View article

The faces and hands of caregivers and other social partners offer a rich source of social and 8 causal information that may be critical for infants’ cognitive and linguistic development. 9 Previous work using manual annotation strategies and cross-sectional data has found 10 systematic changes in the proportion of faces and hands in the egocentric perspective of 11 young infants. Here, we examine the prevalence of faces and hands in a longitudinal 12 collection of nearly 1700 headcam videos collected from three children along a span of 6 to 32 13 months of age—the SAYCam dataset (Sullivan, Mei, Perfors, Wojcik, & Frank, under 14 review). To analyze these naturalistic infant egocentric videos, we first validated the use of a 15 modern convolutional neural network of pose detection (OpenPose) for the detection of faces 16 and hands. We then applied this model to the entire dataset, and found a higher proportion 17 of hands in view than previous reported and a moderate decrease the proportion of faces in 18 children’s view across age. In addition, we found variability in the proportion of faces/hands 19 viewed by different children in different locations (eg, living room vs. kitchen), suggesting 20 that individual activity contexts may shape the social information that infants experience. 21