Cognitive Problems and Learning Robots

When I'll be talking at SCM next month, my talk will most likely make the connection between the type of studies currently undertaken in Robot learning and how they can be used towards diagnostics in people with cognitive problems. Generally, this knowledge is applied the other way around. An example of that approach can be found in this article on a Bayesian model of imitation in infants and robots. In it, one can read:

In particular, we intend to study the task of robotic gaze following. Gaze following is an important component of language acquisition: to learn words, a first step is to determine what the speaker is looking at, a problem solved by the human infant by about 1 year of age (Brooks & Meltzoff, 2002). We hope to endow robots with a similar capability.

This is fascinating because one of the reason kids with autism have difficulty learning is that their 'gazing' behavior is not optimal, which in effect stops them from learning by imitation. Another aspect of the connection between the two approaches is how imitation can be a quantifier of Autism at an early age. Video made by families of kids with autism show charateristic behaviors at an early age (9-12 months.) More recent papers show the behavior can be seen earlier at 4-6 months.