The brain is a large machine. We explore dynamical and statistical properties of human memory to understand how it can be updated and be collapsed. To understand these properties, we rely on machine learning theory. Then, theoretical predictions with computer simulation are examined by behavioral experiments with large robotic manipulandum (TX-ARM) and VR systems
The idea of "coevolution" between the mind and the motor skills hints us that the neural representation of the mind (e.g., decision, communication, thought) might be able to seep through to sensorimotor actions. We explore a neural representation of decision making in terms of the spatiotemporal properties on the coordinate systems of the body schema.
We propose a computational model of motor recovery through neurorehabilitation training in order to develop a new interactive rehabilitation techniques implemented into the rehabilitatoin robot (TX-ARM).
A communication between the brain and the machine often written in Cyberpunkish literature might be realized in the near future. We develop a brain enhancement technology in terms of the motor function, the memory stability, and the decision performance by using the transcranial direct cortical stimulation (tDCS).