1-044 - Finding Structure in Time: Computational Methods for Characterizing the Micro-dynamics of Interaction

Session Type: Paper Symposium

Integrative Statement

Our symposium is inspired by an influential paper by Jeff Elman highlighting the importance of characterizing the temporal structure of behavior for understanding human cognition. We believe this is even more true for development. Micro-dynamics of interaction shape childrens’ longitudinal trajectories across domains, from motor development to psychopathology. Behavior unfolds across many modalities and multiple timescales: from seconds (a gaze shift) to minutes (reading a storybook) to years (a secure attachment relationship). New technology allows us to collect multimodal data at a fine temporal scale, leading to the curation of massive multimodal corpora of behavior. However, along with new opportunities come new challenges. Theories are often underspecified as to the exact nature of these unfolding interactions, and researchers in developmental psychology have limited training for working with high-density datasets.

This symposium will provide developmental scientists an introduction to a variety of techniques for studying high-density interaction data. The presentations will cover basic qualitative and quantitative techniques for discovery of structure in high-density data (data visualization and clustering analyses; presentations one and two) and then move on to advanced algorithms for quantifying patterns of interaction between individuals or modalities (Recurrence Quantification Analysis and Granger Causality; presentations three and four). The presenters will provide conceptual and quantitative introductions to each technique illustrated with specific applications and empirical examples drawn from developmental psychology research. Finally, scripts associated with each presentation will be made available to all session participants on a citable Github repository available after the session.

Sub Unit

• Methods, History, Theory

Chair

• Kaya de Barbaro, The University of Texas at Austin

Individual Presentations

• Getting To Know Your Data: Visualization and Quantification of Temporal Dynamics of Interaction - Presenting Author: Kaya de Barbaro, The University of Texas at Austin

• Tapping into the Temporal Structure of Developmental Data using Burstiness Analysis - Presenting Author: Drew Abney, Indiana University

• Microgenetic Research using (Cross-)Recurrence Quantification Analysis - Presenting Author: Ralf Cox, University of Groningen

• Discovering Temporal Dependency and Directional Influence among Multimodal Behavioral Variables with Granger Causality - Presenting Author: Tian Linger Xu, Indiana University Bloomington