Is obesity a drug addiction?

Next year the psychiatrist’s bible, the Diagnositic and Statistical Manual (DSM) will be revised into its 5th edition. It may have a dull moniker, but the DSM is more than a medical handbook. Central to its construction are the prevailing winds of societal attitudes – consider that as late as the 1973 homosexuality was still labeled a mental disorder.

These influences are illustrated by recent calls that obesity should be included in the DSM. Proponents of this position point to evidence that obesity research, both in animal models and in humans, implicates neural pathways involved in drug addiction. Drug addiction is in the DSM, so why not food addiction?

The “disease model” of drug addiction is that drugs hijack the midbrain dopamine system. Different drugs get to dopamine via diverse molecular pathways, which probably underlie their distinctive subjective properties. But all studied drugs of abuse lead to dopamine increases in the ventral striatum, one of the principal targets of the midbrain system.

Our current understanding of this brain circuit is that it is exquisitely sensitive to “prediction errors”. Imagine that you start visiting a new coffee shop on the way to work. On your first visit, you have neutral expectations of how the coffee will taste – it might be good, it might be bad. On that first morning, it tastes good, leading to a positive “error” signal with which to update your expectations. On the next morning, it’s still good, but that’s what you expected, hence no error signal.

A large body of research has linked the midbrain dopamine system to calculation of these prediction errors. When you taste the coffee for the first time, there’s a positive prediction error, and a dopamine spike.

Now, imagine that on the second morning (when there should be no change to your expectations) an evil (or friendly, depending on your outlook) scientist stimulates your midbrain dopamine system just at the moment you take that first sip of coffee. It will seem to you as though the coffee tastes better than it should have done. Various drugs, from cocaine to nicotine, may act in this way, hijacking the natural dopamine response and leading to a series of positive prediction errors, even when there’s nothing to be learnt. Over time, this may result in long-term plasticity in the striatum and other targets of the dopamine system, sometimes leading to addiction.

That, in an over-simplified nutshell, is the dopamine hypothesis of drug addiction, one that receives increasing support from neurobiological studies. Its application to obesity relies on one further logical step. Drugs hijack natural reward circuitry, so why shouldn’t potent natural rewards to do the same? As Marc Lewis puts it in his recent book Memoirs of an Addicted Brain:

 Whether in service of food or heroin, love or gambling, dopamine forms a rut, a line of footprints in the neural flesh.

There is some truth to this well-turned phrase. Sugar rewards given to rats result in dopamine spikes similar to those seen in response to drugs, and cocaine and methamphetamine can suppress appetite by interfering with the reward system. There is evidence of dopamine receptor changes in obese individuals similar to those seen in drug addicts, and such changes correlate with body mass index.

But evidence of food addiction contributing to obesity does not imply that obesity is equivalent to food addiction. Obesity is the consequence, not the cause, of an unhealthy balance between eating and exercise. Brain mechanisms involved in the processing of reward and satiety will no doubt be involved in setting this balance, and may be more involved in some cases than others (such as Binge Eating Disorder). But so will a wide range of social and economic pressures, as indicated by the graphic below. In 2009, 36% of adults in the United States were obese, and this number is rising. If obesity were to make its way into the DSM, a huge swathe of the population would be handed a psychiatric disorder overnight.

Reproduced from Ziauddeen et al. (2012), Obesity and the brain: how convincing is the addiction model?

Fortunately, wisdom is prevailing, and this is unlikely to happen (binge eating, on the other hand, will probably be added). There is an urgent need to understand the biological and social factors contributing to both obesity and addiction. As biological understanding increases, many debilitating conditions that were traditionally thought to derive from individual failure or lack of willpower (such as drug addiction) become re-interpreted, as neural hijacks. The same may be happening to obesity. This is generally a good thing: stigma is reduced, and options for treatment and therapy are increased.

But there is also a danger of fatalism, both on a societal and individual level: we currently interpret new research through the lens of a see-saw worldview: as biology dominates, personal responsibility recedes. Instead, for modern neuroscience to have a useful impact on society, it must construct explanations for complicated conditions that allow for biology and responsibility to co-exist. Redefining a social and biological phenomenon as a “brain disease” is not the way to do this. After all, absent a ghost in the machine, the owner does not melt away when the brain comes into focus.

I have just returned from UPenn where I spent some time as a Fellow in neuroethics. This blog post is adapted from a talk I gave during that time.