Judith Grisel, both a neuroscientist and recovering addict, is pretty well positioned (to say the least) to give an overview of drug addiction that ties together the underlying neurological basis with the actual subjective experience of addiction. Needless to say, this made for a really interesting read.

She covers most of the different major classes of psychoactive drugs and their mechanism of action (so far as we understand them now) but the really interesting part is how this fits into a sort of “grand unified theory” of addiction.

The gist is that our brains have control systems that try to maintain homeostasis around certain “set points”. If you perturb that system by, say, ingesting a drug which floods the brain with a particular neurotransmitter then there is a compensating action which tries to bring the brain back to it’s set point, just like a sudden burst of warmth would trigger a thermostat to turn on the air conditioning. Now, if you ingest said drug repeatedly then the brain will actually lower the set point so that the effect of the drug will result roughly in the original, pre-drug abuse “normal” state.

In practice that means that whatever the effect of a drug is on our subjective experience, repeated use will result in our brain moving the baseline sober state in the opposite direction. For example, opiates work by mimicking endogenous neurotransmitters which bind to particular receptors and block pain signals (as well as induce a sense of calm and euphoria). But our brains also have endogenous anti-opiate neurotransmitters which bind to those same receptors and have the opposite subjective effect: amplifying pain signals and causing anxiety and dysphoria. When you flood your brain with opiates, the natural response is to increase the amount of endogenous anti-opiates to counteract the opiate effect. And if you frequently ingest opiates, your brain will up-regulate the baseline amount of anti-opiates. The result is what we see in opiate addiction, where addicts need their fix to feel normal and their baseline state is agonizing pain and dysphoria.

So far this all make sense but I was left with a few questions:

The way this is organized conceptually is as if there is a control structure outside of the brain, but that is not how it actually works. The brain is a self-contained system and whatever mechanism is maintaining homeostasis is mediated by the same biochemical processes which mediate everything. So why don’t drugs also mess with the regulatory functions of the brain? Or is it just that the particular drugs we have now don’t? Would it be possible in principle to create drugs (or cocktails of drugs) that have the intended effect but also block the homeostatic response that leads to addiction?

Along those same lines, there seems to be a tremendous amount of individual variation in addiction. Some people have “addictive personalities” and seem to have an immediate compulsion to use drugs soon after their first experience while others seem to be able to use drugs, even highly addictive drugs, on an infrequent or opportunistic basis and never have a strong compulsion to increase the frequency of use. Do the people prone to addiction just have a stronger homeostatic response?

I would really love to see how this model fits into computation models of neuroscience. How would Karl Friston or Andy Clark describe addiction?