Using Symbolic Dynamics to Reveal a Strategy-Based Control Parameter When Learning From Text

• In Event: 61.015 - Embracing and Modeling the Complex Dynamics of Motivation and Engagement: Contextual, Temporal, Dynamic, and Systematic

Abstract

Theoretical Frame & Purpose
Self-regulated learning (SRL) involves volitional engagement towards a learning goal (Pintrich, 2000; Winne & Hadwin, 2008; Zimmerman, 2008). Behavioral sequences during SRL episodes are thought to be meaningfully related, but few studies have investigated behavioral sequences or how they may be impacted by the use of metacognitive strategies. We investigated the role of learners’ decision to read versus read and take notes on the dynamics of visual attention, which served as a proxy for the focus of the cognitive-metacognitive system. To isolate sequences in visual attention we applied a symbolic dynamic analysis technique called Orbital Decomposition (OD), which detects patterns in sequences of nominal data derived from successive measurements of a system (Guastello, 2008). We modeled the learner’s SRL behavior as a dynamic system within an attractor plane, and used symbolic dynamic analyses to establish the nature of repeated gaze iterations among materials.
Methods & Data Analysis
Each of twenty-four undergraduate students (n=12 female; mean age=22.14) viewed three pages of information, one sheet of note-taking paper, and a sheet of instructions. Self-paced studying behavior was recorded and sampled each second for up to ten minutes. A letter symbol was assigned to each gaze location—A through C: Sheets 1 to 3, D: Note-taking page, E: Instructions. Optimal symbol sequence lengths (Guastello, 2000) of repeated gaze locations were calculated from raw data strings (e.g., ABABAB) by the ORBDE 2.4 OD software (Peressini & Guastello, 2014). Results were plotted using symbol sequence plots and analyzed with ANOVA.
Results
The decision to read versus read and take notes did not influence posttest performance or time on task, but it did significantly impact the attention given to various materials (F(2.91, 63.90)=9.06, p<0.000). Whereas learners who did not take notes spent 47% of their time inspecting the first page, those who read and took notes spread their attention more evenly among the different materials. The first page served as an important point of origin for repeated gaze sequences for both groups. It was the prominent point of origin of repeated gaze sequences for learners who did not take notes. However, for note-takers, the note-taking page served as the most powerful point of origin. Reading without note-taking was associated with a high frequency of moving back and forth between the first and second informational pages (39.2% of sequences). Learners who also took notes most often repeated a gaze cycle (19.7% of sequences) between the first page and the note-taking page. Note-takers were significantly more likely to initiate repeated gaze sequences from the instructions.
Significance
To our knowledge, this is the first study to analyze SRL behavior sequences using symbolic dynamics. Our results suggest that metacognitive decisions during SRL episodes act as control parameters that systematically shift learners’ behavior within the overall attractor plane of possibilities (Stamovlasis & Sideris, 2014). Future research is needed to investigate the relations between gaze sequences and SRL processes, and the possibility that gaze sequence structure and organization may change according to varying learner or task features.

Authors

• Joanna K. Garner, Old Dominion University

• Daniel M. Russell, Old Dominion University