2013 Fiscal Year Final Research Report
Neural mechanism of prefrontal cortex for computation of information value
| Project/Area Number |
22500377
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Fusional social brain science
|
|---|
| Research Institution | Tokyo Institute of Technology |
|---|
Principal Investigator |
NAKAMURA Kiyohiko 東京工業大学, 総合理工学研究科(研究院), 教授 (10172397)
|
|---|
| Project Period (FY) |
2010-04-01 – 2013-03-31
|
| Keywords | 認知科学 / 神経科学 / 情報工学 |
|---|
| Research Abstract |
Humans seek information, that is, want to know more and more. This study focuses its efforts on analysis on neuronal mechanisms that calculate amounts of information. Preceding studies have proposed different definitions of the information value, which include Shannon information, probability gain in correct response, and change in belief represented by probability distribution. We examined which definition is consistent with neural activities of moneys performing information-seeking behavior. The obtained result is that in the prefrontal cortex many neurons exhibited activities consistent with Shannon information and some neurons with probability gain. Also, we showed that different properties of simple cells and complex cells could be reproduced by a computational model of the primary visual cortex under the Shannon information maximization.
|
