task of judges and juries, however, is not to explain or
usually even to help but to judge. Is there anything you've
learned from this form of behavior analysis that you think belongs
in the courtroom?
DR. COCCARO: I think perhaps only in the penalty phase.
I don't think any of this work really can, you know, free
someone of the responsibility of their behavior. So I feel very
strongly that if you're going to do anything with this work,
it really ought to be in the penalty phase, not necessarily they're
going to get less of a penalty, but perhaps they get treated,
or perhaps instead of getting the death penalty they get life
and get treated, or something along those lines. That's how
I feel about it.
PROF. WILSON: Thank you.
CHAIRMAN KASS: Diana, follow up on this?
DR. SCHAUB: Yes. What about the disease jurisprudence?
Do you see it playing a role there? I mean, you said that you
would prefer that these interventions be voluntary. But could
you imagine situations in which there would be involuntary interventions?
What would be necessary to justify an involuntary intervention?
DR. COCCARO: Well, you know, I'm going to agree
with Dr. Morse. That's not a decision for a scientist, to
be frank with you. But that's something for society at large
to determine. I think it goes to behavior. How often is somebody
doing something, presumably that will correlate with things in
the brain. But we're not really there yet. I would prefer
that we talk about voluntary treatment rather than involuntary
DR. SCHAUB: But you also mentioned that you know something
about the lithium haters.
DR. COCCARO: I'm sorry?
DR. SCHAUB: You mentioned that you know something about
the lithium haters, people who value their aggressiveness.
DR. COCCARO: Yes. Well, I mean, I think it's —
I think it's — this may skirt the issue, but I think
it's up to the legal system to make the decision about whether
or not they should be forced to take lithium or not. I mean,
I think perhaps they should. It's going to depend on a case-by-case
basis. I'm not prepared, as a scientist, to endorse that
point of view at this moment.
CHAIRMAN KASS: Bill Hurlbut, then Paul and Rebecca.
DR. HURLBUT: I'll wait until he goes.
CHAIRMAN KASS: Okay. Paul, and then Rebecca.
DR. McHUGH: Yes, I enjoyed that presentation very much.
And you are following, in my opinion, the most likely and fruitful
design of describing things in terms that ultimately can be translated
into brain functions and understood in brain — both in the
sense of brain structure and function, but also in the brain's
responsiveness to learning.
But I had just a couple of questions. As I've been following
your work and have been fascinated by it, this is an opportunity
for me to ask some questions that might be useful to the other
members of the group.
The first thing I just want to understand is you described this
categorically in DSM-IV terms, but do you mean it categorically,
or do you mean that this is an extreme of a dimension that we
might understand and ultimately see that by focusing on this group
out at one dimension see something that is otherwise a bell-shaped
curve in the population.
The second thing, given that kind of thinking about this matter,
one of the striking features about your talk is that you haven't
used the ordinary personality or temperament terms to describe
some of these features. You talk about this as being cluster-B,
but are the features, particularly the ones that might be related
to women and to other people, related more specifically toward
dimensions such as the combination of extroversion and neuroticism?
And if that's the case, are these more likely — these
patients responding to the serotonin to be the ones that are more
extroverted and neurotic versus the ones that don't, turn
out to be more either not — don't have such high neuroticisms
and even more extroversion problems.
And, finally, you say nothing about the learning process. Are
you thinking of this learning from — in the developmental
process out of conditioned learning or out of more social learning
of an (Albert) Bandura type or things of that sort that are playing
a role in developing and shaping the brain in this process?
DR. COCCARO: Okay. The first thing is that I am more
of a dimensionalist. And what we're doing with IED is really
focusing in for maybe, to some degree, convenient biomedical terms,
looking at the extreme dimension.
The curve is actually a J-shaped curve, depending on the thing
you're looking at, because most people aren't that aggressive
and then you get these extremes out there.
The second thing has to do with extroversion, the classic personality
dimensions. And, I mean, I've looked at these, and they don't
correlate with serotonin measures the way — at least in
my studies, the way one would expect. So it does seem to be more
of an aggressive impulsive kind of thing.
As far as the learning issue, I must confess that I'm starting
to get into this myself more these days. And so I haven't
really explored scientifically or empirically what kind of social
learning we're talking about. I'm still looking at —
I'm still sort of dissecting things out as I go along. But
certainly the learning process is in there.
I mean, as people grow up, they — you know, if kids —
if you have a parent who is aggressive, it sets up a whole bunch
of problems, because that person probably has genes that predispose
them to being aggressive. So they're going to pass those
Then, they're aggressive to the kids, so the kid gets aggressed
upon. And then the trauma itself is going to have profound changes
perhaps in brain structure and function. And then, at the same
time, they learn that this may be one way — and it may be
an appropriate way — that one should respond to frustration.
The parent is angry at the kid because they're frustrated
at work or something at home, so they beat the kid up. So they're
sort of learning in that sense. But I'm still learning about
learning processes myself.
CHAIRMAN KASS: Rebecca Dresser?
PROF. DRESSER: I had some questions about potential
justification for treatment or some sort of mandatory use of this
— of treatment. To what extent does this behavior have
good effects? Whether it's something related to the neurotransmitters
or this impulsive aggressive behavior could be useful.
I once worked with a person like this who used it quite successfully
to intimidate people. Now it probably wasn't a super extreme
case, but I could see that it would have personal rewards. So
to what extent would someone who is considering a treatment be
interested in it? What would that person be giving up?
And also, are there possibly socially valued behaviors associated
with this? Something — I don't know — creativity
or problem-solving ability. Something that could be useful in
a social way that might be involved with this same event going
on in the brain that might be lost if we were to try to treat
this with a drug.
CHAIRMAN KASS: Like intolerance of injustice.
PROF. DRESSER: Excuse me?
CHAIRMAN KASS: Intolerance of acts of injustice.
PROF. DRESSER: Yes, right. Right. Intervening to assist.
And to what — I mean, what percentage of people are interested
in seeking help? And what percentage find this a useful way of
going about in the world? And I think we probably would have
to confront this issue of when should it be imposed.
DR. COCCARO: Very good questions. Maybe I'm not
imaginative enough, but I can't see a socially justifiable
reason to be impulsively aggressive. I just can't see it.
I mean, if a person you're talking about knows they have —
the question for me about that person you're talking about,
do they know they have an impulsive aggressive temperament, and
just let it run free reign because it gets them something? That
becomes kind of premeditated in the way Dr. Morse was talking
So, you know, and that's not appropriate anyway. I mean,
you shouldn't be using, you know, this kind of behavior to
intimidate people. As far as treatment seeking, part of the problem
with aggression has been that it's seen as bad behavior.
And it's just bad behavior, and there's no biology to
it, there's no science to it, nothing. It's just bad
behavior; we have to punish it.
And because of that, everybody feels — most people in
the community feel it's just — oh, you need an attitude
adjustment. So nobody knows about the possibility — the
general community doesn't know from — this could be
a disorder that has biological, you know, roots that can be treated
in a biomedical context. Therefore, nobody seeks — people
don't really seek treatment until either their — their
whole lives are falling apart or they're forced to seek treatment.
In a study that I've written up — it isn't published
yet — we looked at people in a general psychiatric clinic
who had this problem. About six percent of this population actually
had it. And I don't remember the number precisely, but most
of the people who had the diagnosis actually wanted to be treated
for it, although nobody approached them to be treated for it,
because the caregivers — I mean, their treating physicians
— didn't see it as something that you treat.
It's just bad behavior. It's part and parcel of something
else that goes on. So I think that's part of the problem.
And I think one of the reasons that I wanted to develop IED in
a better fashion was to do the typical things that people have
done for their disorders.
If you don't — you know, sometimes you want to label
things. But if you don't label things, you can't get
a sense of how much is out there. And if you can't get a
sense of how much is out there, you'll never convince anybody
that it's a problem worth studying or treating.
The antithesis of aggression might be social — generalized
social phobia. People are extraordinarily shy. I can tell you
that the drug companies have gone nuts with social phobia. I
mean, they have gotten indications to treat — everybody
is treating. There's nothing wrong with social phobia. It's
a very prevalent condition. It's bad to have it.
But gee, is it any less worse to be impulsively aggressive and
get into trouble? No. But are the drug companies going after
the development of agents to treat it? No. Partly because —
I don't think it's a real problem partly because of the
lawyers, you know, liability issues and those sorts of things.
So part of the reason of, you know, trying to develop IED as
a — and it is in the DSM, but we have just been fixing the
criteria, is to raise awareness of the fact that there's really
a scientific body of knowledge behind this. And so that ultimately
people would realize, gee, you know, this is a problem just like
depression, just like panic disorder, and it can be treated.
It doesn't get you out of anything. It's just something
that, you know, needs to be treated, and I think that's where
I'd like to ultimately go with this.
I've actually gotten a little sidetracked with the IED story
from the neuroscience that I've done. But I feel it's
important to sort of develop the criteria better and develop the
empirical database behind, you know, looking at and studying these
Have I answered your questions? I wasn't sure. There was
one I was missing.
CHAIRMAN KASS: Before we open it up to more general
things, I have a comment and Jim has another. Are there other
people for that queue?
And this might also in a way echo the kind of question I put
to Professor Morse earlier. I know that you don't see this
as altering the way we ought to think about moral responsibility.
Yet as I listen to you, I can't help but think that it might.
You begin with — you present us Phineas Gage, and the
changes that happened to him he's not responsible for. And
you present us porphyria and various other kinds of clear diseases.
We were talking at the break about known temporal lobe seizures,
and it's not clear to me, once you develop some kind of a
syndrome here, admittedly it's got environmental components
and genetic components — and I like very much the rich way
in which you lay this out.
But to the extent to which you see this as an opportunity for
prediction, interventive prevention, and therapy, you are in a
way beginning to assimilate this kind of conduct to a disease,
to assimilate it to the way in which we think about disease.
And maybe it won't affect the judgment of guilt or innocent
in any particular case, because you don't, as you say, have
predictive power in any particular instance.
But it would surely, I think, play a role in mitigation when
people pronounce sentencing. It would play a role in questions
about whether you would recommend, encourage, exhort to, and even
compel certain kind of medical interventions down the road. So
that — I mean, I think there's a kind of perfectly understandable
and cheerful sense that the understanding of behavior, at least
as the law is concerned, is sort of impervious to what science
is going to provide here. But I don't see it.
I mean, it seems to me that there is — and I'm not
complaining about this. I mean, I'm not complaining about
this at all. It seems to me that, to the extent to which we understand
some of these people as being analogous to the cases of temporal
lobe seizures, it's going to be hard I think to simply say
— to remind yourself that, well, if the policeman were standing
there, they might not have done it. That might not be sufficient
for the way in which we've come to think about it.
So am I off base with this?
DR. COCCARO: No, I don't think so. I mean, my point
of view is that we can't use this information. The only kind
of people that I study, because they don't have temporal lobe
epilepsy and these other kinds of things —
CHAIRMAN KASS: Right.
DR. COCCARO: — as explaining away what they did
and their responsibility. And as I was saying to Dr. Morse at
the break, I was involved in a case where somebody had clearcut
serotonin abnormalities, and he killed his wife, and it was in
an act of range. But it was absolutely clear behaviorally that
he knew what he was doing. You know, he just gave in to his frustration,
and he did it.
And the lawyers in the case wanted to make a new law —
a serotonin defense. And I told them, "That's not going
to work." I mean, you should use this in the penalty phase.
This is something that we should be using so that he doesn't
get the death penalty. You know, that there were mitigating circumstances,
and that maybe he should be treated.
So my feeling about this really goes not to explaining away
responsibility. In most cases, I think if you — I think
you could theoretically come up with a case where, yes, the person
really couldn't do it, but even there it's hard, because
people don't act aggressively all the time. Right? But they
act aggressively often enough, and, yes, you're right.
As Dr. Morse said this morning, you know, they know they're
going to do it in certain circumstances, and they need to do whatever
possible — society needs to do whatever possible to prevent
them from getting in those situations.
I think my emphasis is really on strategies for intervention
and treatment. And whether we compel it or not I think is up
to society. You could have a situation where you go, "Well,
I'll let you go if you take these medications, and you can
prove to us that you're taking these medications. But if
you refuse to do it, we're going to put you back in jail,
because we can't trust you out there."
But we're not there yet with the therapeutics anyway. And
predicting — we're not there with predicting. I'm
not sure we'll ever be there with predicting, because all
you need is one bad event. That's it. Well, he was good
99.999 percent of the time. Yes, but one time he killed somebody.
I mean, it's not acceptable — I hope — that we
send people out who have killed people to do it again.
CHAIRMAN KASS: Thank you.
Jim Wilson, and then we'll open it. On this, or in the
queue? On this point? Oh, in the queue. Jim and then Dan.
PROF. WILSON: Now I want to make two rather than one
point. In response to Leon's question, I think it's
important to realize that in the penalty phase of a criminal trial
everything is taken into account — the probation report,
the testimony of friends, increasingly of late the testimony of
victims, whether the person had had jobs, whether the person was
subject to abuse, whether the person had a prior record, the nature
of the prior record, the possibility that some of the things that
a neuroscientist might add to this is real, but it doesn't
alter the situation.
The penalties are determined by judges after a presentence report,
unless the case is so cut and dried, or unless the jury is required
by law to make the decision on the penalty issue, as they are
with respect to the death penalty. So that the existence of neuroscience
does not change the logic of the criminal justice system. They
will take into account all causes of behavior in assessing a penalty,
within ranges set down by the law.
With respect to Rebecca's question about does IED or impulsive
aggressiveness ever have any value, let me report a piece of research
that was done a few years ago, hence may be out of date, though
I doubt it. This was a careful study done of a randomly selected
group of people who had — some of whom had gone to the aid
of a person in distress.
A woman was being abused, a child was being beaten up, somebody
was harassing other persons, and they were studied. And the one
thing that becomes absolutely clear is they were not motivated
by concerns for justice or fairness or the rule of law. They
were mad. They were mad at the offender.
They often got mad, and here the madness was useful. Now, it
would be nice to design a world in which their madness was only
reserved for dealing with people who were being threatened.
Happily, we can't control human behavior, but there are positive
signs of impulsive aggressiveness.
CHAIRMAN KASS: They didn't jump in to beat the victim?
PROF. GLENDON: Jim, my recollection of that study is
that it also suggested that these Good Samaritans were actually
people with a history of rather antisocial behavior. Is that
PROF. WILSON: That's quite right.
CHAIRMAN KASS: Dan Foster?
DR. FOSTER: I just want to make a simple statement that's
probably obvious to everyone. But there seems to be a tendency
sometimes, like this morning, to jump way ahead of the implications
of the science that's involved in the neuroscience and make
assumptions that the answers are there.
We always start — and Dr. Coccaro was very cautious about
— here with correlations, so there are correlations between
the serotonin system and some behavior. And he's very cautious
to say, you know, that that's not the whole answer. You have
to way beyond correlations. You start with them. You've
got to go way beyond that, and the same thing for the imaging.
For example, the PET scans are all — almost all just an
uptake of glucose into an area where you can look at it. Nobody
has even begun to look at the role of other nutritional factors
about glucose uptake — for example, fatty acids and other
things that may alter — that are changing throughout the
day when we eat, and so forth, that may alter PET scans and their
relation to things.
I just — all I want to say is that we need to be very
careful to not put too much emphasis on where we stand now and
its subsequent social or other behaviors. We have to start.
You know, it would be — it would sort of be like looking
at genes before we knew the genetic code.
You know, in other words, the things that we're looking
at are very — one would have to say very primitive, and
I think we just need to be cautious about trying, therefore, to
put a structure for the courts, and so forth, based on this until
we know more about what's going on.
It is — I think Mike would be better to say than I, but,
I mean, it is a very early time here, and we need to be cautious
about the implications of these things. That's all.
CHAIRMAN KASS: Thank you.
I think we should turn to the general comments, if we could.
DR. McHUGH: Can I just say one more thing, just to draw
out the answer to your question about this being a dimensional
matter? It is a dimensional matter that's being studied primarily
amongst people who are complaining about this. You're not
going out and studying the population in the world.
If this is a dimensional feature, then we presume that there
are people with this dimension, lesser or not so severe, that
don't come to doctors' attention, and those people might
be very well — doing very good work out there, and this
dimension may be serving great purposes for them in some other
way. Is that not possible? I mean, we are looking at this from
a clinical — in a clinical population, and if it's a
dimension, a human dimension, then we're getting a skewed
view of the population and of this dimension.
DR. COCCARO: Well, yes, to some degree that's true.
I can say that we don't have an anger clinic. I mean, people
come in because we put out public announcements saying that we're
studying this kind of thing, and they come in to see about that.
So they are sort of clinical, but they're not pure clinical
in the sense that they're coming in and —
DR. McHUGH: But it's not a population survey.
DR. COCCARO: No, that's not a population survey.
The twin studies we do are population-based.
DR. McHUGH: I see.
DR. COCCARO: Yes.
DR. McHUGH: But then, to the answer to the question,
would these perhaps be helpful, we might look at people who are
also asked to be aggressive and see how many of them, in fact,
fall into these kinds of brain scan and biological manifestations.
By the way, just to back up what both Jim and Mary Ann were
saying, even before we were doing brain scans and biological explanations
— again, psychologists like Joe Matarazzo and other people
of that sort were looking at criminal behavior in policemen.
And with dimensional features, they found they were remarkably
similar in their — in these temperamental characteristics,
although they were very different in their social understandings,
and the understandings of what they had ultimately learned over
time worked for them and worked for all of us. Thank goodness
we have them, and they are ready to go out and do good work.
Do you plan to — again, to continue your themes of this
dimension and answer Rebecca's question in this theme, to
start looking at people of that sort who aren't coming to
you complaining of their anger, but are saying, "Gee, you
know, that's what's making me captain or master sergeant
or all those wonderful people that we need."
DR. COCCARO: It would be a very exciting study to do
that. We can probably get at some of that with our twin program,
because we have a population base — the twin program from
Pennsylvania, we bring them in every week. And these people
are not coming in for treatment for anything, but they've
got stuff. They've got lots of psychopathology. And as we
sort out what they have, we can probably do that.
CHAIRMAN KASS: Bill Hurlbut, and then Diana.
DR. HURLBUT: This is kind of a subset of Leon's
question earlier about culpability, and it goes to the root of
the question that Dr. Morse was bringing up implicitly —
the notion that eventually you might have some findings from neuroscience
that reach the level of foundational revision of our understanding
of human nature, and, therefore, altered views of culpability
What I want to ask you about is sort of a general topic. But
when you say that the mind — that the human being is a conscious,
relational, intentional being, there's a lot packed into those
three words. And take for example the word "intentional."
Okay. It's aboutness or relating to something outside yourself,
a purposefulness, a whole context actually. Human intentions
are based on a whole large image of the world in which they live.
And in a sense, as each of us develops individually, we undergo
a process of personal construction, where we put together a large
view of the world based on experience, observational and interactional
kind of relational dynamics. So we eventually become rational
in a sense of a resonance along a sort of alignment of the ratios
of the way the world is.
And our behaviors, then, are aligned with the realities we see.
We develop images and ideals according to those, and ultimately
I suppose it's fair to say that our motivations and, therefore,
our actions are deeply rooted in our most foundational aspirations
or images of what life actually is.
So, then, for the healthy person at least, one could say that
one's construction is inseparably related to one's moral
disposition, and goes finally down to the level of what you might
call a philosophy or a faith, a kind of deep story about what
our existence is.
Now, do you accept all that as a reasonable start?
DR. MORSE: I accept a lot of it, and enough as a premise
that I'm happy to go with you.
DR. HURLBUT: Okay. What I really want to ask goes to
another question about that. If one takes that constructed identity,
and, therefore, constructed moral mind seriously, then what I
really wanted to ask you about is the question of Phineas Gage
or, to put it more relevantly, the analogy that we would use with
that obvious pathology extended to a much more prevalent pathology
of sociopathy, which, according to some people, is as high as
50 percent of people on death row and 20 percent of people in
the penitentiaries. Some people say that's what the incidence
of those is.
Okay. So what I want to ask you is this. When you use the
word "rational," aren't you smuggling in already
a huge premise about the way the mind is as though we can stand
outside of the world and look at the world the way you've
said we couldn't? Is not the rational mind actually a strangely
emotional relational mind? And what do you make of — well,
put it first, isn't the development of childhood moral responsibility
what you'd call the infancy excuse or — somewhere that
was used in our readings.
What's going on in development of a child? There is a development
of the moral mind, among other things. And what happens if your
capacities for this are diminished. And here I'm getting
at the broader category of sociopathy as focused by studies of
Phineas Gage like people done by Hannah Damasio, where she found
that if infants had certain lesions, they would never develop
moral capacities, whereas if those same lesions took place in
late adolescence, they would have at least routine moral alignments.
So what I'm asking is a broad question. Might all this
revise our sense of how the moral mind is developed, such that
we start to see that there are some people who are almost what
you might call morally color-blind, that it's not just that
everybody is rationale but that the rationality of being is actually
as strangely — I don't know what a right word would
be — but relational, dynamic, that some — that it's
not just impulsivity but a fundamental moral incapacity that might
be underneath a lot of crime.
And you know what I'm asking here, right?
DR. MORSE: I do.
DR. HURLBUT: Okay.
DR. MORSE: Well, that's a very, very hugely broad
and deep question, and there's also a specific question.
The specific answer would I think also allow people to infer the
broader answer. And since it would be immoral to keep people
from lunch, let me do the shorter form.
You've asked about sociopathy. I know I'm dealing with
a sophisticated crowd here, but let me just briefly say that there
is a difference between the DSM-IV antisocial personality disorder,
which is largely behaviorally defined. You get in trouble with
the law, you have problems paying your bills, it's all behavioral.
And then what's known as the classic psychopathy concept defined
most well-known by Hervey Cleckley in the Mask of Sanity,
which is more a psychological concept having to do with analogy
of the short form, failure of guilt, failure of empathy.
And we, in fact, can measure that now very, very well behaviorally
with something called the Hare Psychopathy Checklist Revised,
which is an enormously powerful and robust measure for picking
out these people who seem to lack the moral capacity, and where
there is an enormously powerful association between antisocial
conduct and high scores on the psychopathy checklist.
All right. Having said all of that, so now we know people out
there who lack these capacities. Now, think about what it is
that allows you, in the face of temptation, not to do things you
know you ought not to do. It's not just there's a policeman
at your elbow, so that's, in a sense, an unfair test, and
Dr. Kass is right.
It's what you have a capacity to think about in terms of,
gee, I'm going to feel guilty. What would it be like to be
a victim of that myself? You have empathy. You have these internal,
if you will, checks. Now suppose you don't have those checks.
It's going to be much more likely that you are going, in effect,
Okay. So now comes the question. We have these people. They're
highly — and they are overrepresented in prison. And by
the way, the overlap between antisocial personality disorder and
hare psychopathy is large but not perfect by any means. And it's
assumed that somewhere between 40 to 60 percent of people in prison
have antisocial personality disorder. A somewhat lower number
are — suffer from psychopathy as defined by the hare.
Should these people be held morally responsible? The answer
of the law is yes, for precisely the reason Mr. Hurlbut said.
They know what they're doing, they're instrumentally rational.
They're A to B rational. You know, they have a premise, they
can get there from here. They're not out of touch with reality
in any gross way, but they lack this particular capacity.
The philosophers are almost uniform in believing that they ought
not to be held responsible, because one of the necessities for
moral rationality as it were, was the ability to feel guilt, to
have some empathy. You don't, in other words, have the tools
you need to reason well about what you have reasoned to do if
you don't have guilt and you don't have empathy.
I'm on the side of the philosophers on this one. I would
say these people are not morally responsible for what they do.
By the same token, I'm then entitled to control them if they
are particularly dangerous. I mean, they are psychopaths who
are CEOs of major corporations, and the like, and —
— perhaps we don't need to control them. People who
are, let's say, murderers, rapists, and the like, maybe we
do. But it's — again, having the science of this does
not answer the question of what we ought to do with these people,
either morally or practically.
As far as your deeper question, I have one sentence on that.
Yes, our identities are socially constructed. Yes, the concepts
of rationality and intentionality are normative. There is no
obvious right answer to what counts. But I would like to think,
since we are at least in some fundamental ways all the same, wherever
and whenever we have been born, wherever and whenever we've
been born, that there are certain human capacities that we do
have in common that have sort of a fundamental base, or what I
mean common sensically by rationality and intentionality.
CHAIRMAN KASS: Diana, and then with one last intervention
DR. SCHAUB: I just want to ask you for a prediction.
I mean, you sketched out that we have dessert jurisprudence, and
we have disease jurisprudence. Presumably, in the past, the disease
jurisprudence was a very small slice of that.
Even without some kind of foundational discoveries or hard science
to support it, do you think we will increasingly move towards
the disease jurisprudence? I mean, the scientists might be telling
us to be cautious, but will citizens be cautious? And that this
shift towards disease jurisprudence might not be out of softheartedness
at all, but out of a desire for more social control.
You've already said that we see this in the case of sexual
predators. We're very frightened of sexual predators. Maybe
we're not as frightened of, you know, impulsive aggressive
people. But do you think we will see a shift in that direction?
DR. MORSE: I hesitate to make a prediction, so all I
can do is sort of talk about what will be the forces at work.
The reason these mentally abnormal, sexually violent predator
commitments were instituted, we used to have them back in the
middle of the 20th century. They were — they fell into
destitute or were abolished. Then we had them again starting
It was the fear that we can't keep people in prison long
enough sometimes, and we've got to control them when they've
very dangerous. The way you do that is you "medicalize"
the problem. You convert to the disease jurisprudence. You then
claim these people are non-responsible, and then you have warrant
in our system for holding them involuntarily because they're
If you look at the criteria in these mentally abnormal, sexually
violent, predator commitment cases that the Supreme Court has
approved, all you have to do is take out the word "sexual"
in these criteria and just leave in the word "dangerousness,"
and they would be precisely the same.
And in terms of the definition of who qualifies as abnormal,
they would be precisely the same. There is no doubt in mind whatsoever
the Supreme Court would approve such broadened form of disease
jurisprudence if legislatures wanted to do at. At least the present
And by the way, in approving the criterial language for these,
it was nine to nothing in the Supreme Court of the United States.
Everyone agreed. Gee, these people — you know, the legislature
says they can't control themselves, they can't control
themselves. This is okay.
If people began to fear the IED folks that Dr. Coccaro has so
well described to us, could we have expanded disease jurisprudence?
Absolutely. I see no block to doing it constitutionally if legislatures
wanted to do it.
We are particularly afraid of sexual folks — sexual predators
in our society. If we got particularly afraid of IED folks or
anybody else, it is I think, if you will, legally doable.
CHAIRMAN KASS: Thank you.
Before breaking, Mike Gazzaniga, if I might ask, is there something
you want to comment on as the Council's resident guru on neuroscience
DR. GAZZANIGA: Well, both speakers took my position,
— they said it better than I could. So I think neuroscience
has very limited, if any, role in the courtroom. And I think
the neuroscience that is done is — does relate, is probabilistic
and correlative and not at all — do we understand it enough
to have it playing — or being used in a causal role? So
my view was represented well this morning.
CHAIRMAN KASS: Very last question to you and the two
speakers. Does that mean if — if the NIH were to set about
a major study on the neuroscience of criminal behavior, that we
should be for it or against it?
DR. GAZZANIGA: It depends who gets the grant.
DR. MORSE: For me, it would depend entirely on the design
of the study and the purpose of the study. I can certainly imagine
studies that would be useful, and I can imagine studies that would
be a waste of the taxpayers' money.
CHAIRMAN KASS: Since you do some of this research —
DR. COCCARO: Yes, I do it, and I'm funded to do
it. Although I don't really study criminals. I would be
in favor of it, but, yes, I mean, we have a peer review system
that keeps that in check, so that, you know, the study design
is very, very good, and that we're not wasting taxpayers'
But I'm all for, you know, studying more and more about
this, so we can come up with targets for intervention.
CHAIRMAN KASS: Thanks to both of you for very clear
presentations and great generosity and forthcomingness in the
exchange. Please join members of the Council for lunch, if you
And, Council members, we reconvene at 2:00. You have an hour
and 20 minutes for lunch. That should be enough.
(Whereupon, at 12:36 p.m., the proceedings
in the foregoing matter recessed for lunch.)