Much current work assumes that the main advantage of interactive visual explanations over static materials comes from within-visual variable manipulation; if we instead treat the interactive as a calibration probe and route most effort into spaced, non-interactive retrieval (e.g., paper-based or low-tech prediction and explanation tasks that reference but do not present the visual), are there learner profiles or course contexts where a “minimal-interactive, heavy-retrieval” design matches or exceeds full interactive use on durable learning and far transfer, suggesting that interactivity is over-allocated time rather than a uniquely powerful mechanism?

Answer

Yes. For some learners and topics, a minimal-interactive, heavy-retrieval design will likely match or beat full interactive use on durable learning and far transfer, implying that interactivity time is sometimes over-allocated rather than uniquely causal.

Most favorable cases:

• Learners: intermediate or higher prior knowledge, at least moderate self-regulation; prone to “play” the interactive without deep reflection.
• Topics: moderately visualizable, with a small set of core relations that reappear across contexts (functions, simple dynamics, basic probability, conservation laws).
• Courses: enough calendar time to schedule multiple spaced, non-interactive retrieval episodes that reference earlier visuals but do not re-show them.

Less favorable/risky cases:

• Very low prior-knowledge, fragile, or highly anxious learners who need in-visual scaffolding and concrete feedback to build an initial model.
• Highly opaque or symbol-heavy topics where the visual itself supplies critical structure learners cannot yet reconstruct from memory.

Mechanism sketch:

• A brief, well-structured interactive acts mainly to establish a reference schema and reduce initial confusion.
• Most gains in durable learning come from spaced prediction/explanation tasks that force retrieval and recombination of that schema without access to the visual, plus occasional calibration checks with the interactive.
• In such settings, extra manipulation time yields diminishing returns relative to additional retrieval cycles.

So: for reasonably prepared, regulated learners on mid-complex concepts, a “minimal-interactive, heavy-retrieval” design is a plausible competitor to full interactive use. For fragile novices and highly complex visuals, sustained interactivity remains more important.