Computational Neuroscience
(ATR International)

Research Staff

  • Prof. Mitsuo Kawato

    Prof.
    Mitsuo Kawato

  • Assoc.Prof. Jun Morimoto

    Assoc.Prof.
    Jun Morimoto

E-mail { kawato, xmorimo }[at] atr.jp

Research Area

We aim to understand the human brain and to achieve new machine intelligence (artificial intelligence) based on brain information processing functions. We conduct research and educate students on computational neuroscience and cutting-edge machine intelligence (artificial intelligence) with such methodologies as brain decoding, brain machine interfaces, neurofeedback, and robot learning at ATR, an internationally recognized computational neuroscience center.

lab members

Key Features

Machine intelligence for humanoid robot control

The framework for finding optimal behavioral policy can be formulated as a goal-directed decision-making problem. Using data-driven reinforcement learning algorithms, we construct machine intelligence (artificial intelligence) for humanoid robot control to solve this decision-making problem.

Decoding brain signals

Brain signals resemble codes that encode cognitive and behavioral states. We address understanding how cognitive and behavioral information are encoded and processed in the brain using computational modeling and machine learning based on large-scale neural data. With decoded brain signals, we also develop brain-machine interfaces and brain-based communication systems.

Brain-Machine Interface (BMI) in daily life

By measuring brain activities in daily living environments, we develop inference techniques of such mental states as stress and empathy. Based on these, we approach the neural basis of cognitive functions in natural situations for the social applications of neuroscience, including human resource development.

Measurement data

To understand how the brain generates behaviors and thoughts, we develop analysis methods for human brain data. We emphasize multiple integration measurements to overcome the limitations of each individual brain measurement.

Neurofeedback

We integrate psychophysical, neuroimaging, and computational neuroscientific approaches and propose novel neurofeedback methods, developing effective methods for BMI, medical treatment, and communication applications.

Computational models of decision-making

Our goal is to understand how humans make decisions. Reinforcement learning models and economic theorems allow us to build neural computations for human decision-making. We apply them to solve social, economic, and medical problems.

Adaptive shared control for BMI exoskeleton robots

Since robots are expected to work closely with humans, the development of a shared control strategy is becoming an increasingly important research direction. We are constructing an adaptive shared control strategy for our brain-machine-interface (BMI) exoskeleton robot.

Fig.1 Machine intelligence for humanoid robot control

Fig.1 Machine intelligence for humanoid robot control

Fig.2 Decoding brain signals

Fig.2 Decoding brain signals

Fig.3 Brain-Machine Interface (BMI) in daily life

Fig.3 Brain-Machine Interface (BMI) in daily life

Fig.4 Measurement data

Fig.4 Measurement data

Fig.5 Neurofeedback

Fig.5 Neurofeedback

Fig.6 Computational model of decision-making

Fig.6 Computational model of decision-making

Fig.7 Adaptive shared control for BMI exoskeleton robots

Fig.7 Adaptive shared control for BMI exoskeleton robots