The Context
The central question addressed by the project is how humans learn
complex visualmotor tasks. Can we piece together a model of human
learning of such tasks based on purely on the visualmotor inputs and
outputs? We answer this question in the affirmative: from a large
sequential corpus of visualmotor data gathered from human subjects
during learning, we can track the evolution of control policies in the
novice to expert transition in real-time. The visualmotor data is
non-stationary; it is characterized by periods of slow evolution
punctuated by conceptual shifts in which policies change radically.
We have developed algorithms that build and track models of the
policies across these conceptual shifts. These models are rich enough
to capture individual differences in the task, and are simple enough
to learn in real-time. Our goal is to use these objective measurements
of cognitive activity (rather that verbal reconstructions obtained
from subjects) to shape and speed training on such tasks, as shown in the
sketch below. 