Computational Approaches Reveal Developmental Shifts in Exploratory Play.

Area of Science:

• Cognitive Development
• Computational Modeling
• Developmental Psychology

Background:

• Exploratory play is crucial for cognitive development, yet quantitative characterization of its shifts is limited.
• Challenges in coding and analyzing children's play behavior hinder research on developmental changes.

Purpose of the Study:

• To quantitatively characterize developmental shifts in children's exploratory play patterns using computational modeling.
• To investigate how children's exploration strategies change from early to middle childhood (3-11 years).

Main Methods:

• Analysis of behavioral data from 432 children across five diverse exploration tasks.
• Action-by-action coding of behaviors: repeating actions, switching targets, or terminating play.
• Application of computational Markov models to quantify child-specific transition probabilities (stay, switch, end).

Main Results:

• Older children demonstrated reduced perseveration (repeating actions) and increased switching to novel targets.
• Older children were more likely to terminate tasks earlier, indicating a shift in exploratory strategy.
• Developmental effects were most pronounced in novel toy exploration tasks.

Conclusions:

• Markov models effectively quantify developmental changes in children's exploratory play.
• Older children exhibit greater exploratory variability and less perseveration, with context-dependent variations.
• Findings reconcile prior research by emphasizing the interplay of task structure and developmental changes in exploration strategies.