WHO:	Karen Adolph , Professor of Psychology New York University
TOPIC:	Two Learning Mechanisms in the Development of Action
ABSTRACT:	Many everyday problems require highly flexible solutions because the specifics of the problem space are highly variableóholding a conversation, fixing a meal from leftovers, driving a car through Manhattan, and so on. Maintaining balance is the central problem in the development of motor actions. Like social communication, impromptu cooking, and city driving, solving the balance problem requires a highly flexible solution because the biomechanical constraints on balance are continually changing. My colleagues and I have proposed that developmental changes bias infants toward acquiring flexible learning sets for coping with threats to balance. When learning sets are not viable, infants must resort to less efficient simple association learning. These claims are supported by findings from dozens of studies where we challenged infants with novel threats to balance. In some studies, depth cues specified variations in the layout of the terrain (slopes, stairs, bridges, cliffs, gaps, obstacles, barriers, etc.). In other studies, shine, color, and texture cues were available to alert infants to upcoming changes in friction and rigidity (rubber and nylon slopes, rubber- and teflon-soled shoes, foam pits, etc.). We show that infants acquire flexible learning sets for coping with novel variations in depth but must resort to less efficient association learning in the case of friction and rigidity. Two learning mechanisms are necessitated by differences in the available perceptual information
WHEN:	11/17/2003 12:00:00 PM
WHERE:	Meliora 269

Events Homepage