Creativity, Math & Play
Learning by Experimenting, Not Just Reading
Also called: Interactive learning
- Personal interest
- Philosophical question
- Working interpretation
The best learning tools let people change things and discover the relationships for themselves. I like replacing explanations that people only read or watch with experiments they can run themselves.
Which idea would suddenly make sense if a learner could change just one setting?
Why it attracts me
Reading that two settings work together is easy to forget. Watching your own experiment miss the answer because they work together is hard to forget. I like building tools where the learner runs the experiment.
An example
The Factorial Experiment Lab on this site starts with a heat-sealing machine making weak seals. Three settings could be to blame: temperature, how long the jaws stay closed, and pressure. You can try the classic approach first: start with every setting low, change each one alone from there, and keep whatever helped. In the lab’s built-in process, a longer closing time usually looks harmful at the lower temperature, so that approach keeps it short. It never tries the higher temperature with the longer time, so it cannot see the two settings helping each other. Then you run all eight combinations and see the effect directly. The lesson arrives as a surprise you caused, not as a rule you were handed.
What I think (and don't know)
Interaction is not automatically learning. A slider can be fun without teaching anything. The design work is choosing the one setting whose change reveals the idea, and I don’t yet know how to test whether that worked.
Where it connects
Learning Through Play is the same idea for young children: discovery that feels like play rather than a test. Charts and Diagrams as Thinking Tools is the picture side of this. Causes Versus Things That Happen Together is why experiments like this matter: changing settings on purpose is how causes are separated from coincidences.
Questions I'm still exploring
- How do I know whether a learner understood the idea or just played with the controls?
- Which ideas are better taught by explanation than by experiment?
- How much guidance can an experiment have before it stops being a discovery?
Sources and further reading
- Seymour Papert, Mindstorms: Children, Computers, and Powerful Ideas (Basic Books, 1980) — Argues that children learn deeply by building and experimenting.
- Bret Victor, “Explorable Explanations” (2011) — An essay on documents that let readers change assumptions and explore the author’s model.
Working interpretation: drafted from my notes and interests for review. It is not a direct quotation, and I may still change it.