| Project/Area Number |
26870935
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Kansei informatics
Cognitive science
|
|---|
| Research Institution | Advanced Telecommunications Research Institute International |
|---|
Principal Investigator |
Horikawa Tomoyasu 株式会社国際電気通信基礎技術研究所, 脳情報通信総合研究所, 主任研究員 (60721876)
|
|---|
| Project Period (FY) |
2014-04-01 – 2017-03-31
|
| Project Status |
Completed (Fiscal Year 2016)
|
| Budget Amount *help |
¥3,640,000 (Direct Cost: ¥2,800,000、Indirect Cost: ¥840,000)
Fiscal Year 2016: ¥520,000 (Direct Cost: ¥400,000、Indirect Cost: ¥120,000)
Fiscal Year 2015: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2014: ¥2,080,000 (Direct Cost: ¥1,600,000、Indirect Cost: ¥480,000)
|
|---|
| Keywords | 脳情報デコーディング / fMRI / 印象 / 視覚 / 深層ニューラルネットワーク / 感性情報 / 高次認知機能 |
|---|
| Outline of Final Research Achievements |
The brain decoding approach was used to examine whether impressions on visual images can be predicted from human brain activity patterns measured using functional magnetic resonance imaging. The decoding analyses were performed to evaluate the predictability of 1) hierarchical visual features, 2) twelve types of impression scores associated with surface characteristics, and 3) positive-negative scores (valence), which were obtained using computational models and human rating. The analyses suggest that the lower/higher visual features were better predicted from the lower/higher visual areas, the impression scores were successfully predicted from the mid- to high-level visual areas, and the valence score was accurately predicted from the frontal brain areas. These results suggest that the multi-level visual information associated with subjective impressions on visual images can be decoded from human brain activity patterns.
|
