University of British Columbia's Dr. Luke Clark discusses his research on slot-machine gambling, how our environments can affect our genes to play a role in forming addiction, and the function of dopamine in driving our behaviour. He also explains how our expectancies, or associations, can assist in preventing or encouraging problematic habits.
Tracy: Addiction, in general, is a complex and deeply personal subject. Today there seems to be a divide within professional circles between those who believe that solely genetics cause the disease and those who believe that it's only caused by the social environment. What is your take?
Dr. Clark: It's a curious movement. We talk about it a lot in the lab and try to get our heads around it. Anyone who works on brain systems of reward or [looks into] what dopamine does--well, to me, it seems really obvious that the brain system is going to be affected by social context. There are social and environmental risk factors for all sorts of problems and certainly for addictions. But you see this in other areas of mental health research as well, where a lot of people assume that you either have a neurobiological model of what's going on, or you have a psycho-social model of what's going on. You've either got brain factors, or you've got environmental factors. To most of the psychologists that I speak to, [they] view these things as being completely complementary and not at all at odds with each other. So I find the whole argument sort of vexing.
It isn't one or the other; there shouldn't be a divide between the two camps of each idea.
I teach intro Psych[ology at UBC] from January to April each year, and the students start asking [about this] pretty much from the first week once you get into the nature vs nurture discussion. "Okay so what is it then?" [I often tell them there's] a bit of a disappointment ahead, because everything is a bit of each. Any behavioural disposition or trait that has been studied ends up being about 30 to 50 percent genetic with the rest environmental. This works through gene-environmental interactions. If you want to get into the mechanism of how a gene-by-environment works, [there is] this new area called epigenetics that even psychologists don't quite understand. But it seems to provide a mechanism for how the environment can change the gene expression. That's the reality, yet it doesn't seem to come full circle to [debating] the balance of those two factors in most mental health problems.
Your work focuses on the psycho-physiological factors that underpin gambling. Can you give a brief explanation for why it is that, for example, even if someone is thousands of dollars in debt, they still can't stop going to the casino even if they tell themselves not to?
That really gets to one of the things that has always fascinated, and continues to fascinate me about gambling. In a way, the more I study gambling, the more puzzling it becomes. Gambling games have [what is called] a house edge. This is what an economist would call a "negative expected value". The casino or the slot machine is taking a proportion of the amount that is bet and only gives a certain amount back. So the longer they play--purely based on maths and law of averages--the more destined they are to lose money. Now when I talk to gamblers, including lots of problem gamblers, they completely know all of that. They actually know a huge amount; they're not naive and they're not uneducated. They spend a huge amount of time reading about gambling so they know exactly how house edge works. Yet the behaviour persists. As you say, when someone's own personal experience has taken them to the point where they're hundreds of thousands of dollars in debt but they're still plugging away trying to chase losses, [that's] when casual gambling becomes problem gambling. If there's one symptom that people single out as the tipping point, it's that chasing of losses.
So then how do we think about that in psychological terms? People in the gambling field, like Bob [Dr. Robert] Ladouceur in Quebec, who is an authority in this field, call this double switching. It's as if there's a switch in the gambler's brain: when they're at home in their kitchen, they're completely cogent about the game and they understand all of this. But when they enter the game itself, the switch essentially gets "flipped." They go into this hot, intuitive, emotional, risky way of thinking. All of their cognitive processing falls by the wayside and [their thinking] becomes much more distorted. For about 10 or 15 years now, I've been interested in cognitive distortions within gambling in terms of how the gambler mis-appraises the game. You can think that runs of outcomes is telling you that something is about to happen, like a row of losses means that you're about to win, or that a run of reds means it must be black next, or thinking that being able to choose something or throw a ball is giving you some sort of skill over something that is still determined by chance, and those kinds of effects. A lot of those distortions mainly happen when players are "in the game".
When that "switch" has been flipped.
Yes. And there are really two things that might be going on there. One is about the arousal we know gambling has: that big physiological component. It releases stress hormones, it gets your heart rate increased and so forth. For years people have talked about the "thrill" of gambling as being what gamblers are more hooked on rather than the winning of the money itself. In the same way that if you're a stockbroker, you don't want to be making decisions if you've just had a big bust-up with your partner, it's the same where emotions affect all people's decision-making. So maybe this arousal affects gambler's decision-making once they're in the game. Or maybe it's much more. Now you mentioned [B.F.] Skinner earlier, which I see as the other side: conditioning. The Pavlov side, where when you go into the venue itself [and] just the sights and sounds and all these other conditioned stimuli. They might throw that switch.
So either the emotions distorting your thinking or the cues/triggers distorting your thinking. To the first point, what is your opinion on the effects of the nervous system, where, when in a fight-or-flight stress mode, your rational decision-making is overridden by your impulsive urges.
Yes, I think that sounds quite credible as a model here. It's similar to what we think about as a compulsion. A compulsion is a response that initially served a purpose but then the world changed so that it is no longer an adaptive response. And yet the response continues despite that change in environment. In OCD, we know it's useful to check that the oven is turned off once. You don't need to go back to check loads and loads of times. Going back just 10 years or so in the literature, impulsivity and compulsivity, which are both relevant things to addiction, were basically smeared together. 10 years ago we didn't have great definitions of compulsivity. That's changing recently (but more through animal research than human research) with the discussion of the habit system in the brain. Initially yes, rewards are involved, which gets the system working, but as you repeat that behaviour, brain activity shifts and it becomes habitualized. Then it's basically a compulsion. We have quite good tasks [research tools] for measuring that process in animals, and within the animal models of addiction, that all looks quite compelling. But of course we're trying to understand [human] addiction so we need to get some evidence for this in patients, but that becomes much harder.
Journalist Charles Duhigg helped to popularize that MIT Habit Loop research. He wrote The Power of Habit which details the cue-routine-reward loop and applied [that animal research] to human stories. The initial cue releases dopamine (which intensifies your drive) toward something rewarding (meaning it releases endorphins). The cue might be seeing the casino lights when driving by. The routine would be turning into the casino parking lot, walking in and putting money on the table. And the reward would be winning--or maybe near-misses: loosing, but with excitement to bet on the table again.
Yes, we know without question that dopamine is very much involved in learning about cues. That is clear. But if you think about this in terms of gambling, I suppose I would think about a slot machine where the first time you start playing, you've never seen one of these devices before, and initially, you're interested in the wins. Every so often, when you do this thing, you win money. The winning money would be the “unconditioned stimulus” and the spinning reels becomes the conditioned stimulus. The brain quickly [learns] everything that happens before: the spinning of wheels always precedes the winning of money. From what we know about the dopamine system, once the animal has learned all this, the actual reward itself doesn't really drive the dopamine systems as much anymore. It becomes more about [the] cue and less about the reward. That's what gambling looks like in terms of money, which seems to be quite important at the beginning. Presumably, the money must have something to do with it: they start playing because they want to win some money. But when you talk to problem gamblers, the money no longer seems to be the big issue.
There's a really nice book in the gambling field published a few years ago by MIT's Natasha Dow Schüll on the evolution of slot machines in Las Vegas. Her big point is that what [gamblers] want is complete immersion in the game. She talks about regular slot machine gamblers who just want to go to another place where they forget about all their troubles. She has interviews with heavy slot machine gamblers who say when they finally get that big win, they actually find it frustrating because the big win pulls them out of that zone they were in. That's a massive transfer of what the reinforcement is in those cases.
In regards to mental health and addictions, I read that you were doing some research on whether or not depression played into addiction.
Yes, I guess these things are not too linear. What I've been interested in since grad-school is how the brain makes decisions and how that process goes awry into mental health problems. I started off working on bipolar disorder and looking at mania: whether during manic episodes, bipolar patients make a lot of risky decisions. Can we use psychological tests to capture that risk-taking? How does it change throughout the faces of the illness? I still do a bit of work on suicide with a group in Pittsburgh. We're interested in whether you can view a suicidal crisis as a breakdown of decision making and how you prioritize short-term vs long-term or, future, rewards. We're trying to think about suicide in a behavioural economics perspective effectively.
I got interested in addictions since you can classically see addiction in terms of short-sighted decision-making where you focus on the immediate high and neglect all the long-term consequences. Until about 2004 I hadn't really thought about gambling much at all, other than as a task for looking at risk-taking. Around then the UK government changed their gambling laws for the first time since the 1950s with a general liberalization. They essentially had a funding initiative to build up some gambling research within the UK at the time, which was quite small really. The first round [was] where I got on as a researcher and that funded the work we did on near-misses.
Can you briefly explain near-misses for those of us who are unfamiliar?
Yes, this shows that when you loose on a slot machine game, it looks somehow close to a win which drives the same brain response that people show when they really win. In a way, the reward system responds inappropriately - it's almost like it's kind of tricked into responding by these slot machine outcomes. These outcomes don't cost the slot machine designer anything. To put lots of these events in the game [is cheap], but slot machine players keep playing more when they get those games. They drive brain activity in the same areas that respond to the wins.
And back to your work in this area of research.
The gambling field is a very young area of research. You hear people often say that gambling research is about a decade behind alcohol and tobacco research, both in terms of what we know about the brain and also in terms of public policy. There's a lot of research that still needs to be done. It's a small area and it's an area that is sufficiently close to public policy. Public policy is 'eye-on-the-agenda', so research that we do can more or less directly shape decisions that governments are making and psychiatric classification, which is not something I've often experienced as a researcher.
Do you believe there is something--whether it's genetics or environment, or both--that predisposes people to become addicted? Where, even if you listen to society telling you that drugs are bad and you choose to never do them, that you can still find yourself addicted? A lot of times this can appear as problematic gambling, or shopping, or overeating, or video gaming.
Yes, I think there's a disposition. This is a full distribution of risk. We all have some degree of risk: we might be only at the 5th percent or we might be at the 95th percent, but there is a distribution of risk. That distribution conveys risk to all of the drug addictions; it is also the same risk for gambling problems, and it's very much linked to attention deficit disorder, and likely a lot of other kinds of risky behaviours like bungee jumping, and adrenaline type behaviours.
I find the [predisposition] evidence completely compelling. Look at prospective studies for example. You're taking a group of teenagers and measuring how impulsive they are (or some form of risk-taking tendency) on a task. You have a large group of thousands of them; you follow them all up and five years later you see which ones develop addictive problems. In those studies, very, very clearly, high impulsivity predicts a whole constellation of addictive, risky behaviours further down the line. That for me is cast-iron support for a disposition. The playing field is not level. Wendy Slutske [at the University of Missouri] has gone back to age three in toddlers. They talk about temperament as the foundation of personality and [saw that] disinhibited and impulsive temperament in three-year-olds predicts problem gambling and other addictions [up to] 30 years later.
What degree do negative and positive association towards an addictive activity or substance have in the early stages of addiction? For example, one person who has a disparaging association with gambling so initially doesn't care to step into a casino, versus another who views it with a cool, movie-like association so is excited by a place like Las Vegas.
What you're saying there is what I would call expectancy theory. There's loads of evidence for that in terms of alcohol. Teenagers and young adults have quite idiosyncratic expectancies about what alcohol is going to do to them, as opposed to about alcohol in general. One teenager might think it's going to relax them and make them a bit more sociable and another teenager might think from their upbringing that it's going to make them aggressive and abusive, while a third teenage might think it's just going to make them fall asleep in the corner and bring the evening to an early end. And yes, the expectancies that you have affect your risk of going down that path. If you have positive associations, statistically, people would talk about those expectancies as mediating the link from risk to development of a problem. There's been some work on gambling expectancies (but nowhere near as much as on alcohol) and the work shows that it looks pretty much exactly the same. For some people, early exposure to gambling is actually going to prime a negative association or expectancy, which could end up being quite protective. But all of those prospective studies are in big samples, so on average, it seems to come out that there's more risk there than there is resiliency.
Now advertising massively feeds into that, which is why in a lot of countries including in the UK, there's been a big expansion of gambling advertising. It's not too bad in BC, but if you watch sports here, there is some advertising.
Dr. David Kessler brought this up in one of his books in regards to smoking. At the FDA, he wanted to make smoking less of this sexy, Hollywood thing, and more so a yellow-teeth and illness causing thing. So they put in regulations to change the way smoking was advertised to help switch beginner's initial associations.
Yes! Particularly, for problem gamblers, gambling adverts are showing direct cues. Adverts create a craving state and potentially trigger relapses in problem gamblers--likely all the time, realistically. There's a slight difference in BC because province-run gambling (basically a monopoly approach on gambling), means that the adverts you see are only going to be BCLC adverts. In other jurisdictions where you've got privatized gambling, you have a very competitive marketplace. In the UK there are [about] 5 major chains of bookmakers who are in competition and out-advertising each other. I know Australia is clamping down on [gambling adverts]. Queensland, one of the Australian [states] has recently banned gambling adverts in sports before a watershed time.
What are you most excited by looking at your past research in terms of policy?
One of the main lines of work that we've done is on slot machine gambling specifically. This is a similar question to other addictions as well: are some drugs are more addictive than others? People intuitively say, 'Yes', but when you try to rank them and bring data to that, it gets very controversial. There's a Lancet paper from about 10 years ago where Dave Nutt tried to devise a Top 20 of addictive drugs. [We can] classify some drugs as being more harmful and probably more addictive than others.
You can think about the same questions in terms of gambling. State lotteries, like Lotto 649 don't seem to be that addictive. You never see a problem gambler who is addicted to playing Lotto 649. What's at the other end of that spectrum? Generally, wherever you look, slot machines seem to be at the other end. A lot of the casino floor is devoted to slot machines now. If you look in treatment centres, a lot of problem gamblers seeking treatment have problems with slot machines. There are very similar, very complicated, controversial issues attached to that in the same way as with drugs. But my take is that slot machines can be--and have been--engineered. [Especially] over the last 20 or 30 years, they have become much more sophisticated. We have real slot machines in the lab and we do experiments where we try to unpick the different psychological ingredients of those games. We're trying to work out if particular features, like speed, underly why they seem to be more harmful than other games.
We're also getting very interested in online gambling which is actually a very contemporary theme. There are a lot of concerns whether that kind of availability to gamble in the middle of the night, for example--or in your bedroom whenever you want--if it'll draw people in who might not otherwise go to the casino. The worry is if we would see some increase in problem gambling. But the research on that doesn't quite hold up across other areas. In the work that has been coming out recently, there's a lot of chicken-or-egg issues there. But it doesn't look like the online [gambling] is really the cause. We know that problem gamblers like to gamble across lots of different formats, and sooner or later they start gambling online. But the better studies don't look as if the online fired them into having a problem.
There are a lot of policy aspects here. One of the things with online data, which you don't have from most other forms of gambling, is that all of those clicks on the website are saved and recorded. So an [online] gambling provider like BCLC may be able to look at that data and identify who are developing problems. If they can do that. This gets into quite complicated statistical modelling and predictive analytics trying to identify who in the online sphere is at risk. But if you can do that, there is a lot of potential for intervening or messaging to those people to try to get them services.
Do you see hope in intervention programs?
Yes, I think BC is pretty advanced in that sense: treatment for people whose gambling has become out of control. We have self-exclusion programs where you can ban or self-exclude yourself at casinos in BC, and pretty much the world over. You're barring yourself from all of the gambling venues in the province. We need that. Importantly we have psychological treatment services that are free and run at short waiting lists. These offer group and personal treatments which are effective. They involve a bit of cognitive behavioural therapy, usually a bit of financial counselling/debt management, and often some family therapy as well. That is a good program. In a lot of US states, they still just have Gambler's Anonymous (GA), and that's it. So I think we're in a reasonably good state.
I suppose somewhat uniquely, with online gambling, we do at least have all the data in one place and if someone can process the data and work out how to predict who [might] develop a gambling problem, there is a real opportunity for prevention there. So I think it's pretty cool that we got into that work.
Well thank you for the tremendous work that you do Dr. Clark. It was a pleasure to speak with you.