Using Auto-Recurrence Quantification Analysis to Examine Dynamics of Information-Gathering Behaviors and Agency During Game-Based Learning

• In Event: AERA Poster Session 6
  In Poster Session: Games for 21st Century Learning (San Diego)

Abstract

Leveraging behavioral sequences is essential for identifying whether interacting with scientific texts built into a game-based learning environment (GBLE) relates to learning. This paper uses auto-Recurrence Quantification Analysis to extract the entropy (i.e., number of unique sequential patterns) of learners’ information-gathering behaviors (i.e., interacting with a poster, reading a book or research article, talking with a non-player character) during learning about microbiology with Crystal Island, a GBLE. Undergraduates’ (n=82) sequencing of information-gathering behaviors, identified via log files, indicated novel patterns of information-gathering behavior were associated with higher learning gains and limited agency over actions. We conclude there needs to be increased diversity in information-gathering behaviors learners demonstrate during game-based learning to improve learning outcomes.

Authors

• Daryn Dever, University of Central Florida

• Mary Jean Amon, University of Central Florida

• Hana Vrzakova

• Megan Wiedbusch, University of Central Florida

• Elizabeth Brooke Cloude, University of Central Florida

• Roger Azevedo, University of Central Florida