Can't Just Stop

by Sharon Begley

  To overcome the limitations of retrospective analysis, from 2002 to 2004 Mayo’s James Bower and J. Eric Ahlskog asked their Parkinson’s patients if they had begun showing any unusual behavior after being prescribed dopamine agonists. Eleven said they had developed a compulsive drive to gamble, in most cases within three months of beginning the dopamine regimen or increasing dosage, the Mayo team reported in 2005 in Archives of Neurology. Most were taking pramipexole (trade name Mirapex), which locks onto the dopamine receptors that are especially abundant in the nucleus accumbens—reward-circuit central. “This was the paper that showed this is consequential and something we should pay attention to,” said Mayo’s Bostwick.

  One of the eleven patients was a fifty-four-year-old married pastor who had been in the habit of dropping twenty dollars or so during his once-every-four-or-five-year visits to a local casino. But on the dopamine agonist “he began to gamble almost daily and over several months lost about $2,500, which he kept secret from his wife. He reluctantly brought this up to his neurologist,” the researchers wrote. The sixty-three-year-old who began gambling at casinos two or three times a week on the Parkinson’s meds, compared to once every three months before, told his neurologist that he felt an “incredible compulsion” to gamble even when he “logically knew it was time to quit.” A forty-one-year-old computer programmer, who had never gambled in his life, said that after he started taking the Parkinson’s meds he felt “consumed” with the need to gamble online, losing $5,000 within a few months. “In addition to gambling,” the Mayo researchers wrote about this poor soul, “he compulsively purchased items that he did not need or want.”

  Neurologist Joseph Friedman of the Alpert Medical School at Brown University reported a more unusual compulsion in a Parkinson’s patient on a dopamine drug, an accountant who “tallied figures over and over again,” Friedman told the in-house magazine of Massachusetts General Hospital in 2006. “Another patient couldn’t stop trimming her hedges in winter, another obsessively pulled weeds and one couldn’t go grocery shopping because she couldn’t stop reading labels on cans.” In every case, as soon as doctors took them off the dopamine-boosting meds, the compulsions melted away.

  Only a minority of Parkinson’s patients develop a compulsive behavior on a dopamine agonist. When Mayo neurology fellow Anhar Hassan and her colleagues reviewed the records of 321 Parkinson’s disease patients from 2007 to 2009, by which time movement-disorder physicians were keenly aware that the drugs were linked to compulsive behaviors and made sure to ask about them, sixty-nine patients, or 22 percent, had developed a sudden-onset compulsion while on the drugs. The prevalence reached one in three among patients on the higher doses, the Mayo team reported in Parkinsonism and Related Disorders in 2011. Hassan counted twenty-five newly compulsive gamblers, twenty-four patients with compulsive sexual behavior, eighteen newly compulsive shoppers, six suddenly compulsive computer users, and eight “compulsive hobbyists.”

  Among them was the sixty-three-year-old woman who began spending three hundred to four hundred dollars per week buying flowers and twelve hours a day arranging them, an activity she had never before shown much interest in. One fifty-seven-year-old man began staying up late and compulsively building furniture and making pottery. A fifty-one-year-old burned the midnight oil rewiring or painting his house, while a fifty-five-year-old man began compulsively cleaning bathtubs. The men seemed to gravitate toward male clichés, such as the fifty-six-year-old who started purchasing new watches daily at Walmart and the sixty-three-year-old who loaded up on Lamborghinis and Bentleys (yes, plural). “Also compulsive ice-cream eating,” the researchers added. The women’s new compulsions also seemed almost absurdly stereotypical, such as the fifty-three-year-old who began compulsively buying costume jewelry and the seventy-two-year-old who began buying kitchen items so compulsively as to leave Martha Stewart in the dust.

  Scientists are mystified about why one Parkinson’s patient taking a dopamine agonist starts gambling compulsively while another starts compulsively gardening. They have made a little more progress understanding which Parkinson’s patients are most at risk of succumbing to a dopamine-fueled compulsion. Men seem especially susceptible, as do younger patients, those in whom the disease struck relatively early in life, those with impulsive personalities, and people who have had Parkinson’s for two decades or more. But the link is now indisputable: a 2014 analysis in JAMA Internal Medicine, using a Food and Drug Administration database of adverse drug reactions reported by physicians, found that the proportion of reports involving compulsive behaviors was 278 times higher for dopamine agonists than for other drugs. “The likelihood of a causal connection,” wrote Joshua Gagne of Harvard Medical School, “is high.”

  The What Circuit?

  If the dopamine-fueled pleasure-and-reward circuit were as simple as originally thought, the Parkinson’s patients would not be in this book. Their behavior would be described as an addiction, motivated by a drive for the hedonic hits that gambling can bring, not a compulsion motivated by a desperate need to avoid or quiet anxiety. But pleasure circuit, it turns out, is a misnomer.

  In 1954 at McGill University in Montreal, Peter Milner and James Olds were conducting an experiment that involved inserting an electrode into a rat’s brain. Their target was the reticular formation, which regulates sleep-wake cycles. The electrode curved off its intended path, however, and landed just above the hypothalamus, which is part of the limbic system’s Emotions Central. The landing site thus received the stimulation intended for the reticular formation. In the experiment, whenever rats pressed a lever their limbic system received an electrical stimulus from the implanted electrode. How much did rats like this? If a rat had an opportunity to press the lever while en route to chow, it lost all interest in dinner and instead hit the lever, stimulating its brain “frequently and regularly for long periods of time if permitted to do so,” the researchers reported in 1954 in the (now defunct) Journal of Comparative and Physiological Psychology. “The control exercised over the animal’s behavior by means of this reward is extreme,” they continued, “possibly exceeding that exercised by any other reward previously used in animal experimentation.”

  The McGill team found the same effect when the electrodes were implanted in the nearby nucleus accumbens, Olds wrote in a 1956 article for Scientific American called “Pleasure Centers in the Brain.” Even if rats had been deprived of food all day and were lured down a ramp by an enticing amuse-bouche, given the opportunity for electrical stimulation of the limbic system they never made it to dinner. Instead, they happied themselves nearly to death rather than taking a moment from their lever-pressing (up to two thousand times per hour) to scamper over to the food dish. Later, given a choice between pressing a lever to stimulate their brain’s “pleasure center”—let’s defer to Olds’s 1956 nomenclature for now—or one that would warm up a cage cold enough to hang meat in, the rats opted for the former. When research in the 1970s established that the neurons in the regions Olds called the pleasure center run largely on dopamine, dopamine was crowned the brain’s “pleasure chemical.” And lo, the idea of a “pleasure circuit” became dogma.

  Then things grew more complicated. Think back to the McGill rats, whose behavior was interpreted as pleasure-seeking. But rats, like people, engage in repetitive behaviors for lots of reasons. What if the rats, which were a few ganglia short of being able to articulate their feelings, were not sensing pleasure every time they pressed the lever but experiencing anxiety if they didn’t?

  Looking back on decades of such experiments in 2010, psychiatrist Morten Kringelbach of Oxford University and psychologist Kent Berridge of the University of Michigan wrote in the journal Discovery Medicine that “[t]he pleasure electrodes may never have lived up to their name.” Although studies “also found compulsive lever pressing in some patients, it was never clear from these patients’ subjective reports that the electrodes did indeed cause real pleasure.” Instead, electrical stimulation of the nucleus a
ccumbens produced a mildly pleasurable feeling, but rarely with an intensity that the rats apparently experienced. That has led researchers to suggest that stimulating the dopamine circuit never caused intense pleasure at all, but rather a compulsive drive to obtain the stimulation. And thus was brought to the fore the distinction between liking and wanting: one can want, even need, to do something but derive no enjoyment from it. Just ask someone with OCD.

  The dopamine circuitry, rather than living up to its original moniker as a pleasure center, is more like a prediction machine. It forecasts how rewarding something will be and then compares reality, once it arrives, to that prediction. If reality falls short, you feel a sense of incompleteness, dissatisfaction, of being left anxiously hanging by the behavioral equivalent of waiting for the E flat after the G-G-G in the opening chords of Beethoven’s Fifth Symphony (as I described in Chapter 4). In that situation the nucleus accumbens generates a feeling of wanting to try again to make reality live up to its dopamine-created hype—by playing another hand of poker, trying another bite of cheesecake, hitting the mall again. The compulsion arises from the need to make reality accord with prediction, to complete the quest for reward, to satisfy the expectation. But if and when that happens, the result is not happiness as we usually conceive it but, at best, relief, a sense of alright-ness. It’s the E flat.

  A study with monkeys showed what was happening in the brain. To establish that the animals’ ventral tegmental area functioned like that of humans, the scientists gave them a drop of sweet syrup, which they love, while recording their brain activity with electrodes. As expected, the reward/pleasure circuit became active and there was a burst of dopamine; the ventral tegmental area indeed seemed to register rewarding, pleasurable experiences, and to run on dopamine, as the human version does. The monkeys were then trained to watch a video screen, learning that when a green light appeared a drop of sweet syrup would descend from a tube within reach of their mouths a couple of seconds later. A red light, however, was the prelude to nothing. Green light = reward, red light = disappointment.

  As the monkeys made the Pavlovian connection of color to reward, their ventral tegmental area became active and awash in dopamine when they merely saw the green light. It didn’t wait for the syrup: knowing the treat was coming was apparently as enjoyable and rewarding as the treat itself. Starbucks addicts likely have a similar experience: if you see that green-and-white mermaid, a feeling of pleasure washes over you even before you take that first sip of iced vanilla macchiato. In the monkeys, there was no second spike when the monkeys actually licked up the syrup. This was evidence that activity in the “reward” circuit, fueled by dopamine, was producing the expectation of reward, not the satisfaction of the reward itself. If it marked pleasure, dopamine-circuit activity would have risen when the monkeys actually received the syrup. But it didn’t.

  The arrival of the dopamine spike before the arrival of the reward is analogous to the seconds between when you set the slot machine’s lemons, cherries, and JACKPOTS! spinning and when they stop. Achieving a reward and merely anticipating it—whether winning new powers in a video game, winning at the slots, or anything else—each produce activity in the ventral tegmental area, the mother lode of dopamine. Once you get a taste of the pleasure that awaits you, your reward-expectation circuitry lights up like a winning slot machine. You feel compelled to continue. Oh, and near-misses make the ventral tegmental area as active as outright wins: missing the green pig with our angry bird or reading a total dud of a text rather than the invitation we were hoping for stokes the dopamine circuit almost as powerfully as true wins. Disappointments cause dopamine levels to crash, making us feel disappointed, anxious, unfulfilled, driven to get the promised pig or text we expected. Activity in the dopamine circuit is not so much about pleasure as about expecting pleasure, and when we don’t get it we feel driven to seek it out, desperately and compulsively.

  * * *

  Dopamine neurons respond to the difference between the prediction or expectation of rewards and the occurrence of those rewards, neurobiologist Wolfram Schultz, then at the University of Fribourg in Switzerland, argued in the Journal of Neurophysiology in 1998. There are, he posited, three possibilities:

  1. If rewards arrive without having been predicted, the dopamine circuit becomes activated, with neurons firing away. Reality surpassed expectations and the dopamine neurons go wild. (There are margaritas at this church picnic? Yes!) You feel euphoric, which is why Jamie Madigan described the effect of winning unexpected loot in Diablo 3 as a dopamine freak-out. Getting a Porsche for your birthday when you expected another tie is way more thrilling than knowing the Carrera was in the cards.

  2. If rewards occur as predicted, the dopamine circuit fires, but less intensely.

  3. If predicted rewards do not come, dopamine neurons’ activity drops off a cliff. This is the situation in the brains of the Parkinson’s patients: the meds stimulated the firing of their dopamine neurons, but nothing in reality met the heightened expectation of reward. So they went thrill seeking at casinos or (thrills being in the eye of the beholder) pottery-making. What starts as increased activity in the dopamine circuit from artificially activating it with drugs becomes a compulsive drive toward more and greater rewards that, in what feels like a comment on the human condition, can never match the expectations generated by the brain.

  Since we are not dopamine slaves, that’s not the end of the story. The orbitofrontal cortex and other prefrontal regions can inhibit the neurological activity driving us to execute that behavior. The greater the activity in the striatal regions that house the dopamine machinery, the more likely activity in the prefrontal regions will fall short. Or, conversely, the weaker the activity in the prefrontal regions, the less activity is needed from the dopamine circuitry to drive a compulsive behavior. Which one wins determines whether we can squelch the compulsion or are forced to give in to it.

  Conclusion

  If this book has left you with anything, I hope it is the realization that there is no bright line between mental illness and mental normality. Many psychiatrists use that fact to argue that the bar for diagnosing mental disorders is too low, and that too many people who should be diagnosed as mentally ill are slipping through the net. Studies that raise the estimated prevalence of depression or PTSD or ADHD not only ratchet up the pressure to increase funding for research on that particular illness; they also implicitly plant in society’s collective brain the suspicion that experts are underdiagnosing other mental disorders, too. That notion gets picked up by the media and turned into listicles of “warning signs” for this or that psychiatric ill, and presto—more and more of us have a mental disorder, and soon the exploding numbers call into question the very notion of mental normality.

  That point has been made countless times by experts who have pushed back against ever-expanding psychiatric diagnoses. This is not the book to re-litigate that case.I But the fact that anxiety has become the most prevalent psychiatric disorder holds lessons for those of us who do not believe that everyone is, even a little, mentally ill. Most important, it sheds light on the seeming ubiquity of compulsive behaviors. Not the extreme compulsions of OCD and hoarding, to be sure. Instead, by recognizing that anxiety is (to paraphrase the insight of medical historian Roy Porter quoted in the Introduction) the lunacy that our age deserves, the puzzling and the disconcerting become understandable. The compulsive behaviors that anxiety drives us to range from organizing kitchen cabinets to our most idiosyncratic specifications to acquiring things we don’t need, from saving a single fading dried flower from a funeral to desperately swiping through our phones for fear of missing out, all in an effort to keep the angst at bay. The saddest thing I came to understand in researching and reporting this book is that so many of our behaviors draw us into them not because they bring joy but because they promise to quiet anxiety. But the most heartening thing was the realization that the ability of compulsive behaviors to quiet anxieties great and small is one of the
greatest gifts our brains can give us.

  * * *

  I. For the best presentation of the case against the creeping expansion of mental diagnoses, read psychiatrist Allen Frances’s 2013 book Saving Normal: An Insider’s Revolt Against Out-of-Control Psychiatric Diagnosis, DSM-5, Big Pharma, and the Medicalization of Ordinary Life. The subtitle says it all, so I’ll add only that Frances oversaw the DSM-IV, and feels morally obligated to make amends.

  ACKNOWLEDGMENTS

  I AM ENORMOUSLY GRATEFUL TO the people, named and unnamed, who agreed to tell me about their compulsions. They did so in the hope that others would understand why they do what they do, and why saying “just stop” is not only pointless but heartless. If I have not furthered that understanding, it is entirely my fault. My enormous thanks, too, to the psychologists, psychiatrists, neurologists, and neuroscientists who patiently explained their research to me and did not throw me out of their offices when I asked, for the umpteenth time, what qualifies a behavior as compulsive rather than addictive or impulsive. Finally, let me acknowledge those who helped me at the very beginning and at the very end of the work on this book. Thanks to Douglas Main, who helped me find enough scientific research to feel confident that there was a book to be written about compulsions, I was able to put together the proposal that became the basis for this book. And, thanks to my editor, Karyn Marcus, the book you read is much better than the manuscript I originally submitted.