2003 LECTURE SERIES

Opiates, Brownies, Sex and Cocaine: Seeking the Brain Signature for Desire

Dr. Anna Rose Childress
VA Addiction Treatment Research Center
University of Pennsylvania
January 29, 2003

I'm wired, and I'm also interested in giving this lecture tonight.

Let's see if the slides work. An imaging talk without images is really not what it should be, so let me see if I can get this to slide show and get our first slide up here. I tried to get colorful slides. I know that for you it's only 7:30, but for me it's 10:30, so I need to be able to stay awake for this.

I'm so very glad to be here this evening to talk with you about this topic that's so close to my heart and that's been really the topic of my research in one way or another since the graduate school days that Dr. McGaugh mentioned.

I'd like to thank Dr. McGaugh for giving me this opportunity, this honor to be here to kick off this year's lecture series, and I'd like to also thank Lynn Brown for helping arrange the many, many things that enabled me to actually get here. This included preventing snow in Philadelphia, this included guaranteeing sunshine in California, this included finding a surrogate laptop when my laptop was knocked off a stack of luggage at airport check-in, and then I was beginning to think, how would it be to have an imaging talk without any images, and it didn't feel good. For a while that's what I was thinking. And of course, I would like to thank, the lights are very bright but, Dr. Larry Stein, one of Irvine's own now for two decades, but in his former East Coast life, he was the mentor who really made the brain [inaudible] award the obviously most rewarding target for my research energies for all this time, so without Larry Stein, you would be having a different lecture tonight, and it wouldn't be me giving it, so I'm very glad to be here.

So my topic this evening, ah, yes, my topic this evening are the profound desire states for reward, which undoubtedly give us some of our best moments as human beings, but when poorly managed, they can also get us and our patients in a whole heap of trouble. Now, we have something to blame for all this. We have evolution to blame for the pull of these desires. From an evolutionary perspective, our brains have evolved to keep us alive and perpetuate the species so it evolved systems which help us quickly learn to see the signals that mean danger so that we can run and fear, or fight if we need to, to avoid becoming some other creature's lunch, but on the positive side of things, and that's really where I'm going to be most of the evening, the happy side of motivation.

The brain has evolved systems that are highly conserved through evolution, which give natural rewards such as food and sex a powerful pull. They give them almost an imperative, gotta-have-it quality. Now, for sex, this is so we can continue as a species and for food, so that we can, well, live long enough to have some more sex and reproduce. And fine, I can just hear you saying okay, fine, food and sex, no problem there, what's the problem, sounds good. Ah, but here's the rub. The brain has to decide when to answer the call of the reward system, when this call is to be answered and, conversely, when it's too dangerous to answer the call, when it's disadvantageous. This is a lot more difficult; this is where we have more trouble. Let me go to the next slide and see if I can illustrate this a little bit. Let's get concrete for a moment.

Ah, let's return to ancient times. You can tell we're talking ancient times here because of the primitive quality of this cartoon that I drew. So here's the situation. We're gonna have an ancient Everyman here, and this ancient Everyman has had an incredible romantic encounter with the receptive female in Primitive Hut A here, see the little heart? And that's a positive thing. And so this very night, across the fires that light the hut, he saw her and there was a chemistry. He knew that she was probably receptive, he could remember what it was like when they were together, the incredible encounter, it was glorious, how she smelled, how she tasted, good heavens, the touch of her skin-he's definitely feeling a pull. Ah. Only one little problem that's not so little. Here in Primitive Hut B is the very mean alpha male, for whom she is a favorite, and so our Everyman now has had a problem. He's been beaten soundly by this guy at least once before. The next encounter could be the last one, he could get into real trouble.

So the pull is very strong, the distance is relatively short, she's really pretty hot, but the potential costs are very high. So here's the dilemma. Does he go with the pull, get the girl, ride into the sunset, or does he call it a day, wait 'til the alpha male is sick or drunk, another time to make his move? Or does he make a move, get killed by the alpha male in his attempt to mate, to reproduce, to help the species go on? You decide.

But here we go, with apologies to Shakespeare, the real issue here, for most of us, is really to go or not to go, right? It's the dilemma. It's not only his dilemma, it's really truly our own, basically weighing the pull of reward against the consequences of pursuing it. It's really our daily evolutionary legacy. So this legacy can help us understand puzzling behavior, all kinds of interesting things, why a politician might risk, oh, say, his presidency for an appealing natural reward.

But it's this broad framework that I'm going to use this evening to try to understand, in these terms, our patients' difficulties and the disorders of desire, the addictions, to go or not to go. Now, with regard to the addictions, our primary questions really have been in the area of vulnerability. These disorders of desire, vulnerability, how is it that people get addicted? Not everyone does. And among those who become addicted, some people have a lot more difficulty leaving their substance or leaving their activity. There is a difference in the rate of relapse, though rates of relapse are really miserable for addictions in general. For most people who manage to leave a substance, or manage to lose weight, or manage to stop a compulsive, addictive sort of pattern or behavior, the rates of relapse are high. They hover at around 80 percent at 6 months, so relapse is the big clinical issue, and it's really the place that we entered this picture.

As I hope you'll see by the end of the hour, I think there's an overlap in these two arenas. Understanding the vulnerability to relapse may also help us understand something about the vulnerability to addiction itself, who gets there and who doesn't. And just a note here, for those of you who came for one of the other of these substances or activities, I promise I'll do a little bit about each one, but since we have 50 minutes, now 45, it'll be a survey, and I'll take speaker's prerogative here and I'm going to start with cocaine, because that's really been the place that most of the research in my lab has been in the last ten years. We've done work with opiates, we've been doing work with nicotine, alcohol, and of course, we constantly work with chocolate in our lab, at least informally.

So, let's talk a little bit about cocaine and let's talk a little bit about the issue of relapse, because that's really our entrée into this topic. Relapse rates are very high, relapse rates are high not just for people who get all the press, like Matthew Perry or Robert Downey Jr., but they are for high for all of our patients. And when you think about the addiction cycle, it's important to recognize that most of the relapse actually occurs long after withdrawal, so long after people have stopped smoking or stopped cocaine, this return, this tendency to return, is profound. So it's not just the drug is still in the body or the person is getting sick. Long after any observable signs of withdrawal from the drug, long after the drug has left the body, the vulnerability to relapse is there. What triggers it, what precipitates it, what precedes it, was our clinical question.

Our patients would give us stories about profound desire states, I'm going to use the word craving, they don't even always use that word, that often preceded relapse. And importantly, these desire states also were not confined to a period of when they were acutely leaving the effects of the drug, but could occur at long periods post-drug, months, even years. What kind of state is this? And so I want to introduce this state to you since we're going to be talking about it for quite a bit, to give you a feel for what it's like, the cocaine desire state, and I could give you a description, but I'd like to offer you one that actually was written by Patti Davis, who is the Reagans' daughter, and she wrote it on the occasion of Robert Downey's re-arrest. This was in May 2001, and here is her description of a desire state, and she does a really good job with it, and it really parallels most of what our patients describe, so I'd like to read it aloud to you. She's talking about cocaine as her lover.

"My lover was cold and cruel and hardly faithful, but I never fell out of love. Every time I see a movie in which people are doing coke, I want it. I can almost taste it in the back of my throat. I still love that taste. You don't get over the drugs, you don't ever fall out of love."

So there are some interesting points about this quote. One is, she'd been abstinent for some long period of time at this point, so there's a persistence of this vulnerability. Under the right circumstances and with the right video, the right movie, she can still get into this state. Also, it's no accident that she's using romance language here. She's talking about this in a very profound, positive, appetitive, chocolate croissant, wanna-have-it, sexual object kind of way. So this is not a sickness state, this is not a withdrawal state, this is a profound pull toward this substance.

And I want to make that point because basically, she's describing something that is really based on the desire for euphoria, so I'm going to co-opt this word for the time being, for tonight, this craving word, for this fuchsia side of the screen. It is used in this field, in this addiction field, to mean a lot of different things, and it's a free world, people can use the word any way they want, but for tonight I'm talking about this side of the motivational coin, the pull toward reward, and I'm not using it in this other way, though these may be reasons that people could use a drug, to get away from a withdrawal or a discomfort or some other negative state, but I'm focusing on this side of the coin.

The other thing that is interesting about her language, if we go back for just a moment, if I'm lucky, is that basically she is using these sexual terms, these love terms, the circuitry that we talked about a little earlier with our ancient caveman, is really an ancient circuitry, been around for a long time, and these drugs of abuse are relatively recent, right, in the history of our evolution. So they have to act somewhere-where do they act? Well, they're acting, it looks like, on this old circuitry and activating it in a very profound way, in a supra-normal way, and what her quote illustrates is that cues for the arrival of that drug in time, acquired this really interesting ability to trigger a throbbing, pulsating desire state. That's why it's in fuchsia. Which, of course, is clinically important because it may lead to relapse.

Instead of saying throbbing, pulsating desire, which takes a lot, especially if you're going to say it a lot, I'm going to call it a Go state, and that's nice and short, and basically, you'll be hearing me talk about Go the rest of the evening, and I'll be talking about how these cues that we've been looking at can trigger this state and how we studied it under control conditions in our laboratory.

So how is it that these cues, like the video, or someone talking about cocaine, or the sights and sounds related, how is it that they acquire this interesting ability? Well, as you heard Dr. McGaugh mention earlier, basically, Pavlovian conditioning, or what you can think of as cat-and-the-can-opener conditioning. You have your can-opener, the cat hears the sound, approaches, intensely excited, expecting the food. These drug cues, and they can be very varied - the people with whom you've used cocaine, the place that you've often purchased cocaine, paraphernalia lying in the gutter between our addiction center and the VA medical center, sort of like a trail, Hansel and Gretel - you basically have lots of things that have been very closely associated with the arrival of the drug effect. They are reliable predictors of the drug, just as the can-opener sound is a reliable predictor of tuna or whatever you purchase.

So after a while, these cues, through disassociative learning, acquire the ability to trigger this interesting cascade of symptoms. She describes some of them. Our patients can feel their tummy rumbling, they described the cocaine taste in the back of their throat, even though they haven't used it yet. They can smell it, though it's not in the room. Their ears may be tingly, their head is light, their heart is palpitating, they may feel a little bit of euphoria, almost as though they've had a taste of the drug, a very drug-like state, and of course, importantly, they feel the pull. And it's the pull that we wanted to understand.

So we had this thought that this big Go state that the patients described to cocaine could be one reason it's so hard to say no. So, yes, it's a rhyme, but it's not so easy for our patients to say no. It's particularly not easy for them, and we need to explain it, so our first thought about this was the Go state, and we wanted to try to be able to study it, but our patients' biggest Go states don't happen in our presence. They've never used cocaine with us, hopefully, and they haven't used it in our center, and so everything about our center is the signal for cocaine is not coming. So how do you get to study this Go state if you wanted to look at its brain substrates, how do you get to study it in a controlled setting?

So what we've really done over the years, and we did this work first with the polygraph lab, but most recently in the brain imaging setting, is have our patients become collaborators, and they actually have helped us create things that are reminiscent of cocaine out in the real world. So, we've had paraphernalia that they bring in, we have audiotapes of them talking street talk about their cocaine experiences, we have video tapes, and we use a lot of those in the imaging setting of people simulating the buying, selling, and use of cocaine. And most of what I'll be telling you tonight is about smoked cocaine, which gets to the brain in about seven or eight seconds and produces this powerful euphoria that patients liken to orgasm, but they put an exponent on it. They say it's orgasm times 10,000, or higher, if they are very hyperbolic in the way they describe it.

So they've helped us as collaborators, really, create the stimulae that would help us study the state that's so problematic for them, allowing us to ask the question, what are the brain substrates of the Go, and what would be the candidate structures. Well, I gave you hints about this earlier, and this is not going to be an anatomy lesson with a very hard quiz at the end. There will be a quiz, but not a hard quiz - no quiz.

We're not going to do too heavy on the anatomy, but I do want to mention that the candidate structures here are the brain systems that seem to be important in the direct effects of cocaine, and from animal work we know, for a very long time, that basically we have an assenting circuitry here with cells that begin in the (haha, let me see if I can see here) yes, in the mid-brain, and they have projections to regions, in the temporal lobe, for example, they stop off at the amygdala. What's the amygdala? Amygdala, little almond-shaped structures in the middle part, medial section of the temporal lobe, very important for assigning plus or minus hedonic value to incoming stimulae, saying go or no go, get the heck out of here - so, very, very important for assigning emotional valence on up. Also in the ventral striatum (that's this other little section here), very important in animals, showing that cocaine causes a very large increase in the transmitter dopamine in the striatum, and also in the amygdala, and these associated structures that may be important for the reinforcing properties of the drug.

And interestingly, in animals, it's also been shown that cues for cocaine and for other natural rewards, food, sex, also cause an increase in dopamine in this area here, the ventral striatum. So other areas that have projections from these same cell bodies that are projecting up include the orbital frontal cortex, why is that important? Well, you know, we have the caveman dilemma here. You have a region that is actually helping you to decide about the reward and the consequences, reward and consequences, and it's a higher-up region here, and it can talk to these others, these are inter-connected, to help you make the final decision to go or not to go. And on up, anterior cingular, medial prefrontal cortex, so this circuitry that ascends here from these cell bodies and goes up all the way to the frontal cortex, is important in the direct effects of cocaine, and also, potentially, we thought, in these effects of the cues for cocaine, based on the animal work and based on these patients' descriptions of this being such a cocaine-like state. So how do you study it? Well, we had a PET camera. What kind of camera is that? It's a specialized device that actually detects minute amounts of radiotracers, the radioactivity from them. And basically, what you can do is have a little bit of radioactive water that is dripped in and mixes with the bloodstream, and is essentially indexes where blood flows more. Why would you care? Where the brain is more active, more blood flows, probably to take fuel there in the form of glucose, so it's a good index of general activity. So you can ask with blood flow tracers, where the action is. It doesn't tell you the neurochemistry, which is something that you really want to know, but it does tell you where the action is, and you can see if the anatomy follows some of the circuits that I've described.

And we basically had access to a PET camera, and we were going to use the PET camera, Positron Emission Tomography, with oxygen-labeled water, radioactive water, in cocaine patients, and in patients who had no cocaine history, and let them watch our videos, watch our videos that we had made. I make my own videos, with our patients' help, of people simulating the buying, selling, and use of cocaine, and also nature videos from public television. And so we wanted to see the difference in what would happen in the brain. Our bet was that we would have some of these structures that I've described to you applauding in response to the cues for cocaine, particularly the amygdala and the anterior cingulate, very important for emotional reactivity, and all these structures holding hands, allowing our patients in the real world to both appreciate the arrival of the drug, but also, importantly, to anticipate it in response to signals, whereas we didn't think that that would happen when our patients were watching hummingbird videos, right? Most of them, central Philadelphia, not been really watching, you know, hummingbirds while using cocaine, it's not a cue, so we thought we were safe in that regard.

So the first question is, you saw what this PET camera looks like. Looks pretty innocent there, but our patients have to come in, lie down, their head is in a head holder, their arms are stiff beside their body, they're told not to move. People coming in, taking blood samples, asking them questions in white coats, and now we're hoping that the cocaine videos are going to carry the day, and that they will actually crave cocaine in these circumstances. And everything about that situation is fairly hostile, so you have to hope that you have some pretty good videos.

Now, we had been doing video work for a number of years in the polygraph setting, so I had things that had already worked there, but the polygraph setting is nothing. People are in a comfortable room, you know, basically, not all these constraints, not all the sterility and the harshness, so we had our fingers crossed - could we get the Go state going in this relatively hostile environment? because if we didn't, all bets were off. No matter how much the PET scans cost, and they cost a lot, it wouldn't make any difference if you couldn't get the state.

So, the good news is, we were able to get the state, and the state was, as Patti Davis Reagan had described it, it was cocaine-like, and that's what we were hoping for. So the patients, and they're the black bars here, described that when they were seeing the cocaine video, as compared to the hummingbird one, there was an increase in cocaine-like high feelings, they felt a little twinge of high, there was an increase in cocaine desire or craving, and a wish for rush. Importantly, and happily for me, there was not a change reported by them in the wish to get rid of bad feelings or withdrawal or how relaxed or how good they felt overall. If anything, they felt a little bit better after seeing the cocaine videos. Probably there was a reminder that cocaine was still somewhere out there in the world when they were finished with this experiment a week or so later.

But the other question then became, of course, all right, we got the state, can we see limbic activation. These were the first studies of this type, and as you'll hear in a few minutes, several of the labs have begun to do this work, but we really didn't know if we would be able to see these structures responding to these cues. Would the amygdala be applauding? Would the anterior cingulate say yes? And we asked the question and, sure enough, let me orient you to these slides, because I'll show you three-dimensional ones in a moment, but these are a little bit unusual for your daily reading.

What these are, are hatband slices through the brain at this level, and basically this is an MRI, Magnetic Resonance Image, of the gray and white matter structures, and over here are the PET images, the functional images. No, the brain, of course, doesn't really have these beautiful colors, we color-coded, and we asked it to say red is where the activity is greatest - remember how we said the radioactive water was going to go where things were most active? And basically, what you can see here, and this was very exciting to us, is that the anterior cingulate is definitely applauding when patients watch the cocaine video as compared to the nature one, and the amygdala, here it's particularly prominent in this particular patient on the left side, but actually it occurred on both sides. This is a three-dimensional view of those data. Here we're looking at the bottom of the brain, so we're bottom-up, and you can see amygdala on both sides, extending up to the basal forebrain, and here is the anterior cingulate in this slice that is just down the middle of the brain, and you can see that this is very much activated.

This was pretty exciting to us, because it was the first evidence of this kind that, basically, we could peek inside the brain and see what is happening during a big desire state. This state is big. So, it was exciting to us, and we actually - here is a graphic to go along with those pictures just to let you know that the amygdala was the winner here in terms of actually having the biggest response in the structures that we examined, and the anterior cingulate next, and the surrounding temporal pole around the amygdala. In our hands, in this paradigm, the hippocampus, which is famous in some other kinds of memory, not so much this emotional reliving, but sort of more factual or declarative memory, specific recall, was not differentially activated. After all, we weren't really making demands on people in terms of specific recall; they were describing this very much as if reliving the cocaine experience. And the orbital frontal cortex had a little bit of a curious response, which I'll talk about later. One side responded a little bit in the patients as compared to the controls; the other didn't. And I'll come back to that later.

Just a word, technology evolves, and what we've begun to do in the recent two years is actually use another technology which also measures blood flow response, but does it without any radioactivity, and this is called perfusion FMRI, and these, again, are hatband structures, hatband slices showing you structures that are activated in response to the cocaine as compared to neutral cues. And what you can see where the yellow arrows are, is that we have a lot of activation here detectable with a non-radioactive technique, and it's the usual suspects. So we have here the ventral striatum, which we're able to see in this technique, we weren't really able to ask about it. Though it's ever-so-famous in animal work, it wasn't really visible in our PET studies in a way that was comfortable for the anatomists to agree upon. We have the anterior cingulate, we have the orbital frontal cortex, and we have a structure here called the insula. You see how this little area here as the cortex is folding in bright red? it is the area that registers reports from your viscera, so in terms of feeling aroused, excited, it would be registered there and sure enough, here it is. So we have a lot of these regions that are part of this system that allows you to respond to rewards and, importantly, the signals for them.

What's encouraging for us is that since this work has begun, there are half a dozen labs now with different kinds of ways of trying to trigger craving--some are vignettes, some are videos, some are scripts- that basically have been able to show that these candidate structures, these limbic structures, these old brain structures that are so important in our natural rewards, have been essentially co-opted by these drug cues so that now they respond to cues for drugs, as though, if you think about it, to the brain, from what the brain knows, this is the biggest activation it has ever seen of this system. These drugs activate this system in a way that's far, far larger than natural rewards. You can see that from animal studies. The increase in this chemical messenger dopamine is huge to amphetamine or cocaine compared to even a very good chocolate brownie, you know? It is orders of magnitude larger and basically, the response to these cues in that same system is also large.

So we've got our little diagram up here, and I won't go through all the anatomy again, but what I will say is that it's encouraging to have that replication for cocaine, and that's really where more of the work has been done, but I would like to also summarize, really quickly, these drug cues can be used, they trigger a Go state, and you can study it in an imaging setting and you can see limbic activation.

Now, for those of you who are here for things other than learning about cocaine, what about other drugs? It's an important question because, after all, if we could find an intervention that would work for one, wouldn't it be nice if it worked on several? As it turns out, many of the major drugs of abuse exhort their effects in a final common pathway through this same system that I've described to you. They do it in different ways, which I won't go into here with time, but basically, that's an interesting thing, so you wonder, well, with opiates, for example, with heroin cues and heroin patients, would we see some of these same structures activating. And we've looked at that, we've done that a little bit with PET and we're now doing it with FMRI, and we've completed an FMRI study for - I don't have slides for that yet, but I do have some slides from the early PET work, and with opiates, and I'll mention this study which has been completed in the past year by another group. Basically, we saw orbital frontal activation, so this is the area that is trying to make the decision, here is the promise of reward, do I, don't I, and also the anterior cingulate, and in another patient the anterior cingulate.

And I will go back, really quickly, to - sorry- to the fact that we - oh, gee, I wanted to show you - that a group in England, Sel, et al, have actually shown also, corresponding to what I just showed you with our slides, that this frontal region here, down here in the very bottom of the frontal lobes, is making the decision about reward, go/no, and insula and [inaudible] are all showing some activation. So this overlap, it's not 100 percent, and the studies are still early, and there aren't enough of them, but what you can say is that there is, even now in the human data, some interesting and basically suggestive overlap between the structures that are activated by cues for cocaine, and those that are activated by cues for opiates.

And, I will say a quick word about nicotine. The nicotine work is very new. This study - we were collaborative on this study. We provided the nicotine videos that we have used in our polygraph work, and we are doing our own nicotine work with FMRI now, but Art Brody at UCLA actually used our nicotine videos and did a study with glucose metabolism, looking at the areas that would be activated, and sure enough, some of the same culprits, right? Anterior cingulate and, in this case, orbital frontal cortex correlate with craving, as well as the dorsolateral prefrontal cortex and again, the insula, places registering all these reports from your viscera.

Ah, yes. The high school group is still up there, right? You guys are still awake? All right. So, a word about natural rewards. Now, as I talked earlier, it is no accident that our cocaine patients use sexual language to describe both the high and the desire. These circuits are shared. So what about sex? We thought someone ought to do a study, you know? It was a hard undertaking, but it was really something that had to be done, and we had research assistants, we even have sometimes high school work-study students who come and work with us. They volunteered their time to go through video material and try to select things that they thought might be evocative for our patients. It was selfless, and basically, they spent hours, and we decided to try to do a study with, I use the term loosely, normal Philadelphia males, and see if these same circuits might light up in response to these carefully selected erotic vignettes from the video store on 43rd and Market.

So, a little bit about sex. So, the easy part of this story is that for about half the men, the anterior cingulate was applauding, and the amygdala, but you say well, what about the other half? Well, the other half didn't like those videos. But you couldn't always depend on what they were saying, and this is the complex thing about studying sex, I think, I've proven by now that I'm not a primary sex researcher. We basically found that sometimes the patients - patients, they weren't patients [inaudible] a lot of craving, but their amygdalas weren't agreeing, and we had other people who were saying, well, not really my cup of tea, you know, politically correct response for these triple X videos, but their amygdalas were going whoa! What do you mean, a two? This is not a two, this is a nine!

So we had a mis-match between self-report and brain, and so clearly, what we needed to do, and what we should have done, and what we will do is some peripheral wiring. So I actually - no, no - I had ordered the equipment before the study began, took the equipment a while to get in, then the question was, oh, my God, we're halfway through, to wire or not to wire. If people are wired, will they feel uncomfortable, self-conscious, more self-conscious than they already are?

Sex research is very difficult. So you might think it's just a walk in the park, but it's really, really difficult. So we thought, "shoot, we won't wire," so we didn't have a nice objective measure of arousal. I'm talking penile tumescence here, for those of you who are wondering about the wiring thing. And it's a self-placed loop, it's not me, not me, not the research - no, no, place it themselves. And I should have done that, though, because then I would be able to sort these data out and really see if I could correlate the amygdala with the actual objective response without getting the self-report in the way. The self-reports are really, really tricky.

This is just to show you that overall, the group gave themselves a five on the zero to nine, so overall the group was fine with the videos, but there was a lot of variability and, as I said, it didn't map well onto self-report. But this group really put their hearts into it. This group is a group in Montreal, Karama, et al, and it just came out this past year, and they spent a lot of time. They selected videos, they tried them with men and women, and they had to work a long time because they were selecting videos that the women felt were least disgusting. This is - so they wanted to use the same videos with the men and women, but this was a problem because, in fact, in the literature, and there's a lure, that these videos are not quite as evocative for women as for men, and so they were trying to equate for that, and what they did was a pretty good job. They couldn't get exactly to the point that the arousal was equal for the two groups, but they did end up with a set of videos that they could use for everybody without the women being totally grossed out.

And the study here was actually finding the same circuits that we've been talking about, the same structures here - anterior cingulate, the medial prefrontal cortex up here, that we talked about, the orbital frontal cortex, weighing of reward, the insula again, the occipital temporal cortices in terms of the vision and the emotional memory, and the amygdala, and the ventral striatum. Now, it is not an anatomy lesson, but you have to believe me that these are, in fact, the locations that we already highlighted. So it is a very nice study that shows a really, really strong overlap between the areas activated in response to these sexual videos and those activated by drugs of abuse like cocaine.

So, you're saying, forget this sex stuff, let's get to something interesting, like food, you know. And that work has also just begun. Most of these areas of research are just beginning because the imaging tools have not been that widely available. We have a study going on at Penn right now where they're looking at the craving for chocolate. We're collaborating on that and the craving for savory things like chicken fingers, and they basically came running over to me and said we've got to collaborate, we have amygdalas in response to the chicken finger craving. I said chicken fingers? I don't know. Chocolate I could go with, but we're working with them, we'll be able to see more what that is going to look like within probably the next six months.

But there is good animal data that shows, for example, overlap in brain regions that are activated by morphine and chocolate in rats, and there has been interesting data that shows that a drug that blocks opiate effects will also block the consumption of a highly palatable food. So rats will still eat their regular chow, but if you - rats love Cocoa Puffs, they love chocolate too, and if you give them Nalexone, it will actually reduce their consumption of a highly palatable food like that.

So there are links between food and these drugs of abuse, and one of the things that has been just done that is a really fun study, was done by Dana Small, published in Brain in 2001, was actually a chocolate eating experiment. This is the one that, you know, we wish we had done first, but they did it already, in which they had people eating chocolate, and they looked at the regions of the brain that were activated in the very early phases, when the people were just tasting the chocolate and still finding it pleasurable, and - these were chocolate lovers that were recruited, by the way, also selfless in giving themselves to science and studying.

And then they watched the brain areas that were activated as the chocolate eating continued, and it was no longer quite so uhhh, and they went all the way beyond satiety into the point that it was uncomfortable for the people, and they looked at the regions that were activated, and in fact, you have some of the same structures, it's not 100 percent overlap, but you have an interesting thing here. This region right here, again, right above the eyebrows, is this orbital frontal cortex in the medial section, was activated when people were still in the pleasurable anticipation and enjoying the effects, but as they got over into the sort of a more [inaudible], the activation moved too, and they had a nice dissociation in space of the positive and the negative aspects of chocolate eating.

So the area that we're looking at here, so far, talking about drugs and abuse, natural rewards, what have we got? Yes, the cues can elicit profound act-effective positive states of desire, Go, whether it's cocaine, whether it's opiates, whether it's nicotine, sex, or chocolate, and this can be used to study these brain substrates, and we're seeing lots of suggestions of limbic activation to the cocaine/opiate/sexual cues, and also to food.

One of the really interesting things to ask, well, once you see this signature, is there something that could be helpful? Our patients are really struggling with these desire states, they are almost overwhelming to them. Is there some way you could modulate this system? And, I won't go into this in great detail, but I'll tell you that we're very interested in the category of medications that is known to sort of tune down the volume on the dopamine system. It essentially reduces the amount of dopamine that is released in some of these critical, terminal regions of the system.

And these medications, the representative of these medications, these GABA B agonists, is called baclofen. Some of you may have heard about it. It is actually an anti-spastic, it's been used for years and years and years for paraplegics as sort of a primary therapy, but in rats, it was shown that it caused disinterest in rats who otherwise avidly took cocaine. Normally, rats train, take cocaine in their night cycle, they go over, it's dark, they hit the lever as soon as the lights go down. If they are given this medication baclofen, they would be disinterested in cocaine for several hours, they would, you know, go out for a danish, a newspaper, shower, coffee, come back, and then take cocaine. It is relatively short-acting, and I saw these data presented in '95 by Dave Roberts at CPDD, a conference on addiction, and I thought, you know, this is a medication that is already approved for humans, because it's always a struggle to get a new medication approved. Would it be something that we could try in cocaine patients?

And we have been trying it. We basically have done it in an imaging setting, pre-treating with this medication for several days, and we have also been doing a little bit of clinical work, and - let's see how our, how is our time, by the way? I don't have a watch here. I don't know if we have time for Edward's story, but if we do, he'll tell you a little bit about what happened when he took baclofen, this medication that actually tuned down the volume on this system. Let's try it and see. If it works, we'll be very tempted to do it. Let me see if I can find the cursor. Let me click on Edward and see what happens.

A.C.: You found out something interesting when you were taking baclofen. What did ...

Ah!

[Group laughter.]

Edward: [Inaudible] spasms, but I would try to use cocaine [inaudible] couldn't feel the effects of it.

Ah, couldn't feel the effects.

Edward: And then ...

A.C.: Did it sort of block the high?

Edward: Yeah. Then one day I said, let me try it without taking the medicine [inaudible] waited two days, I stopped taking my medicine for two days, and I was having bad spasms, but I was getting the full effect of coke.

A.C.: But you were doing the low dose at first, and that was 10mg four times a day?

Edward: Yeah, and ... A.C.: And then you upped the dose. Tell me why you upped the dose.

Edward: I upped the dose because I found that the 10 dose wasn't taking care of the muscle spasms.

A.C.: Right.

Edward: I would still get up and then my legs were still just ...

Increased his dose to take care of his spasms.

Edward: So I said, well, maybe if I take two of these, it'll calm it down, and it did.

A.C.: Now, when you left off the doses, either the one or two days, or even for two or three doses, what about the pull towards cocaine? Like, the basic feeling of gotta-have-it that you were describing to me. How would that change?

Edward: Well, when I was taking it, I didn't have any cravings, you know, I could be there, sitting watching TV, but when I didn't take it ...

A.C.: Right.

Edward: For them couple days with them doses, I felt myself wantin' to get high more.

A.C.: Did you ever try to get over the high dose by taking more cocaine? What happened? Did you ever try that?

Edward: I tried it, but I found myself wasting money.

A.C.: Wasting money 'cause it didn't work.

Edward: Yeah, I wasn't gettin' high like I ...

A.C.: I hear you.

Edward: Thought I would. Stopped taking them for two days to try to [inaudible].

A.C.: Yeah.

Edward: Stopped taking it, and I mean, as soon as I get that first bag, I was high.

A.C.: Again.

Edward: As I should have been getting.

A.C.: Yeah.

Edward: With the ...

So, we wondered, well, what about Edward's brain? Could we see that pre-treatment with the medication would actually reduce this activation that we had seen with cues usually, and so we looked at Edward, and we also looked at a handful of other patients, and what I've given you for reference down here is the three-dimensional scans that you saw before with activation of the amygdala and the anterior cingulate. But up here are the scans from patients who were pre-treated with baclofen, and what you can see is it takes the red out, right? But basically, there's actually less activation here, and the anterior cingulate is not activated in cues any longer.

So we're very interested in whether or not this would be a potential candidate medication that might be helpful for cocaine. Interestingly enough, the patients who were taking it for cocaine, many of whom were cigarette smokers, also reported spontaneously, and they were in the halfway house where there wasn't really much else to do, that they were smoking fewer cigarettes. And so we were very interested in that, and we're now actually finishing up a protocol to give it to our human subjects committee and to do a trial with smoking, because it hasn't been done, and yet this class of medications, these GABA-ergics, in animals it looks very promising with regard to nicotine administration, and also in terms of blunting cocaine-related behaviors.

So we're really interested in this sort of kinder, gentler way of modifying the dopamine system in terms of candidate medications. Other things have been tried, including really blocking that system, which has a lot of sort of aversive side effects that I won't go into here, but I think that this is actually a promising path.

Let's summarize the Go, and then we're going to move towards stopping. Summarizing the Go, these drug cues, yes, they elicit the Go, yes, we see amygdala and anterior cingulate, medication and behavioral tools that I'm not going to tell you about tonight seem to be able to potentially blunt the Go, but maybe you've begun to have a nagging feeling in the back of your head. Now, gosh, she's going on and on and on about the Go. Is it the whole story? I mean, after all, you're saying to yourself, we've all got Go, you know?

We all went through that same evolutionary thing and we've got a well-conserved rewards system, and we're exquisitely responsive to rewards and maybe some of us are more responsive than others. It seems a little unfair to think that someone would get more out of chocolate or sex than someone else, but that could well be true, but we're certainly not all compulsively pursuing those things. In fact, many of you are here tonight, hundreds of you, attending a lecture instead of out looking for cocaine, you know, or a novel sexual partner, it's very flattering to me, and I actually - so this is the way that I think about it, is that basically we have to explain for ourselves the fact that we all have healthy reward systems because as a species, if we didn't, we wouldn't be around very long, and most of us engage in these activities, whether it be gambling, chocolate, sex, even experimenting with rewarding drugs, without getting into trouble. Many of them do them regularly, some of them get into some trouble but stop on their own, but think about the people that we actually see in treatment, and this is going to be loud, but I apologize.

Treatment-seeking populations, clearly, it seems almost a truism, they've got big stopping problems, right? That's why they're there with us, they are a sub-group of all people who are exposed to these activities and these substances. So we had to think about that, and we had to acknowledge that craving, as powerful as it is, didn't seem to account for all the variability in outcome that we see. We got some pretty good cravers, probably in the audience tonight, and yet you're here, you're not out there pursuing cocaine, and we would all acknowledge it's a powerful reward.

What is it that makes the difference? We began to think about the possibility that some of our patients might literally have bad breaks. It's been shown, and in fact, there's a neurological literature that is sort of made a big thing about this in the last few years, that people who have lesions here in this most bottom point of the frontal lobes, right in the middle, so-called ventrimedial prefrontal cortex, seem to be impaired in taking the consequences of their actions into account, so that they are actually pulled by the immediate reward, but they are not thinking very much about Monday morning on Friday night.

You know how it is on Friday night, right? Monday is a year away on Friday night. Well, for our cocaine - I'm talking to the high school students now, I have one at home and one in college - and Friday night seems often eons away from Monday morning. But as you get older, you're supposed to be able to think a little bit about the fact that Monday morning will come and if you haven't done the studying or the preparation for your test, you'll have consequences. Our cocaine patients have a great deal of difficulty in thinking about the consequences on Monday morning of Friday night's cocaine use. Monday morning could be a positive urine that will result in them losing their children or their house or their job or their spouse, and they may have lost all those things already, and yet, on Friday night, the pull, the cocaine use, still continues.

And so it's a puzzlement sometimes, to their families, to themselves, to their caretakers, why is it that after so many negative consequences, repeated negative consequences, that the behavior continues, almost as though those consequences didn't register in some way. Well, patients with lesions in this region seem not to register those consequences in terms of making future choices. They are pulled by the immediate reward, they don't seem to take these consequences into account, and there are actually tasks that were developed to discriminate between patients with lesions here and normal controls, and stimulant abusers, drug users, do more poorly on these tasks. So it was a piece of evidence that behaviorally, they are not as good at doing a long-term conservative strategy, thinking about Monday morning on Friday night.

But what about their brains? We had patients' brains because we had actually taken all those high-resolution gray and white matter images when we were doing the PET work, and when we looked at the gray matter concentration in those patients, compared to controls, this yellow region is a region that actually is where gray matter is less dense, less concentrated in our cocaine patients. Less gray matter where they need it most, where the braking action may be, in this ventrimedial prefrontal cortex and stripping up toward the anterior cingulate.

And over here, in our PET study, we were able to look at the blood flow to that region, and it's stuck in first gear. It's lower here and in the orbital frontal cortex, and again, stripping upward. See how similar these two images are? It's as though this may be the structural defect underlying this functional defect. Wow. This is the last thing that our cocaine patients need, right? They've already got a big Go, they've been exposed to a drug that is very powerful, and now, some of them, and this is group data, actually have got less in the way of stopping ability.

So it's a bad news story, it's a difficult story. It sounds like, that in addition to the Go issues, that our cocaine patients may be compromised, at least some of them, enough to carry a group effect in their ability to put on the brakes, and we think that these defects may be really important in understanding the whole picture about addiction vulnerability.

I'd like to say a little bit more, just for the sake of discussion afterward, about these frontals, and people who are well endowed frontally and have good brakes, and people who are less well endowed. These stains are actually not from addicts, these are from men and women. And what we're looking at here, and it's a little hard to see this, I borrowed this from Ruben Gurr, who's an investigator at Penn who looks at sex differences in the brain in response to emotional stimulae, and a bunch of interesting things, and what he did was actually look at, essentially, a ratio of how much you got up front, the brakes, to the Go, if you want to think of amygdala that way, and basically, it's a ratio. So if you've got more braking tissue, literally, this is volumes of tissue, compared to the amygdala, maybe you're in better shape, and if you've got less, maybe you're in worse shape.

You're saying well, what's this men and women thing? Well, what this is, is a distribution of the ratio of your brakes, your stop to your Go, for men and women. And some of you may find this quite unsurprising. Women have, as a group, better brakes than men, and basically, men - now again, these distributions, each one of these little openings, each one of these little circles, is a person, so you can see that here in the middle ground that there are people who are very similar in their braking ability that are men and women. But there are more men down here at the bottom with a lower ratio, worse brakes, and there are more women up at the top, so that the means of the two groups are significantly different.

And you say, well, I've known this all along, you know? This is nothing new. Well, but it is really interesting to have a brain basis, potentially, for some of the behaviors that you see. And the other piece that I'll give you about this is that this ratio also, your frontal tissue here, declines with age. So now you've got a situation where you've got - we must be compassionate. Males start out with poor brakes, and things only get worse with time. Some people will feel this explains more to them than anything else that I've said tonight, but, so, but when we're talking about addiction, that's sort of an overlay, this is just normal folks, but it's important to think about the importance of these things that we see, behavior that we have a hard time explaining. He what? He ran off with his 18-year-old secretary? What was he thinking? Well, you know, so ...

So our patients, though, may have, then, therefore, two areas of difficulty. They may have throbbing, pulsating amygdalas, lots of Go, but they may have bad brakes. In the lab we call this withered frontals. And they could have both, you know. The world's not, life's not fair, basically, you could be dealt a rotten hand. And this is one way of thinking about it, the difficulty that they may have. Most of us, we've got Go. We have to get Go or we wouldn't be here, and basically, most of us are probably pretty well endowed frontally. You're here instead of out looking for cocaine. Clearly, you're inhibiting some things in order to come here tonight.

And then we have, potentially, a sub-group of people who may not be so well endowed frontally, and they may be vulnerable because they've got Go, and there's animal data that suggests that the Go may get bigger with stimulants, and that basically, your frontals may actually be eroded in their inhibitory function. They have been able to show with animals that if you give them continued exposure to cocaine, that they are poor at tasks that require them to inhibit something that they've previously learned. So it's a bad situation, you go from bad to worse here.

One of the interesting questions, of course, is well, you know, did these people, you know, you see them just after they've used cocaine for all those years. Did it start out that way? Was it this way? How could - from our one-shot studies, you can't tell that, but it's a very important question to ask, and you could ask it in populations where drug abuse is not even an issue, right, for gambling or other things, you could look for these kinds of measures and look at it neuro-psych-wise and look at it brain tissue-wise.

So, in closing here, I just want to say that whether you're thinking about opiates or brownies or sex or cocaine or gambling or other things that you'd like to share with me in the discussion session, that basically, as you move from desire to disorder, I think that you have to think about both these systems, the throbbing, pulsating Go, but, very importantly, the stop. And I'll stop. And this is the lab, this is actually a large endeavor. We have a number of dedicated research assistants, you've heard about their selfless activities in screening our videos, clinical staff that are wonderful, substance abuse fellows, investigators, consultants, it's a large group of people who have put in a lot of effort on this work, and we're also well supported by the National Institute on Drug Abuse, your tax dollars. Thank you very much.