How do the statistics of the environment shape the infant mind across domains?

Session Type: Paper Symposium

Abstract

Infants’ early experiences contain rich statistical regularities across time, space, and multiple domains. These regularities in infants’ environments change not only over seconds, minutes, and hours, but over the course of development, tracking with children’s developing minds and abilities. Such regularities are pervasive in early experience, from low-level visual features, to complex transitions between syllables, to predictable sequences of actions, and even to patterns of emotions in children’s households. How does the infant mind leverage this statistical information to scaffold development, to learn, and to adapt to the local environment? And which statistics (among nearly endless possibilities) are best suited for uptake by the developing mind?
The goal of this symposium is to integrate research on how infants’ processing and learning are shaped by the statistics of their lived experiences in the domains of vision, speech, action, and emotion. Talk 1 will examine how changes in low-level visual statistics over the course of infancy serve to train the infants’ developing visual system. Talk 2 will demonstrate how toddlers’ experience with the transitional probabilities between syllables in speech scaffolds their ability to map words onto corresponding referents. Talk 3 will show how infants use statistical regularities in the actions they observe to form predictions. Lastly, Talk 4 will examine the correspondence between infants’ processing of transitions between emotions and the statistics of their own caregivers’ emotion transitions in the home.
Together, the four talks advance our understanding of the pervasive role of statistical regularities in shaping developmental processes across domains and timescales.

Sub Unit

• Attention, Learning, Memory

Chair

• Mira Nencheva, Princeton University

Individual Presentations

• The developing visual system biases the statistics of the infant visual environment - Presenting Author: Erin Morgan Anderson, Indiana University - Bloomington; Linda B. Smith, Indiana University - Bloomington

• A Changing Role for Transitional Probabilities in Word Learning Across Development - Presenting Author: Jill Lany; Ferhat Karaman; Jessica Sari Fleming Hay, University of Tennessee - Knoxville

• Visual statistical learning in the action domain - Presenting Author: Claire Monroy; Tommaso Ghilardi; Marlene Meyer, Radboud Universiteit Nijmegen; Sarah A Gerson, Cardiff University; Sabine Hunnius, Radboud University

• Infants track the statistics of emotion transitions in the home - Presenting Author: Mira Nencheva, Princeton University; Diana I. Tamir, Princeton University; Casey Lew-Williams, Princeton University