A CU-Boulder team studying Parkinson's disease patients found strong evidence that dopamine in the brain plays a key role in how people implicitly learn to make choices that lead to good outcomes, while avoiding bad ones.
The finding could help researchers understand more about how the brain works and could lead to a better understanding and treatment of brain disorders like schizophrenia, according to CU-Boulder psychology graduate student Michael Frank, who led the study.
A paper on the subject by Frank, CU-Boulder psychology Associate Professor Randall O'Reilly and Lauren Seeberger of the Colorado Neurological Institute's Movement Disorders Center appears in the Nov. 5 issue of Science Express, an online version of Science magazine.
Often people will get a "gut feeling" that allows them to make a choice depending on how often it was associated with positive outcomes in the past. But people with Parkinson's disease often have difficulty making these kinds of choices, Frank said.
To understand why, they developed a computer model of the effects of Parkinson's disease and the medications used to treat it in the brain. From this model they predicted that Parkinson's patients would differ in their decision making depending on whether or not they were taking their medication, which they confirmed in a subsequent study.
They found that patients on their medication were overly influenced by positive outcomes, while those who were off their medication were more influenced by negative outcomes, according to Frank.
"Because Parkinson's disease is caused by lower levels of the brain chemical dopamine, and the medications increase concentrations of this chemical, these results provide strong evidence that dopamine levels play a critical role in developing our likes and dislikes," Frank said.
Much of what is known about the brain comes from studying diseases that affect it, according to O'Reilly. So to understand how healthy people learn from their decisions, it helps to determine the conditions where learning is degraded.
"Studying Parkinson's patients helps us to understand how healthy people learn, by showing us what's going on under the hood," O'Reilly said. "It's kind of like when your car makes a funny noise and you discover how the fan belt works. In this case we looked at a disease that has showed us more about dopamine's role in the learning system of the brain."
Dopamine also plays a critical role in many other neurological conditions including attention deficit disorder, schizophrenia and drug addiction, so these findings may have broader implications, according to Frank. For example, by more precisely understanding the detailed effects of dopamine in the brain, drugs could be designed to more directly target beneficial actions, without as many unwanted side effects.
"This research clearly showed that current Parkinson's medications have undesirable side effects on learning and decision making," Frank said. Other researchers have documented a sudden onset of gambling episodes associated with patients taking these medications. These episodes may have been caused by the increased sensitivity to positive outcomes caused by the medications, together with an insensitivity to losses, he said.
"When experiencing a loss, dopamine levels are normally low in the brain, and the medication may prevent this from happening," Frank said.
By exploring the effects of potential drugs on different brain areas and circuits, the computer model can be used to find drugs that restore dopamine balance without causing these or other side effects. Similar applications to ADD are also possible. Currently, ADD is treated with drugs such as Ritalin that cause more dopamine to remain available in the brain, but these drugs also may block certain forms of learning, causing unwanted side effects.
"This work helps us understand how the brain learns from experience, and translates this learning into decisions, the kinds of decisions we make on a daily basis, often without much conscious thought," Frank said. "Understanding how this process works in detail may help in developing strategies to generally improve our learning and decision-making abilities."