| Project/Area Number |
10480080
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
情報システム学(含情報図書館学)
|
|---|
| Research Institution | TOHOKU UNIVERSITY |
|---|
Principal Investigator |
YAMAMOTO Mitsuaki Graduate School of Information Sciences, Tohoku University, Professor, 大学院・情報科学研究科, 教授 (40004618)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
NEMOTO Yoshiaki Graduate School of Information Sciences, Tohoku University, Professor, 大学院・情報科学研究科, 教授 (60005527)
SHIRATORI Norio Graduate School of Information Sciences, Tohoku University, Professor, 電気通信研究所, 教授 (60111316)
ITO Takayasu Graduate School of Information Sciences, Tohoku University, Professor, 大学院・情報科学研究科, 教授 (80124551)
NAKAO Mitsuyuki Graduate School of Information Sciences, Tohoku University, Associate Professor, 大学院・情報科学研究科, 助教授 (20172265)
INOUE Shojiro Institute of Biomaterials and Bioengineering, Tokyo Medical and Dental University, Professor, 生体材料工学研究所, 教授 (70013860)
|
|---|
| Project Period (FY) |
1998 – 1999
|
| Keywords | 1 / f fluctuations / computer network / neural network / brain / wide-spread information system / hierarchy / control strategy / biosystem |
|---|
| Research Abstract |
This study explores the way of coexistence of human being and information system by investigating biological information systems such as brain and biosystems.
Simulations were performed using interconnected neural network models consisting of spiking neurons, where neuronal spikes could be regarded as an information carrier in the brain (packets in a computer network). Resulting cluster patterns of neuronal activities are similar to those of packet flow in the computer network. In addition, enhancing mutual connections between neurons augments the clustering. This result indicates that enhancing inter-node connections could result in localizing highly activated nodes in the computer network.
Possible control strategy employed in cardiovascular system is explored through modeling, which underlies 1/f fluctuations in heart rate and blood pressure. The results clarified that the control strategy is not a rigid one prohibiting deviations of control variables from an operational point, but that it is a loose one allowing the variations to some extent. Simultaneously, the hierarchical control strategy is shown to be essential that switches global activation level of the control variable.
Resulting dynamics of hybrid system coupling an actual hippocampal neural network and nonlinear oscillator is investigated for varied coupling mode. This study suggests that a system of essentially high degree of freedom happens to exhibit behavior of lower degree of freedom dependent on control strategy. Brain is considered to realize such a diverse dynamics in the neural network by varying controls.
Thus clarified control strategies employed in the brain and biosystem could provide a novel framework for network control and information access control in the internet.
|
