Affective Response Categories-Toward Personalized Reactions in Affect-Adaptive Tutoring Systems

Area of Science:

• Educational Technology
• Cognitive Science
• Human-Computer Interaction

Background:

• Affect-adaptive tutoring systems adjust learning experiences based on detected learner emotions.
• Such systems can enhance training, especially in safety-critical domains, by fostering coping strategies and resilience.
• Effective adaptation requires understanding how different emotions impact learning performance.

Purpose of the Study:

• To introduce the Affective Response Categories (ARCs) model for classifying learners based on their emotion-performance relationships.
• To investigate interindividual differences in how emotion (valence and arousal) affects performance in a simulated task.
• To revise the ARC model based on empirical findings for improved adaptive tutoring.

Main Methods:

• An experimental study with 50 participants performing a simulated airspace surveillance task.
• Emotional valence was measured via facial expression analysis; emotional arousal was measured using pupil diameter.
• Cluster analysis was used to group participants into ARCs based on emotion-performance correlations.

Main Results:

• Three distinct clusters of learners were identified based on their emotion-performance relationships.
• One cluster showed no correlation between emotion and performance.
• The other two clusters demonstrated performance benefits from negative arousal, with varying valence-performance correlations (positive or negative).

Conclusions:

• The Affective Response Categories (ARCs) model provides a framework for understanding individual differences in emotion-performance relationships.
• The findings support the revision of the ARC model, highlighting the benefit of negative arousal for performance in specific learner groups.
• This research has implications for designing more effective affect-adaptive tutoring systems, particularly in high-stakes training environments.