#3: Reward and Learning

The most important part of addiction for the addict is the phenomenon widely known as craving. It’s the hardest part about recovery, so it’s what we need to pay the most attention to when we’re thinking about treatment. What I’m about to describe is what occurs in the brain before the phenomenon of craving is ever consciously experienced. I hope that it will shed light on the addict’s situation, and help us rethink some assumptions about the moral status of addiction.

In the last post, we saw that genes explain some vulnerability to addiction. However, genetic vulnerability doesn’t fully explain why some people become addicts and others don’t. I’ll explain here how addiction begins in the reward system of the brain, and how the brain learns to be addicted.

Me, myself, my amygdala:

There are two major aspects of the brain: the Frontal Cortico-Striatial Circuit (frontal cortex) and the Amygdalar-Cortical Circuit (midbrain). The frontal cortex does all the executive functioning. It cares about the consequences of your actions. It cares about being socially appropriate. It can pay attention, make goals, solve problems, and say no when it ought to. It remembers how badly things turned out last time you did that stupid thing, and plans to avoid such behavior in the future. Notes to self are posted in the frontal cortex (note to self: don’t use actual thumbtacks for this). It is in charge of all moral reasoning.

The midbrain, however, does not give even one fuck about your future. It is entirely socially inappropriate (not wearing white shoes after Labor Day egad! inappropriate, but sniffing people’s crotches inappropriate). It’s impulsive and does not do stuff like “think.” It is in charge of survival activities (food, fight, flight, sex), and only for the next 15 seconds.

It’s easy to caricature the midbrain as a stupid Lizard Brain who sometimes makes you hump people’s legs inappropriately. But it recognizes danger and makes you do things that aid survival, like eating when you’re hungry, and being afraid of bears. Even your high-falutin’, holier-than-thou frontal cortex can see how this has prevented your death a thousand times this week.

Addictive substances don’t work in the frontal cortex. They work in the midbrain. They overload your midbrain with the message that they aid survival. For instance, we know rats will press the Heroin Lever instead of the Snack Lever until they’re dead. Why would they do that? It’s not because they did some long-term moral reasoning, weighing the relative merits of drug use against negative consequences. Rats and mice don’t have a developed frontal cortex, so they aren’t making moral choices and yet they can become addicted. That’s because 1) drugs are really good liars and 2) drugs work in the midbrain, not the frontal cortex. So how come some people can use a substance recreationally, like just on Fridays or when the Cowboys are on, and others end up addicted? CHEMISTRY SCIENCE

The midbrain ranks and “rewards” survival-related activity. All survival-aiding behaviors are reinforced when the midbrain releases dopamine. Dopamine is often described as a neurochemical of pleasure. But it’s better described as a neurochemical of salience: dopamine tells your brain that whatever you just did is good for survival. More accurately, dopamine signals that predictions about what behaviors and substances aid survival are correct.

In this way, the reward system in the midbrain isn’t about liking something; it’s about wanting. You should want to do things that aid your survival. Dopamine reinforces that wanting. If you’re starving, high-fat-content foods are better, right? That’s why you get more dopamine from high-fat foods. So, you want them more. This makes it more likely that you won’t starve. Sure, you might also get too much fat, but your midbrain doesn’t care about that. It just wants Cherry Garcia, like, now.

Addictive substances and activities (shopping, gambling, etc.) tell your midbrain that they are better than expected. The midbrain ranks them as highly salient to survival, and reinforces them with huge surges of dopamine. More dopamine = more important = more wanting. The drug or behavior climbs the ranks of survival imperatives until it becomes the most important thing, more important than eating or fighting. An addict’s drug becomes survival itself.

But this alone doesn’t explain why the addict keeps on seeking out the drug or activity involved when non-addicts stop. We’ve said that it’s not about liking, it’s about wanting. Why do you keep wanting? Apart from the initial dopamine reward-signal, the drug or activity is reinforced through memory.

Memory and Learning: Why do I remember every word to “Total Eclipse of the Heart,” but not my own goddamn phone number?

We’ve said that the midbrain only pays attention to the next 15 seconds. It’s the frontal cortex that handles memory. This is where the neurochemical glutamate comes in. When the midbrain releases dopamine, the frontal cortex pays attention and marks it as a memory worth keeping by releasing glutamate, which solidifies that memory and lets you know you want that thing again. Over time, this is learning:

This activity is soooo rewarding! (dopamine) –>Hells to the yeah, let’s do this again sometime! (glutamate) –> [repeat rewarding behavior] WOOO (dopamine) –>Let’s totally remember to do this as much as we can! (glutamate) = LEARNING

Glutamate associates memories with the salience-reward of that drug. Over time, this creates a feedback loop that doesn’t go away. That’s good for things you actually need to survive. But for an addict, that’s one of the feedback loops that gets broken. There’s something wrong with how the midbrain interprets the drug, and its use is rewarded inappropriately, due in part to genetics. With some drugs, genetics don’t even have to be involved–anyone who uses those will become addicted eventually because it overloads the midbrain so effectively. Meth is particularly good at doing this very quickly.

So, all memories associated with an addict’s drug are preserved as cues to use that drug again. Remember barbecuing with friends with a drink in your hand? The beads of perspiration on a glass of white wine? The first sip of beer at 5:30 after a bad day at work? These are evocative memories even for the non-alcoholic. Over time, every single one of those memories starts up that learning-loop again. Just noticing that it’s 5:30, your memory says “hey it’s time to drink…FOR SURVIVAL” and your brain gets all excited and rewards you for even thinking about it. 5:30 will nearly always make you think about a beer, and you’ll always remember it as being better than it really was.

Addictive substances and behaviors damage the brain’s ability to judge their effects and to make choices regarding the use of that substance/ behavior. Remember, choices happen in the frontal cortex. Drugs act on the midbrain. Combined with genetic vulnerability, this information leads to three crucial consequences for our thinking about addicts:

First, for the active addict, the drug is survival. The addict will always remember the drug as being better than expected. Anything that is an obstacle to using the drug cannot be associated in the memory with using the drug, because doing so would impede survival. So, it’s not that an addict refuses to look at the negative consequences of her drug use. That happens in the frontal cortex. Drugs act in the midbrain. So, no matter what, her most powerful memory will likely be how good the drug is, every time she’s around it. Addictive drugs specifically make that happen.

Second, you probably understand now why the addict in recovery cannot ever go back to the drug, not even casually. There is a great deal of neural plasticity, yes, so we can adapt to new ways of living. But the pathway that gets carved out by an addictive drug over time doesn’t go away. Those drug cues and memories will always be there, every damn time. An addict cannot just go back to casual drinking after a month or two of abstinence, for instance—she can’t just “cut back” in the long run. She’ll end up right back where she was, because the brain does not forget what aids its survival, and drugs are really, really good liars–they short-circuit the brain’s ability to associate anything with them other than survival-salience.

Finally, and most importantly, all of this happens in the addict’s brain before any negative consequences arise. By the time an addict starts experiencing negative consequences, the brain has already “learned” the drug’s salience to survival. Using or not using the drug isn’t even open to moral reasoning by the time the addict is addicted. Moral reasoning, choices, and planning happen in the frontal cortex. But, and I feel like I should say this in bold, drugs act on the midbrain. Rats don’t reason through the moral consequences of pressing the Heroin Lever. They just press it. So, understanding this, we have to rethink where we assign moral responsibility for addiction. It’s definitely not the first time using a drug. It’s not the second. I’d venture to say there’s no choice involved at all, at least not in getting addicted.

But that doesn’t explain why some addicts are able to stop. Why can some people resist cravings, but others can’t? Doesn’t this mean that the addict who quits is somehow morally superior to the addict who doesn’t? Maybe the person who quits “wants” to quit more than the addict, and that’s where we can say there’s a moral superiority. But we haven’t even gotten to the level of choice yet. Hang on for the next post, where we’ll finally get to a part of the mechanism of addiction where we have some conscious awareness: stress. Go here for Part 4.


Hyman, S.E., Malenka, R.C., and Nestler, E.J. (2006). Neural mechanisms of addiction: The role of reward-related learning and memory. Annual Review of Neuroscience 29: 565-598. http://www.annualreviews.org/doi/abs/10.1146/annurev.neuro.29.051605.113009?journalCode=neuro

McCauley, K. (2009). Pleasure Unwoven. Film. The Institute for Addiction Study.

Robinson, T.E. and Berridge, K.C. (1993). The neural basis of drug craving: An incentive-sensitization theory of addiction. Brain Research Reviews 18:3, 247-291.

3 thoughts on “#3: Reward and Learning

  1. Again, you never cease to amaze me with your insight to this disease and your ability to relate it in such a concise and witty manner.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s