ScienceDaily (June 30, 2009) — A new neuropsychological memory test is helping
to uncover how Parkinson's disease can alter people's ability to learn about the
consequences of the choices they make. The test was developed by Dr. Mark Gluck,
professor of neuroscience at the Center for Molecular and Behavioral
Neuroscience at Rutgers University, Newark, working with co-researchers at
Rutgers, New York University, and in Hungary.

As reported in a forthcoming article in the journal Brain,* Gluck and
co-researchers Nikoletta Bodi and Szabolcs Keri of Semmelweis University,
Hungary, found that non-medicated patients in the early stages of Parkinson's
were selectively impaired at learning from reward.

Patients in Hungary were tested using a novel feedback-learning task developed
by Gluck and his colleagues: Catherine E. Myers, research professor, Rutgers
University, Newark; and Nathaniel Daw, assistant professor, New York University.
The research was supported by a Dekker Foundation Award from the Bachman-Strauss
Dystonia and Parkinson Foundation.

"What we are seeing in recently diagnosed patients is that prior to being put on
any medications, they exhibit a selective impairment in their ability to learn
from positive (rewarding) outcomes while their sensitivity to learning from
negative (or punishing) outcomes is normal," says Gluck, director of Rutgers'
Center for Collaborative Research on Cognition and Parkinson's Disease.

This selective deficit in learning from reward is not surprising, says Gluck,
because scientists have long known that dopamine is used to carry reward
information throughout the brain. Parkinson's patients, however, have lost most
of their dopamine-producing cells by the time they are first diagnosed with the
disease. This decrease in their ability to process rewarding outcomes could be
one reason why many Parkinson's patients experience depression, says Gluck.
"It's not just that they have an awful disease, but they have lost the ability
to process the rewarding aspects of life."

Gluck and colleagues found that this reward-learning deficit in un-medicated
Parkinson's patients is in direct contrast to what is seen in patients once they
begin treatment with dopamine agonists, a standard therapy for treating the
disease's motor symptoms. On dopamine agonists, a patient's ability to learn
from positive rewarding outcomes improved to normal levels. But there was a
catch – their ability to learn from negative (or punishing) outcomes, which had
previously been normal, was now impaired.

Gluck explains that an increased sensitivity to learning from events that
results in positive outcomes, coupled with a decrease in the ability to learn
from negative outcomes, could explain why some Parkinson patients treated with
dopamine agonists develop impulse-control disorders, including pathological
gambling, hypersexuality, alcoholism, and compulsive eating and shopping. All of
these behaviors can be understood as reward-seeking behaviors in the absence of
appropriate sensitivity to their negative consequences.

"For example," notes Gluck, "if your ability to learn from negative outcomes is
reduced and you play the slot machines and win $10 for a few rounds but lose
many more times in between, what you may recall best is the thrill of winning.
As such, you will be hampered in your ability to learn that gambling can also
have negative consequences."

The ability to test the effects on feedback learning in early onset Parkinson's
disease could provide additional insight into the impact of dopamine loss on
cognition and behavior. It also could pave the way for identifying which
Parkinson's patients are most likely to experience agonist-related feedback
problems so they can be treated with alternate medications.

Other ongoing research by Gluck and his clinical collaborators in New Jersey,
New York, Europe and the Middle East aims to further understand how and why
learning and decision making is impaired by Parkinson's disease and how the
medications used to treat motor symptoms can impair or remediate these essential
cognitive abilities.

Feedback Learning, Dementia, Depression and Drug Addiction

In addition to providing new insight into the effects of Parkinson's disease and
dopamine agonists on learning and personality, the new feedback-learning tasks
developed at Rutgers University are being used to study learning and
decision-making deficits in people suffering from fronto-temporal dementia
(FTD), a degenerative condition involving the front part of the brain's cortex.
FTD is associated with dramatic changes in personality, behavior and thought
processes, which can include inhibition, social withdrawal and compulsive
behaviors.

With co-investigator Murray Grossman of the University of Pennsylvania, Gluck is
studying changes in positive and negative feedback learning in those diagnosed
with FTD. That research, funded with a grant from the Association for
Frontotemporal Dementia, has the potential to help in identifying methods to aid
in correcting behavior for FTD patients.

"If, for example, FTD patients learn better from reward then punishment, this
would mean caregivers might be advised to avoid reacting negatively to
inappropriate behaviors and to focus instead on rewarding patients for not
engaging in inappropriate behaviors," says Gluck.

In related research, Gluck is working with Palestinian and Israel doctors, as
part of a joint Rutgers-sponsored Israeli-Palestinian brain research project, to
study cognitive deficits in Israeli and Palestinian patients suffering from
Parkinson's disease and depression.

* Advanced access published May 4, 2009 -- doi:10.1093/brain/awp094