Deliver AI-generated visual explanations adaptively based on learner engagement profiles rather than uniformly¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The article recommends moving beyond a single representation for all learners: "A single rep- resentation is unlikely to benefit all learners equally, and adaptive delivery based on, for example, engagement profiles may be crucial to realize the full potential of automated visual explanations." Because engagement profiles moderated immediate performance effects, delivery of GATs could be tailored to low-, mid-, and high-engagement learners instead of applying one modality to every student. The authors frame this as a direction for future adaptive, learner-aware strategies.

Design Implications¶

Context¶

Requirements¶

• Measurement of learner engagement profiles, e.g., via CAP self-report instruments

Constraints¶

• The moderation analysis was exploratory and relied on self-report engagement measures, so results should be treated as suggestive rather than confirmatory

Target Learners¶

• Novice programmers in introductory CS1 courses, particularly in large, heterogeneous courses

Target Learning Goals¶

• Understanding program execution and immediate post-intervention programming task performance

Related Strategies¶

• Adapt e-learning platforms to distinct student engagement profiles with adaptive recommendations and guidance
• Use CT profiles and behavioural metrics to differentiate pre-service teacher instruction and design a learning analytics dashboard
• Develop and adopt AI feedback systems that integrate visual and auditory feedback and automatically adapt students' practice plans
• Move beyond static skill-targeted rewriting toward interaction-aware adaptive error explanations

Examples¶

-

Key Sources¶

• Yuri Noviello, Naaz Sibia, Anastasiia Birillo, Thomas Overklift Vaupel Klein, Michael Liut, and Gosia Migut. 2026. AI-Generated Traces for Novice Programmers: Learning Effects and Learner Differences in a Multi-Institutional Study. Proceedings of the 31st ACM Conference on Innovation and Technology in Computer Science Education V. 1 (ITiCSE 2026). https://doi.org/10.1145/3803400.3809346