Human cortical representation of object categories in short-term memory of natural visual images

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K16017
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Cognitive science
• Research Institution: National Institute of Information and Communications Technology
• Principal Investigator: NISHIDA SATOSHI 国立研究開発法人情報通信研究機構, 脳情報通信融合研究センター脳情報通信融合研究室, 研究員 (90751933)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 短期記憶 / 意味知覚 / 視覚 / 脳内表現 / 脳情報デコーディング / ニューロマーケティング / fMRI / 自然言語処理

Outline of Final Research Achievements

To elucidate neural substrates for visual working memory, which is a neural function for maintaining visual information temporarily after its disappearance, I tried to develop a new mathematical method to model neural representations and to quantify neural representations of categories (e.g. “woman”, “talk”, and “cute”) during visual working memory. The developed method not only visualized neural category representations successfully but also was applied to a neural decoding technique to recover perceptual contents from brain activity in the form of words. Using this method, I found the cortical localization of category representations during working memory and the commonality of category representations between when visual information is directly perceived and when visual information is stored in working memory.

Free Research Field

神経科学