Receive regular updates via email

The neurophysiology of gambling

If you read my previous post on the neurobiology of risk, you may recall that the ventral striatum regulates risk and rewards. In most people, thinking about winning money increases dopamine in the ventral striatum, and thinking about losing money decreases it. This is where gambling comes in.

This excellent post on the neuroscience of gambling describes a series of experiments done on monkeys that show that dopamine neurons learn when to expect rewards. Dopamine increases when the reward comes, decreases when it’s supposed to come but doesn’t, and wildly increases with unexpected rewards. And unexpected rewards are the principal attraction of gambling.

Instead of getting bored by the haphazard payouts, our dopamine neurons become obsessed. When we pull the lever and get a reward, we experience a rush of pleasurable dopamine precisely because the reward was so unexpected. (The clanging coins are like a surprising squirt of juice [for a monkey trained to expect juice as a reward]. It’s operant conditioning gone berserk.) Because our dopamine neurons can’t figure out the pattern, they can’t adapt to the pattern. The end result is that we are transfixed by the slot machine, riveted by the fickle nature of its payouts.

Thus, the unexpected reward essentially makes the ventral striatum very happy, and it can’t figure out how to become very happy again except by continuing the same behavior that led to it before. And there you are, two hours later, out of money at the blackjack table because you were waiting for that rush.

However, most of us have self-regulating systems in the brain that will eventually tell us that logically, we can’t spend all our money chasing the hope of an unexpected reward, that this isn’t enough of a reward overall, and we get into our cars or onto our flights from Vegas and leave the gambling table and its ventral-striatum-enticing allure behind.

But in pathological gamblers, the ventral striatum doesn’t react the way it’s supposed to (perhaps because the unexpected reward becomes expected?). Riba, Kramer, Heldmann, Richter, and Munte (2008, PLoS ONE) gave volunteers dopamine-increasing drugs and found that not only did the subjects make riskier choices, but parts of the basal ganglia and midbrain, which are important parts of the reward system in the brain, showed decreased activity after unexpected rewards.

As it happens, the dopamine-increasing drugs Riba et al. used were intended to treat Parkinson’s disease. It’s known that such dopamine agonists, as they’re called, can trigger pathological gambling behavior. Riba et al. suggest that pathological gambling–at least in Parkinson’s patients–comes from a need to overcome a dulled response in the reward systems of the brain.

This is similar to the concept of drug tolerance, where regular drug users must use more drug than they previously did in order to get the same effect because their system has become less sensitive to the drug. Essentially, gambling addicts are like any other addict: they don’t get the normal feelings that non-addicts do when pursuing their behavior of choice, so they do it longer, harder, and more in order to achieve the reward they crave.

The neurobiology of risk: the posterior cingulate cortex

The cingulate cortex sits on top of the corpus callosum, the thick cable that connects the two halves of the brain. It’s connected with the amygdala, which coordinates perceptions of feeling and emotion, and divided into the anterior and posterior parts. The anterior cingulate cortex has been implicated in effortful decision-making (Mulert et al., 2008, Neuroimage), and the posterior cingulate cortex (PCC), relatedly, in decision-making and risk.

Decision-making and risk are important parts of the animal world; any organism needs to know whether a particular activity is going to get it killed, and decide whether that activity is worth it. Watson (2008, Annals of the New York Academy of Sciences) suggests that sensitivity to risk helps animals survive. The PCC hasn’t been studied much until recently, but it’s possible that it’s the center for risk-related brain activity.

Watson found that the PCC in monkeys showed sensitivity to risk, and its strength, in decision-making tasks. McCoy and Platt (2005, Nature Neuroscience) found that PCC in monkeys activated when monkeys made risky choices, and became more active with more perceived risk.

This article describes how the PCC judges value of rewards as circumstances change. Researchers (Platt et al., 2003) trained monkeys to do a task and rewarded it with juice, so that the monkey learned to expect the juice when it delivered. When the monkey performed the task but got no juice, the PCC fired very strongly, giving what the researchers described as a large “reward-prediction error,” a comparison of a predicted reward with the actual result. This isn’t specifically risk-related, but it does demonstrate that awareness of rewards and their absence is important–otherwise there wouldn’t be a segment of the brain devoted to it. And risk is all about evaluating reward.

Perhaps the main thing to take away from the neurobiology of risk is that it’s inextricably tied into emotion. The PCC, like the ventral striatum, contains dopamine-releasing neurons, and it’s also part of the limbic system, which regulates emotion and motivation. And fear, which may be particularly important in the PCC–learned fear may be stored there–regulates risk-taking behavior and perceptions of risk in general. This, too, makes sense; it doesn’t really matter what the objective risk of a given activity–darting into the open to get that delicious plant, say, or using $100 to buy stock in Google rather than putting it in the bank–actually is if you don’t have some emotional investment in the outcomes.