Modeling the Infant Exploratory Process

bannerImage
  •  A critical issue in the study of infant development is to identify the processes by which task-specific action emerges from spontaneous movement. It has been proposed that infants learn to tune their spontaneous movements to generate task-specific action through reward learning, which is characterized by exploration preceding learning, and accompanied by task specific reduction in variability. However, the contribution of exploration to the early learning of task-specific action has little empirical evidence during infancy. We are computationally modeling our infant kicking data to predict the movements associated with learning the task. The difference between the model and the task-specific movement trajectory data from each infant is considered to be the exploration related variance. Our findings will provide the first quantification of the contribution of exploration to the early learning of task-specific leg action of infants. 


    Collaborators

    • Nicolas Schweighofer, PhD, Associate Professor, Division of Biokinesiology & Physical Therapy; Department of Neuroscience; Department of Computer Science, USC
    • Barbara Sargent, PhD, PT, PCS, Assistant Professor of Clinical Physical Therapy, USC