1-092 - Big data, little kids: Findings from novel large datasets in developmental psychology

Session Type: Paper Symposium

Integrative Statement

Throughout the social sciences, novel methodology and improvement in computational techniques have increasingly made it possible to work with large datasets. Developmental psychology faces unique challenges due to subjects’ age, but stands to benefit greatly from the ability to explore small or graded effects, compare infant behavior to model predictions, and explore individual differences.

Innovative approaches for collecting and analyzing large datasets are presented in the four papers in this symposium. The first paper investigates everyday visual environments of infants and toddlers using over 15 million frames of infant-perspective video from body-mounted cameras at home. The second paper presents a novel online system for running developmental studies on the Internet. Several replications demonstrate comparable online and in-lab results using verbal responses and looking measures. The third paper presents an open, extensible database for aggregating information about children’s vocabulary growth, currently containing ~4000 children. The fourth paper investigates infant speech perception using community-augmented meta-analyses of developmental studies.

Whereas the first two papers develop new methods to remotely collect larger datasets, the latter two support efficient sharing of data so that effective sample sizes are larger than in individual studies. Each approach yields new insights about cognitive development and expands the scope of potential research endeavors.

Sub Unit

• Methods, History, Theory

Chair

• Kimberly Scott, Massachusetts Institute of Technology

Individual Presentations

• Everyday objects: Discoveries about the structure of the learning environment using head-cameras at home - Presenting Author: Caitlin Fausey, University of Oregon; Linda B Smith, Indiana University

• Moving the lab home: validation of a web-based system for developmental studies - Presenting Author: Kimberly Scott, Massachusetts Institute of Technology; Laura E Schulz, Massachusetts Institute of Technology

• Studying children's vocabulary development at scale - Presenting Author: Michael C. Frank, Stanford University; Daniel Yurovsky, Stanford University; Rose Schneider, Stanford University; Virginia A. Marchman, Stanford University

• Keeping up with child development: community-augmented meta-analyses - Presenting Author: Christina Bergmann, École Normale Supérieure; Sho Tsuji, RIKEN Brain Sciences Institute; Alejandrina Cristia, LSCP