Keynotes | K&Ts | GACs | Talks | Posters | Search

Poster E in Poster Session E: Thursday, August 6, 10:30 am – 12:15 pm, Kimmel Center, Shorin & Rosenthal Rooms

Building a model of infant development using experience, brain, and behavior

Cliona O'Doherty¹, Emily M. Chen¹, Sarah S. Tung¹, Michael C. Frank¹, Daniel LK Yamins¹, Cameron Ellis¹; ¹Stanford University

Presenter: Cliona O'Doherty

Infants quickly become proficient in complex tasks, such as mapping spoken labels onto objects in their world. Independently, functional neuroimaging of awake infants, head-mounted recordings from children’s egocentric viewpoints, and deep neural network modeling of infant vision have shed light on infancy as a period of complex cognitive and brain development. However, no single dataset has yet integrated these modalities to support building a comprehensive model of early development. Here, we present Project Simsom, a densely sampled longitudinal dataset comprising awake functional MRI, head-mounted camera recordings, and behavioral measures of language from a single infant aged 2 to 12 months. Preliminary results demonstrate that models trained on the head-mounted camera data can acquire physical world understanding, highlighting the promise of dense longitudinal research for Developmental NeuroAI.

Topic Area: Development, Individual Differences & Clinical Populations