Threat-related AI anxiety and engagement in live-streamed AI courses: a moderated serial mediation model of extrinsic motivation and teacher support

Read full text
Authors: Ming Lei, Yue Zou, Xin Wang
Cited by: 4
References: 33

Abstract

Background: As artificial intelligence becomes increasingly embedded in education and work, live-streamed AI courses have become a common format for timely, interactive, and practice-oriented instruction. At the same time, AI-related uncertainty has intensified learners' anxiety. This study examined how threat-related AI anxiety is associated with engagement in live-streamed AI courses through extrinsic motivation for AI learning and perceived continuous teacher support, and whether this process varies by AI learning self-efficacy. Methods: A questionnaire survey was conducted among 487 adult learners in China who had participated in at least one live-streamed or synchronous AI-related course within the previous 6 months. Data were analyzed using PROCESS Model 87 with 5,000 bootstrap samples. Results: Threat-related AI anxiety was positively associated with extrinsic motivation for AI learning. Extrinsic motivation was positively associated with both perceived continuous teacher support and engagement, and perceived continuous teacher support was positively associated with engagement. The serial indirect effect through extrinsic motivation and perceived continuous teacher support was significant. AI learning self-efficacy significantly weakened the association between perceived continuous teacher support and engagement, such that learners with lower self-efficacy depended more heavily on teacher support for their engagement. Conclusion: Engagement in live-streamed AI courses is associated with a linked process involving emotional pressure, motivational activation, and instructional support. The study extends research on AI anxiety from motivation to engagement, identifies continuous teacher support as a key mechanism in live-streamed AI learning, and highlights the importance of accessible and supportive course design, especially for learners with lower AI learning self-efficacy.

Fields of Study & Subjects

Teaching and Learning Programming Ethics and Social Impacts of AI Online Learning and Analytics

Publication details

Journal / source
Frontiers in Psychology
Publisher
Frontiers Media SA
Volume
17
Pages
1831785–1831785
Publication date
29 April 2026
Language
EN