Puzzles, Problems, and Programs
summary
Argues that aligning mental models happens by iterative informative failure and shows it across puzzles, problems, and programs. Demonstrates with Bongard problems and Cosmic Express examples, builds a PuzzleScript game (lasers mirrors sensors) that exposes emergent skills like beam lock and names a feedback skill, measures predictions and postplay explanations from screenshots before play, and sketches solver‑aided and procedural generation tools that synthesize constraints into levels.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 1:09Bongard problems
- 3:37Research focus
- 9:13Mental models
- 15:15Mental models in games
- 19:47Puzzle games for learning
- 25:26Problem-solving games
- 30:20From puzzles to programs
- 34:11Tools for mental models
- 39:03Collaboration with automated systems
related talks
18mInferring Loop Invariants through Gamification
50mHow to Teach (Almost) Anything
48mStop Fighting Your AI Engineering Prompts That Actually Work
30mCognitive Shortcuts: Models, Visualizations, Metaphors, and Other Lies
43mA moment's reflection
35mDebugging the Mind: A Programmer's Guide to Critical Thinking
48mWhither Problem-Solving Environments?
41mUnleash your programming creativity