Awareness, Not Willpower: How Habit Loops Actually Unwind
A patient walks into Dr. Judson Brewer's office and says, "I did 15 years of therapy, and it didn't do anything. But then I tried your app and boom." He has heard this sentence enough times that he can quote it from memory.1
The version of boom in that sentence is doing a lot of work. The mechanism behind it is one most people never get taught, and once you've seen it, the standard advice for changing habits starts looking like trying to win a chess match with a checkers strategy.
On The Mental Fitness Podcast, Dave and Luke sit down with Dr. Brewer, a neuroscientist and addiction psychiatrist at Brown University whose lab has spent two decades testing what actually changes behavior. The findings are unusual. Willpower is largely a fiction. Anxiety is a habit. And the lever that does the work is something most of us have been taught to set aside as soft.
This piece is about that lever.
The habit loop your brain is running
Every habit, helpful or otherwise, runs on the same three-part loop the brain has used since long before language showed up. There is a trigger, a behavior, and a result. The brain logs the result and repeats whatever produced it.
Brewer roots this in evolution. An ancient ancestor goes out foraging, finds berries, eats the berries, and gets a dopamine signal from the stomach that says calories are here. The brain encodes the location of the berries as worth remembering. The next time hunger shows up, the brain reaches for the same path. This is positive reinforcement.
The other version is negative reinforcement. Same loop, different polarity. You see a predator; you run. The predator doesn't eat you. The brain encodes running as the right move. The behavior survives because doing it again next time keeps you alive.
This is the system that runs beneath every modern habit you have. It's also the system that's running underneath anxiety.
Anxiety is a habit, too
Anxiety is not a personality trait or a chemical imbalance you have to manage forever. Brewer's argument, which he traces back to a humble two-page paper by psychologist Thomas Borkovec in the mid-1980s, is that anxiety is the same trigger-behavior-reward loop as any other habit, just with different content.2
The trigger is uncertainty. The behavior is worry. The reward is a small sense of agency, a feeling of doing something, even though the worry isn't actually solving anything.
Brewer puts the question directly to his patients. What do you get from worrying? The most common answer he gets is nothing. People often look at him in disbelief. The loop kept running because the brain mistook the relief for a real reward, and the cost accumulated quietly underneath. The mechanism is indifferent to whether the behavior is rational. It cares whether the behavior was repeated.
Why willpower keeps losing the fight
This is where Brewer's findings get controversial.
There is, he argues, almost no neuroscience evidence for willpower as a distinct mechanism. The reinforcement learning models his lab uses, derived from Robert Rescorla and Allan Wagner's formulas from the 1970s, do not include willpower as a variable. The math works without it. You can predict how quickly an animal forms or breaks a habit without ever invoking the concept.3
The famous research on ego depletion (the popular idea that willpower fatigues across the day like a muscle) has fared poorly under scrutiny. When researchers pulled together the body of work in a meta-analysis, roughly half the studies showed a positive effect, and roughly half showed no effect, which is the statistical signature of chance. The Stanford marshmallow study, cited for a generation as evidence that self-control predicts success, was originally run on Stanford faculty's kids. When researchers controlled for socioeconomic status, the predictive effect largely disappeared.
What looks like willpower is mostly the prefrontal cortex temporarily overriding an older, stronger learning system. The override works on a good day. Under stress, fatigue, or hunger, the prefrontal cortex goes quiet first, and the older system takes over. White-knuckle effort works until it doesn't, and the moment it stops working is almost always the moment you most need it to.
Telling people they failed because they didn't have enough willpower also makes them feel broken, which is a profitable thing to sell against. It's a great business model if you want to sell people stuff that they have to keep coming back to, Brewer says. The model gets people to come back and try again. It doesn't get them anywhere.
There's a deeper problem, too. Suppressing an urge does not update the loop. The brain still thinks the behavior pays off. As soon as the suppression fades, the loop fires and wins.
What actually unwinds the loop
If willpower works against the brain's older learning system, what works with it?
Brewer's answer, refined across two decades of clinical research, is awareness. The specific, embodied kind that lets the brain finally see, in real time, what the loop is paying out.
The protocol is unusual. When an urge fires, you don't suppress it, substitute, or distract. You pay close, curious attention to what the behavior actually delivers in the moment after you do it. Not the story your mind tells you about it later. The actual physical and emotional state, in the body, in the minute after.
In one of his lab's studies, people who had overeaten for decades were asked to do exactly this. Pay attention while overeating. Notice what it feels like in your body during and just after. Within five to fifteen conscious attention events, the reward value the brain had assigned to overeating dropped below zero. The behavior stopped feeling rewarding. The brain updated its prediction, and the urge weakened. The decades didn't matter. The new evidence did.
Brewer calls this a negative prediction error. The brain expected one level of reward. The actual reward came in lower. Dopamine fires in the opposite direction, and the orbitofrontal cortex (the part of the brain that holds your set-and-forget reward hierarchies) reshuffles where that behavior sits.
In randomized controlled trials, the approach has held up. Brewer's first trial in generalized anxiety disorder produced a 67% reduction in anxiety scores against 14% in the usual treatment control group, with a 64% remission rate against 3%.4 A separate trial in anxious physicians showed a 57% reduction. An earlier mindfulness-based smoking cessation program was roughly twice as effective as the gold-standard Freedom from Smoking program at four-month follow-up.5 These are unusual numbers from a model that doesn't include willpower anywhere in its equations.
The three gears
Brewer often teaches the protocol as a bicycle gear shift, and the metaphor matters because it puts the effort in the right place. A bicycle on a steep hill in the wrong gear is exhausting and barely moves. Shift down, and the same hill becomes climbable.
First gear: identify the behavior. This is the part Brewer has recently simplified. His earlier work had people map the full trigger-behavior-reward sequence for every loop. His newer protocol drops the trigger entirely. Patients kept getting stuck trying to chase down why they were doing the thing, often from old therapy reflexes that said the root cause had to be uncovered first. From a neuroscience perspective, the why doesn't matter, Brewer says. It just sets the loop in motion. It's not what actually changes behavior. So you simplify. Just notice what you're doing. Am I stress eating? Am I worrying? Am I procrastinating? Am I doomscrolling? The naming is the first move. If you don't know you're doing the thing, the loop keeps running automatically.
Second gear: ask what you're getting from it. Pay attention while the behavior is happening and just after. Feel into the body. What am I actually getting from this? The question is not intellectual. The answer cannot be reasoned. The thinking brain can know that the gummy worms aren't satisfying and still want them. The orbitofrontal cortex updates only when the body registers, viscerally, that the payoff isn't there. Brewer tells the story of his own gummy worm habit, which he held onto for years before paying genuine attention to what they tasted like and how he felt afterward. They don't actually taste good, he eventually noticed. The craving collapsed once the body had the evidence.
Third gear: find the bigger, better offer. Once the loop has been disenchanted, the brain is open to a higher real reward. The cleanest version, Brewer says, is simply not doing the thing. Not worrying feels better than worrying. When that's too sparse, the lever he uses most often is curiosity itself. What does this urge feel like in my body right now? How does it change if I watch it without acting? Curiosity activates the brain's reward system and gives it a more rewarding place to go. The gears are not stages you graduate from. You shift between them as the loop reappears.
Two types of curiosity
Curiosity in this sense is a specific thing. Brewer distinguishes interest curiosity, the kind that turns toward present experience (what is this actually like, right now?), from deprivation curiosity, the information-seeking urge to know more, research more, gather more. Both have their uses. Only the first one updates the brain's reward value.
The distinction matters because people who hear often become curious and default to the deprivation form, which keeps them in the head. They go looking for more information about their habit and never actually delve into what's happening in their body during the loop. The information doesn't update the orbitofrontal cortex. The felt experience does.
Where this falls apart
Even with the right model in front of you, there are a few traps that will pull you back into the old pattern.
The first is what Brewer calls applying a Western mindset to mindfulness. I have to do this. I have to pay attention to my breath. The effortful, grit-based posture creeps into the practice itself, making it another arena to fail in. Brewer admits he did this for a decade, including crying on a meditation retreat manager's shoulder because he couldn't do the practice the way he thought he was supposed to. The fix is the same as before. Curiosity, not effort.
The second is impatience. The five-to-fifteen attention events that contributed to overeating in his lab study didn't happen in one sitting. They happened across days and weeks of repeated practice. The work is to keep showing up to the same loop with the same curious attention until the brain catches up.
The third is substitution as a final answer. Replacing one habit with another (cigarette to candy, drug to overexercise) is a Band-Aid in Brewer's framing. It stops the immediate bleed without cleaning the wound. When the substitute isn't available, the brain falls back to the original because the underlying loop was never disenchanted.
The fourth is mistaking rumination for awareness. Worry is future-oriented, and rumination is past-oriented, and both pull you out of the present moment, which is the only place the loop can actually be updated. Interest, by definition, is about right now.
Five small experiments for this week
1. Name one loop. Pick one habit you've been trying to change and identify the behavior in plain language. I doomscroll. I stress eat. I worry. I pick fights. That's first gear. Don't try to change anything yet. The naming is the move.
2. Ask the question once. The next time the loop fires and you do the behavior, slow down and ask what am I getting from this? Feel into the body in the minute after. Not the story your mind tells you. The actual physical and emotional state. One conscious pass this week is enough to start.
3. Run the curiosity experiment whenever an urge arises. Instead of acting or suppressing, drop into the body. Where is the urge? What does it feel like? How does it change if you watch it for two minutes without doing anything? Treat the urge as the experiment, not the enemy.
4. Map your anxiety loop, if you have one. Most people do. The trigger is some flavor of uncertainty. The behavior is worry. The reward is a tiny false sense of control. Once it has a name, the worry loosens its grip, because awareness pulls it out of the background where it's been running on autopilot.
5. Download the habit mapper. Brewer has a free habit mapper on drjud.com you can print and use as a tracker. Even one week of conscious notes is enough to start the updating process.
What this actually means
The reason willpower keeps losing is not that you're weak. It's that the strategy itself is built on a mechanism that neuroscience doesn't really endorse. You've been asking the youngest part of your brain to overrule an older, stronger one on its own home field, in the worst possible conditions.
The way out is to look more closely. The loop weakens when the brain witnesses, in real time, that the payoff isn't what the prediction promised. A new pattern doesn't start with effort. It starts with attention.
Footnotes
1. Quoted in the Mental Fitness Podcast episode and in Brewer's clinical work. The patient was describing a shift to the Unwinding Anxiety app program Brewer's team developed.
2. Thomas D. Borkovec, "The nature, functions, and origins of worry," in G. C. L. Davey and F. Tallis, eds., Worrying: Perspectives on Theory, Assessment and Treatment (Wiley, 1994). Borkovec's earlier short paper in the mid-1980s introduced the negative-reinforcement framing of worry that Brewer's lab later operationalized.
3. Robert A. Rescorla and Allan R. Wagner, "A theory of Pavlovian conditioning: Variations in the effectiveness of reinforcement and nonreinforcement," in A. H. Black and W. F. Prokasy, eds., Classical Conditioning II: Current Research and Theory (Appleton-Century-Crofts, 1972). The foundational reinforcement-learning formula that current habit-change neuroscience still uses, with no willpower term in the equation.
4. Judson A. Brewer et al., "Mindfulness Training for Generalized Anxiety Disorder: Open Trial and Randomized Controlled Trial of an App-Based Intervention," JMIR mHealth and uHealth and follow-up RCT publications. The 67% anxiety reduction and 64% remission rate cited are from Brewer's first randomized controlled trial in generalized anxiety disorder, with 14% reduction and 3% remission in the usual treatment control group.
5. Judson A. Brewer et al., "Mindfulness training for smoking cessation: Results from a randomized controlled trial," Drug and Alcohol Dependence 119, no. 1–2 (2011): 72–80. Mindfulness-based smoking cessation showed roughly twice the abstinence rate of the gold-standard Freedom from Smoking program at four-month follow-up.