Cognitive Load and Learning-Related Outcomes in Article-Based Pharmacology Learning with an Instructor-Provided Highlighting with Structured Question Prompts Tool: A Classroom Evaluation

Abstract

Objective: To evaluate the use of an Instructor-Provided Highlighting with Structured Question Prompts (IHSQP) tool in article-based learning in a Principles of Pharmacology course by examining IHSQP use by learners, cognitive load, learning-related outcomes. Methods: This study was a single-group post-activity educational evaluation. Participants were 107 second-year pharmacy students enrolled in the Principles of Pharmacology course in the second semester of the 2025 academic year. Students received the IHSQP tool in advance as a reading scaffold for an English-language article on glaucoma pharmacotherapy. Data were collected using an online questionnaire to assess IHSQP use, four subdimensions of cognitive load, retrospective pre–post perceived understanding. All scales were measured on a 7-point Likert scale.  The researchers also collected posttest scores and recorded artificial intelligence (AI) usage as a contextual variable. Results: The mean IHSQP use score was 4.05 ± 1.17. The mean score for artificial intelligence use and its perceived usefulness was 5.28 ± 0.98. Germane load had the highest mean score (5.29 ± 0.84), followed by intrinsic load (5.18 ± 0.77), language-related extraneous load (4.84 ± 0.98), and design-related extraneous load (3.89 ± 0.89). Median perceived understanding after the activity was significantly higher than that before the activity (with difference of median at 1.20; IQR 0.80–2.40), P < 0.001) The mean self-reported posttest score was 7.47 ± 1.62 out of 10. No significant associations were found between cognitive load dimensions and learning-related outcomes. Conclusion: The classroom evaluation indicates that IHSQP can serve as a reading scaffold to direct learners’ attention to key evidence and support information integration for pharmacological reasoning. Design-related extraneous load was lower than language-related extraneous load and intrinsic load. However, because this was a single-group evaluation and learners used artificial intelligence alongside the activity, the findings should be interpreted as classroom evaluation data influenced by multiple factors, rather than as definitive causal outcomes attributable solely to IHSQP.