Computational motor control as a window to understanding schizophrenia.

Area of Science:

• Neuroscience
• Computational Psychiatry
• Cognitive Science

Background:

• Schizophrenia presents with diverse symptoms, including sensorimotor deficits beyond cognitive and emotional disturbances.
• A key hypothesis suggests impaired self-agency inference, where individuals misattribute actions and their consequences.

Purpose of the Study:

• To investigate the neurobiological underpinnings of schizophrenia symptoms by modeling the self-agency mechanism.
• To develop a computational framework explaining how altered agency perception contributes to motor abnormalities in schizophrenia.

Main Methods:

• Developed a novel computational model of agency experience within a Bayesian decision-making framework.
• Integrated internal models for action-outcome prediction using optimal feedback control and Kalman filtering.

Main Results:

• The model successfully predicted various motor abnormalities observed in schizophrenia.
• These predictions were based on the assumption of a deformed internal model in the schizophrenia's brain.

Conclusions:

• The proposed computational model offers a unified explanation for sensorimotor and agency deficits in schizophrenia.
• This embodiment-focused approach highlights a potential central neuropathology underlying schizophrenia's behavioral manifestations.