Instructional affect and learner motivation in generative AI-restrictive and permissive classrooms

Abstract

This study investigates the impact of classroom policies on generative artificial intelligence (AI) on students' instructional affect and motivation within a Philippine higher education context. It compares student experiences in generative AI-permissive vs. AI-restrictive sections of a university-level Public Speaking and Persuasion course. Using quota sampling, 156 undergraduate students were surveyed via validated instruments: McCroskey's Instructional Affect scale and Jones' MUSIC Model of Motivation Inventory. Results reveal that students in AI-permissive classrooms reported higher levels of positive instructional affect, expressed as greater trust and engagement with both instructors and course content, alongside greater motivation across the MUSIC dimensions (eMpowerment, usefulness, success, interest, and caring). Conversely, restrictive environments were associated with lower instructional affect and motivation, potentially stifling engagement and perceived autonomy. The findings underscore the significance of clear, supportive generative AI policies that foster student agency and positive emotional responses, while addressing ethical concerns. The study also provides actionable insights for educators and policymakers to design classroom environments that optimize the benefits of generative AI, enhance student motivation, and maintain academic integrity in the evolving landscape of higher education.