If you have any problems related to the accessibility of any content (or if you want to request that a specific publication be accessible), please contact us at firstname.lastname@example.org.
Modeling Oxytocin-Induced Neurorobotic Trust and Intent Recognition in Human-Robot Interaction
AuthorAnumandla, Sridhar R.
Computer Science and Engineering
AltmetricsView Usage Statistics
Recent human pharmacological fMRI studies suggest that oxytocin is a centrally-acting neurotransmitter important in the development and expression of trusting relationships in men and women, distinct from its role as a peripheral hormone related to parturition and lactation. Oxytocin administration in humans increases trust, acceptance of social risk, memory of faces, and inference of the emotional state of others, at least in part by direct inhibition of the amygdala. The cerebral microcircuitry underlying this mechanism remains unknown. Here, we propose a spiking integrate-and-fire neuronal model of several key interacting brain regions affected by oxytocin neurophysiology during social trust behavior. Because modeling social interactions requires near real-time responsiveness, we embodied the brain simulator in a behaving virtual humanoid neurorobot which "sees" a human by way of a camera system designed to capture motion of edges in the environment. Tonic firing of the amygdala is modeled using the recurrent asynchronous irregular nonlinear network architecture. Oxytocin cells are modeled with triple apical dendrites characteristic of their structure in the paraventricular nucleus of the hypothalamus. We demonstrate the success of this hybrid system in learning trust by discriminating between concordant versus discordant movements of a human actor, which leads to cooperative versus protective behavior by the neurorobot when challenged by a new intent from the human. Implications for further research and future design of socially intelligent neurorobotic systems are also presented.