Embodied Computational Psychiatry: Autonomous Robots as Embodied Models of Mental Disorders

Computational psychiatry is increasingly establishing itself as valuable discipline for understanding human mental disorders. However, robot models and their potential for investigating embodied and contextual aspects of mental health have been, to date, largely unexplored. In this ongoing project, we seek to rectify this by creating robot models of mental health disorders. These robot models are not intended to replace other models of mental disorders, such as animal models, computational models, and theoretical models based on neuroscience and/or psychology. Rather, we seek to complement current models where robot models can offer advantages such as:

  • In areas where embodiment plays a key role, such as in interactions with the environment.
  • Allowing the modeler greater control of inner workings of models.
  • Looking at complete agent-environment systems.
  • Allowing access to the inner state of the robot, so that detailed data could be collected during experiments.
  • Allowing experiments that would be unethical if done in animals.

See our ACII 2019 Special Session: Emotion & Affective Technologies for Inclusive Mental Health

Publications

Visit our publications page if a paper is not directly available in the list below:

Funding: This project is funded by an Early Career Fellowship Grant of the University of Hertfordshire.

.