Toward Supporting Adaptation: Exploring Affect’s Role in Cognitive Load when Using a Literacy Game

Document Title

Toward Supporting Adaptation: Exploring Affect’s Role in Cognitive Load when Using a Literacy Game

Document Information

• Subject Area: Gamified Learning, Affective Computing, Cognitive Load Theory
• Keywords: Affect, Cognitive Load, Gamified Learning, Reading Comprehension, Personalized Learning, Student Behavioral Patterns, Data-Driven Adaptation, Self-Regulated Learning, Impact, Affect Prediction

Research Background and Problem

• Identified Issues or Challenges: Emotional states during learning (e.g., positive or negative) significantly influence cognitive processing. However, most studies on online educational technologies overlook the dynamic nature of emotional states and their interaction with cognitive load. Additionally, there is a lack of learning systems or models that integrate affect and cognitive load.
• Significance: Emotional states can impact the effectiveness of learning. In-depth research into the interaction between affect and cognitive load could provide important guidance for optimizing the design of learning systems, thereby improving learning experiences and efficiency.
• Motivation and Related Work: Gamified learning is considered to enhance learner motivation, but its effects are inconsistent. Cognitive Load Theory emphasizes optimizing learning design to reduce extraneous cognitive load. Emotional states (e.g., positive or negative) may regulate learners' allocation of cognitive resources, thereby influencing learning outcomes.

Solution

• Research Methodology: The study employed the experience sampling method, combining self-reported measures of affect (arousal and valence) and cognitive load, and analyzed participants' gaming behavioral patterns and eye-tracking data.
• Innovations:
  • Identifying the relationships between affect and types of cognitive load (intrinsic load, extraneous load, germane load).
  • Revealing potential optimization directions in gamified learning design through dynamic analysis of behavioral patterns in affect and cognitive load pathways.
  • Proposing adaptive learning system design principles, such as promoting positive emotions and managing negative emotions.
• Implementation Steps:
  • Designed an experiment using a gamified learning system with reading comprehension tasks, incorporating real-time eye-tracking.
  • Quantified emotional states and cognitive load through questionnaires.
  • Applied linear mixed models for quantitative analysis, supplemented by qualitative analysis of behavioral patterns.

Research Findings

• Specific Findings:
  • Emotional states (valence) can predict cognitive load: Positive emotions are associated with germane cognitive load (facilitating the construction of learning structures), while negative emotions are associated with extraneous load (increasing additional processing demands and hindering learning).
  • Students’ self-regulated behaviors (alternating between learning and gaming activities) are associated with increased positive emotions.
  • Failures or repeated negative feedback in the game may lead to emotional decline and negatively affect the learning process.
• Advantages:
  • Provides a new perspective on the role of affect in cognitive load, contributing to the development of affect-aware learning systems.
  • Offers a combined quantitative and qualitative methodological framework capable of identifying more complex behavioral patterns.
  • Provides empirical support for design optimizations, such as enhancing gamified learning effectiveness through emotional regulation features.
• Limitations and Future Directions:
  • Limitations include the study sample being concentrated on Canadian university students, which may affect the generalizability of the results. Additionally, the experimental environment involved short-term interactions, without fully considering long-term learning effects.
  • Future research is recommended to explore student groups from other cultural backgrounds and validate findings using data sources beyond self-reports, such as physiological signals.
  • Consider designing better humanized feedback mechanisms and support features to mitigate the impact of negative feedback on students' emotions.