2001 Fiscal Year Final Research Report Summary
STUDIES ON NEURAL MECHANISMS OF DECISION-MAKING AND LEARNING
| Project/Area Number |
12210016
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research on Priority Areas (C)
|
| Allocation Type | Single-year Grants |
|---|
| Review Section |
Biological Sciences
|
|---|
| Research Institution | JUNTENDO UNIVERSITY |
|---|
Principal Investigator |
HIKOSAKA Okihide JUNTENDO UNIV., PHYSIOLOGY, PROFESSOR, 医学部, 教授 (70120300)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
TAKIKAWA Yoriko JUNTENDO UNIV., PHYSIOLOGY, RESEARCH ASSOCIATE, 助手 (90053339)
KAWAGOE Reiko JUNTENDO UNIV., PHYSIOLOGY, RESEARCH ASSOCIATE, 助手 (30138250)
SATO Makoto JUNTENDO UNIV., PHYSOLOGY, RESEARCH ASSOCIATE, 助手 (90255688)
SAKAGAMI Masamichi TAMAGAWA UNIV., RESEARCH INSTITUTE, ASSOCIATE PROFESSOR, 学術研究所, 助教授 (10225782)
|
|---|
| Project Period (FY) |
2000 – 2001
|
| Keywords | Basal ganglia / reward / motivation / eye movement / decision making / dopamin / reinforcement learning |
|---|
| Research Abstract |
Expectation of reward motivates our behaviors and influences our decisions. Recent studies have shown that neuronal activity in many brain areas is modulated by expected reward. However, it is still unclear where and how the reward-modulation occurs and how the reward-modulated signal is sent out as motor outputs. Studies from our laboratory using saccadic eye movement suggest that the basal ganglia play a key role in guiding the gaze to the location where reward is available. First, projection neurons in the caudate nucleus are extremely sensitive to a positional bias of reward, usually preferring a larger reward on the contralateral side. The biased signal is then sent to the superior colliculus through the substantia nigra pars reticulata, mainly as disinhibition. This creates a bias in excitability between the superior colliculi such that the saccade to the to-be-rewarded- position occurs more quickly. Second, visual responses of caudate neurons are strongly modulated by expected reward, thus facilitating the preparatory process of saccade to the rewarded position. A possible underlying mechanism is that cortical inputs carrying spatial signals are strengthened or weakened (with possible synaptic plasticity) by the concurrent dopaminergic input carrying reward prediction error signal. These data lend strong support to the reinforcement learning theories which has been very successful in explaining behavioral learning and neural plasticity. Our results have shown that there exists a neuronal network in the basal ganglia that operates similarly to what the reinforcement learning model predicts. On the other hand, the basal ganglia network shows greater adaptability to behavioral contexts than the current model can do. Further studies using the reward-biased saccade tasks will provide an ideal paradigm for understanding the neuronal mechanism of reward-oriented behavior.
|
