research

LinkIconJapanese page

Research.

Our research is mainly concerned with information processing in the brain. Assuming that dynamics of spikes in neural networks contribute to cognitive functions in the brain, we aim to elucidate the mechanism by which the functions emerge from the neural networks dynamics. We have been studying the relationships between neural network dynamics with synaptic plasticity and emergent properties of spike patterns. For example, we showed the possible memory mechanism through the hippocampal recurrent network. In addition, we showed that the state of neural network dynamics affect the function of recall of information memorized in the synaptic weights. Interestingly, the more chaotic the dynamics are, the more stable the spatial firing rates are. In the cortices, synchronous firing patterns are sometimes observed in vivo. We showed that spike-timing dependent synaptic plasticity (STDP) may contribute to create synchronous spike patterns in neural networks. In numerical experiments, STDP modifies the neural network structure to transform a specific spatiotemporal input pattern into a synchronous output. 

Currently, we are challenging the issues of intention to move, theoretically and experimentally. This concerns how our brain drives our planning to move and switches on the behavior. In other words, this problem includes how the switch-on circuit is self-organized and activated in the brain. We assume that our intention is evoked by nonlinear dynamics of neural networks and transmitted immediately across over the brain. 

Here, we have two approaches: Modeling and Experimental approaches. 


Brain Imaging
Brain Imaging
This is a way of observation on real brains in psychological experiments. We will deliberately analyze the state of the brain using EEG data when the brain begins preparing to move. This result may elucidate the mechanism of evoking voluntary action circuits. 
BrainNonlinearDynamicsSimulation
Brain Nonlinear
Dynamics Simulation		 This is a mathematical modeling approach, which imitates the structure and dynamics of neural networks in the brain. We aim to elucidate the emergent information processing function of the brain by examining dynamical properties of the models.