Skip to content ↓

MIT researcher sheds light on why habits are hard to make and break

CAMBRIDGE, Mass. -- A Massachusetts of Technology researcher who studies an area deep within the brain has uncovered clues about why good habits are so hard to make and bad habits are so hard to break. Her work may also help those who suffer from extreme addictions and certain brain disorders.

"We all live mostly by habit," said Ann M. Graybiel, Rosenblith Professor of Neuroscience at MIT. Habits -- and automatic learned responses such as those used in driving and bike-riding -- may serve to free up the "thinking" parts of the brain for more creative purposes.

As anyone who has ever tried to quit smoking or lose weight knows, habits do not come and go easily. And extreme habits, such as addictions or actions repeated uncontrollably, are the dark side of the brain's ability to relegate tasks to the basal ganglia, three large nuclei of nerve clusters buried below the cerebral hemispheres in the forebrain.

Graybiel is tantalized by new evidence that there may be sensory tricks that break the destructive endless loops that seem to be tied to malfunctions of this brain region. If researchers could come up with a simple antidote to a seemingly unconquerable urge like the nicotine addict's craving for a cigarette, she says, it might help millions break free from the clutches of addictions.

Graybiel hopes that her research will lead to cures or improved quality of life for those with motion control disorders such as Tourette's syndrome. She also is investigating the idea that the basal ganglia may be tied to conditions seen primarily as "thought" disorders, such as obsessive-compulsive disorder (which is like an out-of-control habit) or schizophrenia.

Working with Graybiel are brain and cognitive sciences postdoctoral associates Yasuo Kubota and Naotaka Fujii; postdoctoral fellows Hu Dan, Pablo Blazquez, Juan Canales and Christine Capper-Loup; and graduate student Toshi Sakamoto.


The basal ganglia's work falls somewhere in between that of the cortex, which is active in the "here and now" skills like talking, thinking and learning, and the brain stem, which controls automatic body functions like breathing and blinking.

For a long time, the function of the basal ganglia remained a mystery. It is known that they are involved in the control of movement.

Lesions in the basal ganglia occur in motor disorders such as Parkinson's disease and Huntington's chorea. Neurotransmitters in the striatum, an area deep within the basal ganglia, may also be involved in Tourette's syndrome, depression, attention deficit disorder and addiction.

Certain psychological disorders have physical components. In obsessive-compulsive disorder, for instance, the same useless movement might be repeated over and over. Or Parkinson's patients seem unable to initiate a sequence of movements such as rising from a chair or walking from one place to another, but once the action is initiated, they have no trouble performing it.

Graybiel, whose research team is responsible for much of our current knowledge about neurotransmitter systems and gene expression in the basal ganglia, and her colleagues have uncovered evidence that the basal ganglia is tied to much more than motor control.

They see that its main inputs come from cognitive parts of the brain such as the frontal lobes, so Graybiel is not surprised that the basal ganglia demonstrate a strong reaction to things like reward signals. As an animal is rewarded for learning new behavior, changes occur in the neurons of its basal ganglia.

"Reward is incredibly powerful and drives a lot of the learning we do," she said.

New scanning methods show that this deep section of the brain lights up when we develop and express sequential motor acts, and also in response to rewards. With the new ability for researchers to see the brain's electrical activity while learning is in progress, they can actually see patterns of activity change permanently after learning takes place.


Learning a habit is different from other kinds of learning: often we are not aware of developing a habit, and we develop it slowly over time. "The process doesn't seem to go in reverse, or else we don't have access to the means to reverse it," Graybiel said.

Unlike an association between an object and a word ("Oh, so that's a blue jay!"), learning a habit is very slow. It takes many repetitions, often reinforced with positive feedback, before an action or series of actions become a habit.

Strong positive or negative motivators help develop or break habits. Positive feedback works better than negative. "The brain has an absolutely fabulous system for getting reward signals," said Graybiel. The system is so sensitive that researchers have seen nerve cells fire in response to a single word, evoking a craving long after a habit has been kicked.

Negative feedback, like feeling sick after eating or drinking something, can nip a bad behavior in the bud. In certain behaviors, like drinking alcohol, consequences such as a hangover occur too long after the original binge to do much good as a deterrent.

Many receptors are housed in the basal ganglia, which draw neurotransmitters such as dopamine like a magnet. And like a strong magnet, these receptors don't easily let things go. Once the basal ganglia have been exposed to a powerful addictive agent, it seems to recognize it forever.

Graybiel and her colleagues are looking at how long this kind of response lasts. They have found that even a single dose of an addictive drug will evoke a physiological response after three weeks of abstinence similar to the response that it evoked after a few days. They hope to study the response after as much as a year of withdrawal.


While humans and animals have inborn pattern generators (people automatically swing their arms when they walk like their ape ancestors; birds are born knowing how to peck), humans can develop these automatic behaviors on our own.

When the light turns green, we position one foot to press on the gas pedal, tighten our grip on the steering wheel and get ready to go, even though there is no evolutionary precedent for driving. "The brain is so malleable, we can make our own pattern generators," Graybiel said.

Graybiel suspects, however, that the more fine-tuning aspects of physical habits are accomplished elsewhere in the brain. While pounding out "Chopsticks" on the piano may become automatic, caressing the keys to produce just the right nuances of a Chopin prelude is not the same function, she says.

What we're doing when we learn a new habit may end up triggering the habit itself because, she said, as we develop habits we develop "chunks" of behavior. The process of walking, for instance, involves a series of movements, not just an isolated lifting or lowering of a foot.

When we repeat a behavior, physiological changes may occur not only in the parts of the brain responsible for motor control but also in the parts that deal with more cognitive functions.

The basal ganglia is the only place in the brain to deal with both physical and cognitive actions simultaneously, leading researchers to speculate that the way we program movements and the way we program thoughts may be deeply related.


When animals see a light flash or hear a beep when they get something to drink, they come to associate the light or sound with quenching their thirst. Researchers can see physical changes in their brains as the habit forms. "After a week, you begin to see a change," she said. "The neurons in the striatum actually change what they respond to. The brain changes when you pick up a simple habit."

While nerve impulses travel at lightning speed, genes take a little longer to change, but they too change in response to stimuli. But because brains are as individualized as our fingerprints, no two brains have an identical response to an identical stimulus. While it may take one person one week to develop a habit, good or bad, it may take another person considerably more time.

"Learning more about dynamic changes that occur in the brain as we make and break habits has great therapeutic potential. We may learn, for example, what a harmless habit has in common with an addiction and what is different about the two. This is a subject that interests us all," Graybiel said.

This work is funded in part by the National Institutes of Health.

Related Topics

More MIT News