Grounded in cognitive learning-transfer theory—which distinguishes automatic “low-road” transfer, or similarity-based, from mindful “high-road” transfer, principle-based—this study examined whether strategically designed micro-learning activities can facilitate knowledge from one context to complex problem spaces in high-load university courses. In an Introductory Statistics course, micro-learning units were 2- to 7-minute AI-generated video scenarios followed by retrieval-practice and elaboration-encoding exercises that prompted learners to identify underlying principles, compare isomorphic problems, and enact solution strategies—key mechanisms known to promote both near and far transfer. The videos were generated with Adobe Captivate Classic.

Using a classic pretest-post test control-group design (N = 70), students in experimental condition engaged voluntarily with one micro-learning unit before each weekly assignment, while the control group accessed only the standard course materials (common syllabus, textbook, recorded lectures, and assessments). Outcome measures included cumulative grades, time-on-task analytics, discussion response analytics, participation frequency, and depth of cognitive processing mapped to Bloom’s taxonomy.

Results indicated a 34 percent higher mean course grade for the experimental group (p < .01), accompanied by significantly greater time-on-task and a two-fold increase in discussion posts coded at the “analyze,” “evaluate,” and “create” levels. These findings align with transfer theory predictions: micro-learning reduced extraneous cognitive load, provided distributed retrieval opportunities, and made structural similarities between tasks explicit, thereby enabling learners to abstract and reapply core concepts across contexts.

Implications include the systematic embedding of micro-learning sequences in cognitively demanding courses and the use of generative AI to customize transfer-oriented prompts. Future research should isolate which design features (e.g., analogical comparison vs. spacing) most powerfully mediate high-road transfer.

Keywords: Micro-learning, Cognitive Transfer, AI-generated Design.