Brian Scassellati is a computer scientist who uses models of human development to build humanoid robots that can interact socially with untrained users. As complex robotic technology becomes increasingly integrated into daily life, there is a critical need for these machines to operate comfortably and naturally. "Consciously and unconsciously, parents tailor their actions and the environment to their child. They draw attention to important aspects of a task and generally adapt the task to the child's capabilities," said Scassellati. "By building our machines to elicit this same type of behavior from humans, we hope to exploit the natural structure of social interactions to make robots more flexible, robust and intuitive to use."

Scassellati will address both the technical challenges involved in the construction of the robotic systems and scientific challenges in using these robots as tools to study human social development. Beyond the engineering goal of allowing simple, natural human-robot interactions, this research has also found application in the diagnosis and quantification of disorders of social development, such as autism. "We are currently evaluating systems for speech rhythm recognition, gaze identification in unstructured settings and position tracking and motion estimation as quantitative measurements for diagnosis," he said. "All of these techniques address understanding social cues."

Susumu Tomita is a neurophysiologist who studies how neuronal circuits store information in the brain. Billions of neurons in the brain communicate with each other at synapses using molecular signals called neurotransmitters. "The major neurotransmitte
'"/>

Contact: Janet Rettig Emanuel
janet.emanuel@yale.edu
203-432-2157
Yale University
26-Feb-2007