【レポート】
Workshop on "Open Problems in Neuroscience of Decision Making"
Workshop on
"Open Problems in Neuroscience of Decision Making"
http://www.irp.oist.jp/nc/odm/index.html
From Oct. 15 to Oct. 19, 2008, I joined the workshop on Open Problems in the Neuroscience of Decision Making in Okinawa Institute of Science and Technology, Okinawa. It was a small workshop, about 50 participants from the world (as shown in the photo), and in total had 16 talks in three days. So there was enough time for discussion in each talk. All of speakers are famous scientists in their research fields of neuropsychology, neurophysiology, neuroeconomics, and neurocomputation. In this workshop, four main topics related to decision-making were discussed:
1. Valuation--what affects the evaluation of risk and the timing of reward.
2. Decision--where in the brain are decisions made?
3. Learning--how do dopamine neurons compute reward prediction error?
4. Multiplicity--Why do we use different strategies and which brain parts are involved.
In each topic session, there were several presentations. For example, in first session of valuation, Dr. Mark Walton gave a talk titled "Tracking costs and benefits in frontal and striatal circuits". The author proposed that the fronto-striatal circuits and the mesolimbic dopamine system valuate reward and costs to achieve greater benefits. There were also several interesting talks by Dr. Michael Frank (the title: Temporal integration of expected utility: Neurocomputational and genetic components), Dr. Bernard Balleine (the title: Changes in the cortico-striatal network during instrumental conditioning), Dr. Nathaniel Daw (the title: Multiple valuation systems: a computational view).
In those three days, around the four topics, participants discussed lots about the goal-direct behavior vs. habit behavior, model-based vs. model-free learning, the prefrontal cortex vs. striatum, shared the recent experimental data and ideas. Those topics are much related to my current projects. We want to try to demonstrate from the neuronal level that the prefrontal cortex performs the model-based learning, while the striatum applies the model-free learning. This workshop gave me hints to explain our neural data simultaneously recoded from the lateral prefrontal cortex and striatum of the monkey performing a reward inference task. In the section of general discussion in this workshop, I gave a 10-minute talk titled "Prefrontal cortex vs. striatum: model-based vs. model-free learning.
| 日時 | 2008年10月15日～10月19日 |
|---|---|
| 場所 | OIST本部、恩納村、沖縄県 |
| 報告者 | 潘 暁川（玉川大学脳科学研究所嘱託教員） |
