What is addiction, why does it develop, and how can it be treated? Professor Keith Humphreys tackles these complicated questions in his 2023 book, “Addiction: A Very Short Introduction”. The following edited excerpt focuses on what scientists and medical professionals mean when they apply the word ‘addiction’ to humans. He explains the important differences between animal and human research, the false equivalence between addiction and dependence, and when to invoke addiction as an explanation for behaviour.
Addiction in humans is a hypothesis that a group of things we can observe have an underlying connection, much as there might be a connection under the soil between a group of flowers we can observe above ground.
Among the most creative and important studies ever done in the addiction field were conducted by James Olds and Peter Milner in the 1950s. Working at McGill University in Canada, these pioneering neuroscientists showed that rats given targeted electronic brain stimulation in response to a behavior (e.g. pressing a lever in their cage) would aggressively pursue such stimulation even though it was not essential for survival in preference to activities that are, such as consuming food and water. Subsequent researchers showed that similar behavior could be induced by injecting rats with drugs such as cocaine. Once they had been dosed repeatedly, rats would endure significant pain to keep using drugs (e.g. run across an electrified cage to gain access) and consume cocaine in preference to nutritious food even when they were hungry. Addicted rats will also work very hard (e.g. press a lever 2,000 times an hour) to obtain drugs even though it harms them, that is, they burn enormous calories trying to get drugs that have no nutritive value.
“Addiction: A Very Short Introduction” is published by Oxford University Press
In animal research, addiction is defined by these directly observable phenomena: an animal is considered addicted to a drug when it repeatedly endures negative consequences – up to and including death – in its repeated efforts to use that drug.
Matters become more complex when defining addiction in humans. In humans, other data points enter the equation, specifically the ability of people to articulate their own experience. These include reporting things like “I really planned to stop after two shots of vodka, but ended up drinking the whole bottle” and “I find myself thinking about heroin all day long such that I can’t focus on anything else” and “I know that my cocaine use is wrecking my marriage, but I keep using it anyway.” Addiction in humans also includes choosing not to do other behaviors, for example participating in school, having a job, volunteering in a community (none of this is relevant to lab rats because there is not exactly much to do in a cage).
Addiction in humans includes directly observable phenomena but is not reducible to them. For example, people who are addicted respond to drugs differently from other rewards in ways that can be seen on brain scans. But none of these studies allows a doctor or scientist to point at a brain region with a particular signature and say “There, that is the addiction” as one might say “There, that is the cancerous tumor” on an X-ray or ultrasound.
Neither can we use some count of behavior to define a person as addicted, for example saying that anyone who consumes more than X drinks of alcohol, Y shots of heroin, or Z snorts of cocaine must be addicted.
To understand how science and medicine handle situations like this, imagine you are walking through the forest and you see a cluster of flowers growing. You can observe each of their petals, leaves, and stems, but you can’t see their roots, so you don’t know if they are all coming from one source (e.g. a single bulb, or a single nutrient in the soil) or whether it is just a coincidence that this batch of flowers grow together. You therefore have to make a guess, or what scientists would call a hypothesis, about what you can’t see.
Unless the behavior is genuinely causing harm, for example a person bankrupts themselves buying handbags or stops eating and sleeping and showing up for work in order to play a video game, there is no reason to invoke addiction as an explanation.
Addiction in humans is a hypothesis that a group of things we can observe have an underlying connection, much as there might be a connection under the soil between a group of flowers we can observe above ground. The most important of those observables, just as in animals, is the repeated use of addictive drugs in the face of harm. It also includes people reporting on experiences such as having their “mental real estate” consumed by drug use, experiencing struggles with control over their use, consuming more than intended (e.g., “I’ll just do one line of cocaine” turning into a weekend long binge), and experiencing dependence, tolerance, and withdrawal. Individually none of these things is addiction, but to speak of addiction is to believe they constitute a syndrome. Syndrome literally means “running together.” To say that all these observables are a syndrome is to hypothesize that they run together for an underlying reason, which we are calling addiction.
Tolerance, dependence, and withdrawal are experienced by almost all people who are addicted, but they are not in themselves addiction. You may find that a single glass of wine doesn’t reduce anxiety the way it used to because you have become tolerant to the effects of alcohol, but that doesn’t necessarily mean that you are experiencing harm or that you struggle to control your drinking. Dependence is often falsely equated with addiction, and this was not helped by medical nomenclature in the past using them interchangeably. But many people who take opioids for pain for weeks or longer will experience withdrawal when they quit yet not go on to use opioids in an addictive fashion.
Not all harmful substance use is addiction either. Someone who hardly ever drinks can get intoxicated at the office holiday party and then get a ticket for driving erratically on the way home, but this doesn’t mean they are addicted to alcohol. Indeed, even someone who engages in harmful substance use on multiple occasions is not necessarily addicted.
At the same time, harm is essential to the definition of addiction. We speak colloquially of “being addicted to” many activities that we engage in repeatedly, whether it’s listening to jazz music, shopping for handbags, checking our email, collecting rare coins, or playing a particular video game. But simply engaging in a behavior much more than other people and thinking about it often is not inherently proof of an addiction. Unless the behavior is genuinely causing harm, for example a person bankrupts themselves buying handbags or stops eating and sleeping and showing up for work in order to play a video game, there is no reason to invoke addiction as an explanation.
© Keith Humphreys
Extract from Addiction: A Very Short Introduction published by Oxford University Press in February 2023, available in paperback and eBook formats. Edited for the SSA website by Natalie Davies.
Keith Humphreys (OBE, PhD) is is the Esther Ting Memorial Professor in the Department of Psychiatry and Behavioral Sciences at Stanford University, a Senior Research Career Scientist at the Veterans Affairs (VA) Health Services Research Center in Palo Alto, and an Honorary Professor of Psychiatry at the Institute of Psychiatry, King’s College, London. Queen Elizabeth appointed him an Officer of the Order of the British Empire (OBE) in 2022 in recognition of his service to addiction-related services and scholarship in the UK.
The opinions expressed in this post reflect the views of the author(s) and do not necessarily represent the opinions or official positions of the SSA.
The SSA does not endorse or guarantee the accuracy of the information in external sources or links and accepts no responsibility or liability for any consequences arising from the use of such information.
