Achievement

IGERT trainee Josh Lewis together with graduate student Hector Jasso and IGERT faculty Gedeon Deak and Jochen Triesch have built a 3D simulation of infant/caregiver interaction. The simulation uses a machine learning model of infant attention and uses real data from video sessions of caregiver behavior to drive simulated caregiver actions. The simulated infant learns that it is more interesting to look in the direction the caregiver is looking rather than the opposite direction. Infants with some developmental disabilities often do not exhibit this appropriate gaze-following behavior, and thus our model can help us better understand how these deficits occur and how certain caregiver behaviors might mitigate or exacerbate them. This is an example of how research in human learning and development can be helped by modeling human vision with machine learning algorithms. Josh presented this work at the Society for Research in Child Development 2009 conference.