Neural data analysis and local neural circuit modeling

2014 Fiscal Year Final Research Report

• Project Area: Mesoscopic neurocircuitry: towards understanding of the functional and structural basis of brain information processing
• Project/Area Number: 22115013
• Research Category: Grant-in-Aid for Scientific Research on Innovative Areas (Research in a proposed research area)
• Allocation Type: Single-year Grants
• Review Section: Biological Sciences
• Research Institution: The Institute of Physical and Chemical Research
• Principal Investigator: FUKAI Tomoki 独立行政法人理化学研究所, 脳科学総合研究センター, チームリーダー (40218871)
• Co-Investigator(Renkei-kenkyūsha): TAKEKAWA Takashi 工学院大学, 情報学部, 助教 (50415220) ; KANG Siu 山形大学, 理工学研究科, 助教 (40415138) ; TERAMAE Junnosuke 大阪大学, 情報科学研究科, 准教授 (50384722)
• Project Period (FY): 2010-04-01 – 2015-03-31
• Keywords: 変分ベイズ法 / 積分発火モデル / 対数正規型神経回路 / STDP / 興奮抑制バランス / ブラインド信号源分離 / スパイク相関

Outline of Final Research Achievements

Recent advances in recording techniques enable us to simultaneously record the activities of a large-scale neural population, which in turn enables us to study in detail the circuit-level mechanisms of various brain functions. In this project, we proposed novel mathematical methods to semi-automatically extract the activities of massively many neurons from the data obtained by multi-electrode recordings or calcium imaging. We showed that the non-random structure of local cortical circuits plays active roles in the generation of spontaneous activity and task-related activities in working memory and decision making. Furthermore, we proposed a mathematical model of synaptic plasticity to generate this nonrandom circuit structure in spiking neuron networks, and demonstrated that the proposed learning rule enables a model cortical circuit to separately perceive auditory signals from multiple sources in a noisy environment, thus giving an effective solution to the “Cocktail Party Effect”.

Free Research Field

計算論的神経科学、神経情報学