• Top page
  • Timetable
  • Per session
  • Per presentation
  • How to
  • Meeting Planner

演題詳細

Special Lectures


開催日 2014/9/12
時間 16:00 - 17:00
会場 Room A(Main Hall)
Chairperson(s) 吉峰 俊樹 / Toshiki Yoshimine (大阪大学大学院医学系研究科 脳神経外科 / Department of Neurosurgery, Osaka University Medical School)

計算論的神経科学が世界を変える
Computational neuroscience changes the world

  • SL-2
  • 川人 光男 / Mitsuo Kawato:1 
  • 1:(株)国際電気通信基礎技術研究所 脳総合通信総合研究所 / ATR Brain Information Communication Research Laboratory Group 

Theories and models about brain have developed enormously in the past 50 years and now diverged into the following four seemingly unrelated disciplines. (1) Computational neuroscience in a narrow sense that models various levels of events and information processing in the brain. (2) Brain motivated robotics, where researchers build an artificial system that reproduces information processing of the brain. Understanding the brain by creating the brain is a catch phrase of this approach. (3) Brain machine interface (BMI) connects human brain and body with an artifact based on computational principles for recovery of lost functions or curing disorders. (4) Neurofeedback training that induces brain plasticity and/or specific spatiotemporal patterns in brain by real-time feedback of information related to neural activity. Recent progresses suggest that these once divergent fields can be reintegrated into with higher impacts on science and society. For example, Keio group developed EEG BMI-based neurorehabilitation system for stroke patients, which is regarded as an integrated system of the above second, third and fourth elements. Biomarkers for several psychiatric disorders and related fMRI neurofeedback therapies are examples integrating the first and fourth disciplines. In this presentation, I will mainly concentrate on "decoded neurofeedback (DecNef)" method that integrates the first, third and fourth disciplines to induce spatial voxel patterns in a limited brain region that corresponds to specific information including orientation, facial preference and color. Perceptual learning of specific orientation gratings, associative learning for manipulation of facial preference and creation of phenomenal consciousness of color were successfully realized by DecNef. The future of computational neuroscience in a broader sense could be integration of all the four disciplines, again. A human volunteer wearing a humanoid exoskeleton (2) that is controlled by brain activity measured on one hand (3), and on the other hand feedbacks (4) rich information back to the body and brain of the volunteer while covering wide range of sensory modalities. The purpose of this system could be compensation of lost functions (3), therapeutic treatments of disorders (4), or to induce significant spatiotemporal neural activity and brain plasticity (1).

Copyright © Neuroscience2014. All Right Reserved.