Commentary on “Everything We Think We Know About Addiction Is Wrong – In a Nutshell” (Johann Hari)

What you really need to know about Johann Hari’s claims

The popular Kurzgesagt – In a Nutshell video, based on Johann Hari’s TED talk, makes a couple of very good points. First, the benefits of human connection are indeed a major contributor to wellbeing for all of us.

As a species, we would be wise to steer for deeply fulfilling connection – and away from mindless stimulation, both chemical and behavioral. Second, drug addicts shouldn’t be treated like criminals. They should be trained on how best to manage what often proves to be a chronic disease – a disease of pathological learning, which is accompanied by physical changes in the brain that drive continued use despite negative consequences.

However, neither the benefits of connection nor Hari’s plea for compassionate treatment of addicts justifies his title implying that addiction science is off the mark, or has overlooked either of these points. Hari could have promoted his messages without ignoring or dismissing mountains of solid research published on addiction.

Others have insightfully addressed the weaknesses of Hari’s claim with respect to drug use (chemical addiction). See “4 Things Johann Hari Gets Wrong About Addiction” (The Fix) and “Everything We Think We Know About Addiction Is Wrong – In a Nutshell, Potentially Misleading” (Reddit). After correcting some of the general misinformation in this video, we’ll focus on behavioral addictions that involve supernormal versions of natural rewards.

The video is based on a false premise

The video starts out with a straw-man argument. It claims that if “what we think we know about addiction” were true, everyone given heroin in the hospital would be hooked. Actually, no addiction expert believes this.  Researchers report that only 10-20% of users offered addictive drugs become addicted, in both humans and animals. Hari’s false premise is the claim that every isolated caged rat becomes addicted if given access to heroin or cocaine. It’s more like 20% as this 2010 study reveals (with heroin rates somewhat higher):

“In a study published in the June 25th edition of Science, a team of researchers attached laboratory rats to a device that allowed the rodents to self-administer doses of cocaine—a coke IV of sorts. After a month, the researchers began identifying which rats had become hooked on the drug by looking for the hallmark signs of addiction: difficulty stopping or limiting drug use; high motivation to continue use; and continued use despite negative consequences. Only 20 percent of the rats exhibited all three signs of addiction, while 40 percent exhibited none.”

The difference between the addicted 20% and non-addicted 80% wasn’t lousy parenting or bad living conditions. Instead, it was how the rats’ brains adapted to drug use. Plain old genetics (or maybe epigenetics). The article continues:

“At first, drug use alters the physiology of every user’s brain as they go through a sort of reward-response learning: If you take the drug, you will feel better—certainly a dangerous mindset to be in when you’re wired to an unlimited supply of cocaine. Luckily, in most cases a brain eventually re-learns how to control its intake of the drug. Addict brains, not so much. Unlike their non-addict furry friends, the brains of addict rats lack sufficient “plasticity”—a property of the brain that allows it to adapt to changes over time—to get a handle on their habit. These rats are stuck in a reward-response frame of mind, and with it a downward spiral of addiction.”

Incidentally, 10 – 20% are the rates for situations where the user can self-administer a drug, thus reinforcing the connection between the “high” and using. Reinforcement of this type is different from hospitals, where pain medication is managed, and the presence of pain itself weakens reinforcement (because the body is already producing its own opioids so the drug “high” is less noticeable).

The exception to the 10-20% rate of addiction is nicotine, which is considered by many experts humanity’s most addictive drug. Its use is more socially acceptable and its immediate effects are less debilitating (characteristics it shares with internet porn use). There was a time when almost 50% of adult Americans were smokers. Did all nicotine addicts have attachment issues? Were all these smokers lonely? No. Even today we have millions of Americans who are quite happy and successful, yet cannot stop smoking. This alone refutes Hari’s premise.

While the 10-20% addiction rate may apply to substance use, we will see that supernormal versions of natural rewards (internet porn, junk food) can hook a higher percentage of users. For example, given a choice between sugar and cocaine, 85% of rats forgo cocaine to eat the sweet stuff. From this study:

“A retrospective analysis of all experiments over the past 5 years revealed that no matter how heavy was past cocaine use most rats readily give up cocaine use in favor of the non-drug alternative. Only a minority, fewer than 15% at the heaviest level of past cocaine use, continued to take cocaine, even when hungry and offered a natural sugar.”

If “In a Nutshell” viewers were told the truth, that only a minority of rats become drug addicts, Hari’s message would lose most of its impact.

Rat park experiment not replicated

Hari asks us to take the 1979 “Rat Park” experiment as gospel even though replication of the experiment failed. In doing so, Hari also asks us to ignore nearly 40 years of addiction neuroscience, which has identified cellular, molecular and epigenetic changes that account for the behaviors we recognize as addiction. For example, artificially increasing levels of a single molecule (DeltaFosB) makes rats compulsively crave drugs and junk food. Blocking this same reward-center molecule prevents addiction in lab animals. Similarly, in humans, active cocaine addicts (who died suddenly) had abnormally high levels of DeltaFosB in their brains’ reward centers.

Even more telling, an extensive body of brain-scan research reports that various addiction-induced brain changes are the best predictor of who will relapse (1, 2, 3, 4, 5, 6, 7, 8 ). In fact, in direct opposition to Hari’s claim, the only consistent factors related to either success or relapse were the magnitudes of certain addiction-related brain changes. From one of the studies:

“ER-fMRI data was compared with psychiatric, neuropsychological, demographic, personal- and family- history of drug use in order to form predictive models, and was found to predict abstinence with higher accuracy than any other single measure obtained in this study.”

How could brain changes predict relapse if the only cause of addiction were a lack of human connection?

There’s more to the Vietnam story

The author of this article “Vietnam Heroin Researcher May Have Disagreed With Johann Hari’s Take On The Causes of Addiction” further dismantles Hari’s claim (although he ultimately concludes addiction is a choice, a view we do not share). He points out that heroin was cheap and readily available in Vietnam, with more than 80% of servicemen offered it within the first week. However, the 1974 study reports that narcotic use wasn’t all that rampant:

“Approximately 13,760 Army enlisted men returned to the United States from Vietnam in September 1971. Within the population of 13,760, approximately 1,400 had been found to have urines positive for narcotics at time of departure.”

Only 10% of returning soldiers tested positive for opiates. It’s highly unlikely that all 1400 were heroin addicts, especially when we consider that some would have been given narcotics for pain relief. A ten percent addiction rate is far below the current addiction rate for drugs and alcohol in the US population.

Was widespread heroin use due to the stress of Vietnam or was it due to easy access to cheap heroin? A key finding was that most of the soldiers who eventually did become heroin addicts had prior histories of substance use, which suggests a strong genetic component for these soldiers’ addictions. Said the researcher,

“the greater the variety of drugs used before entering service, the greater the likelihood that narcotics would be used in Vietnam.”

If it was combat stress, why did the men who eventually became addicts generally start their heroin use early in their tours, before being exposed to combat? Why didn’t heroin use correlate with combat action? Said the researcher:

“Those who saw more active combat were not more likely to use than veterans who saw less, once one took into account their pre-service histories.”

Is it really surprising that most heroin-using soldiers stopped when they got back home? Heroin is costly, often hard to get, and interferes with civilian life: finding a job, working, renewing relationships, etc.

What about internet porn use?

Hari’s material has received an enthusiastic response on internet porn recovery forums where many users have been so glued to their screens that they feel socially isolated. Hari’s hypothesis encourages them to ascribe their addictive behavior to lack of human connection. However, Hari completely misses a key piece of information, which in turn leaves internet overconsumers with a major blind spot.

The relationship between human connection and addiction goes both ways, not one way. Many guys who quit discover that their inability to connect was due to their addiction, and that they become social magnets once they stop. That is, although isolation can drive self-medication via addiction, addiction itself impedes connection and mutes its benefits. An addicted brain is altered such that attachment doesn’t generally register normally or feel particularly good, compared with the drug or behavior the user has become “sensitized” to.

Over and over, we see that guys who quit report they become able to connect much more deeply with others, and with much greater satisfaction. Some even discover they were extroverts, not introverts. They are often amazed at how much more enjoyable they find social interaction, sexual activity with a partner, and even climax itself during sex. But they need a period of abstinence from overstimulation before they can fully benefit from the beneficial effects of connection. Their brain’s reward system needs time to rebalance. Hari does not address this need.

The power of supernormal versions of natural rewards

One implication of Hari’s message is that “as long as someone has a good social environment, he/she can engage in addictive behaviors without risk of becoming addicted.” This is just as misguided as a belief that addictive substances are equally dangerous for all users. We see lots of users struggling with internet porn’s effects who have had happy upbringings and plenty of social activity. We see happily married men struggling with it. Let’s look more closely at why internet porn is compelling even for those with good social connections.

Back up for a moment to reconsider drugs. The side effects of most drugs that offer a “high” are aversive. Many alter consciousness, interfere with ability to drive, cause debilitating hangovers, etc. Drugs are also risky to obtain or expensive (or both). Moreover, drugs are a poor substitute for natural rewards. Eons of evolution have tailored mammalian brains to light up for food, sex, bonding, achievement, play and novelty. While Hari informs us that connection is the true reward we are seeking, he ignores these other natural rewards. As psychologist Stanton Peele pointed out in this Psychology Today blog post:

“Rat Park is a classic experiment in which rats, once habituated to a morphine solution, preferred to continue drinking it over water in small isolated cages, but eschewed the morphine in favor of water in Rat Park, a spacious and enriched environment where there were many rats of both sexes. In such an environment the ability to compete for sex quickly took precedence over seeking narcosis – i.e., sex is better than drugs for rats.”

Nor does Hari explain to his viewers that supernormal versions of natural rewards (modern junk food and internet porn, for example) are far more universally appealing and addictive than drugs or alcohol. Supernormal stimuli are exaggerated versions of normal stimuli, but we falsely perceive them as more valuable. This helps explain why 35% of adult Americans are obese and 70% are overweight, even though none of them want to be. With our brain’s reward circuit lighting up, we can easily slam down 1500 calories in burgers, fries and milkshakes. Try slamming down 1500 calories of dried chewy venison and boiled roots in one sitting (or in one day).

Several animal studies have shown that junk food is more addictive than cocaine, (rats prefer sugar to cocaine) and that overeating to obesity can bring about addiction-related brain changes. In fact, when rats are given unlimited access to “cafeteria food,” nearly 100% binge to obesity. The obese rats’ brains and behaviors mirror those of drug addicts. These same rats don’t overeat on regular rat chow, just as hunter-gathers don’t get fat on their native diets.

To say this another way, there are no innate circuits for seeking heroin, alcohol, or cocaine. Yet there are various brain circuits devoted to seeking out and consuming both food and sex. And, while we like a good meal, sexual arousal and orgasm release the highest levels of rewarding neurochemicals (dopamine and opioids). That’s as it should be: reproduction is our genes’ #1 job.

While only a minority of rats become drug addicts; 100% copulate to exhaustion

What happens when you drop a male rat into a cage with a receptive female rat? First, you see a frenzy of copulation. Then, progressively, the male tires of that particular female. Even if she wants more, he has had enough. However, replace the original female with a fresh one, and the male immediately revives and gallantly struggles to fertilize her. You can repeat this process with fresh females until he is completely wiped out.

This is called the Coolidge effect—the automatic response to novel mates. Here’s how the Coolidge effect works: The rat’s reward circuitry is producing less and less exciting neurochemicals (dopamine and opioids) with respect to the current female, but produces a big surge for a new female. His genes want to make sure he leaves no female unfertilized…or exhaust himself trying.

Not surprisingly, rats and humans aren’t that different when it comes to response to novel sexual stimuli. For example, when Australian researchers (graph) displayed the same erotic film repeatedly, test subjects’ penises and subjective reports both revealed a progressive decrease in sexual arousal. The “same old same old” just gets boring.

After 18 viewings—just as the test subjects were nodding off—researchers introduced novel erotica for the 19^th and 20^th viewings. Bingo! The subjects and their penises sprang to attention. (Yes, women showed similar effects.)

Of course, a sedentary mammal experiencing a endless parade of willing females would occur only in a lab and not in nature. Or would it?

Internet porn as a supernormal stimulus

Internet porn is especially enticing to the reward circuitry because it offers an endless parade of sexual novelty. It could be a novel “mate,” unusual scene, strange sexual act, or—you fill in the blank. With multiple tabs open and clicking for hours a viewer can experience more novel sex partners every session than our hunter-gatherer ancestors experienced in a lifetime.

With internet porn, it’s not just the unending sexual novelty that buzzes our reward circuit. Strong emotions such as anxiety, shock or surprise also light up our reward circuit. Unlike scoring heroin on the street corner, today’s porn is easy to access, available 24/7, free and private. Unlike food and drugs, for which there is a limit to consumption, there are no physical limitations to internet porn consumption. The brain’s natural satiation mechanisms are not activated unless one climaxes. Even then, a user can click to something more exciting to become aroused again.

Unlike addictive drug use, porn use is now widespread, and almost universal among adolescent males with internet access. Moreover, many under the age of 30 view porn use as “healthy” and a normal part of “sexual expression.” Young men today use porn because they like it, not necessarily because they lack connection or love. (All neuroscience-studies published to date support the porn addiction model.)

Elephant in the room: the adolescent brain

Hari – who is no addiction expert – does not acknowledge the heightened vulnerability of the adolescent brain to addictive substances and behaviors, which exists quite apart from degree of social connection. For example, studies show that for teen brains, using drugs is far more permanently damaging than for adult brains.

Also, the risk of falling into addiction of all kinds is greater in teens, as is the risk of porn-induced sexual conditioning. Rates of erectile dysfunction, delayed ejaculation and low desire for real partners are soaring in today’s young men. A teen’s brain is at its peak of dopamine production and neuroplasticity, making it highly vulnerable to addiction and sexual conditioning. Adolescent animals produce higher levels of DeltaFosB in response to drugs and natural rewards.

What we have now is adolescents chronically using a compelling supernormal stimulus during the time when their brains are rewiring to the sexual environment. One primary goal of adolescence is to learn everything possible about sex (consciously and subconsciously) in order to successfully reproduce later on. Internet porn can thus alter or sculpt our extensive brain circuitry for sexuality and reproduction –  as well as distract us from learning the very social skills we need for connection.

Inadvertently or not, Hari’s animation leaves the impression that a good social environment prevents addiction. This simply isn’t true, especially for adolescents with their super-sensitive brains. As recovery forum host Gabe Deem points out:

Those rats in the Rat Park could have sex instead of heroin, but what they didn’t have is the option to “fertilize” millions of female rats on internet devices.

• Animal Studies of Addictive Behavior (2015)
• Cocaine Is Low on the Value Ladder of Rats: Possible Evidence for Resilience to Addiction (2010)
• How Cocaine Works in the Addict’s Brain (2010)
• Rats quit nicotine for a sweet reward following an extensive history of nicotine use (2015)
• The science of making drug-addicted animals (2012)