Capstone Projects
Now that you are an expert at automating instructional design you get to show your instructor and your classmates what you can do. This chapter only contains high level guidelines on what types of projects you might take on. It also makes sense for larger projects to have teams of two or more people working on these projects.
Suggestions for Success
For a capstone that combines Intelligent Textbooks + MicroSims, the biggest risk is not technical failure—it’s diffuse scope and shallow integration. The guidance below is designed to help teams of 3–5 students focus on learning impact, coherence, and evidence of thinking, not just shiny demos.
1. Start With a Learning Spine, Not a an Entire Textbook
Before any writing or coding, require teams to define a learning spine:
- Target learner (age, background, constraints)
- One core transformation: “Before this book, the learner could not ; after, they can .”
Have them express this as a single sentence and get it approved.
Why: This prevents the “Copy a Wikipedia article and add random sims” failure mode.
2. Define a Concept Dependency Graph Early
Ask teams to explicitly create a concept dependency graph (even a rough one):
- 15–30 concepts total (30-60 concepts for a course)
- Clear prerequisites
- No cycles
- One capstone concept at the far right
This graph becomes the table of contents, not the other way around.
Rule of thumb: If a concept does not enable another concept, question whether it belongs in the project.
3. Map Each MicroSim to a Cognitive Purpose
Require that every MicroSim declare its instructional role, not just its behavior.
For each MicroSim, teams should state:
- Which concept(s) it supports
- Bloom level(s) it targets (e.g., Understand → Analyze)
- What misconception it helps surface or correct
Example:
“This MicroSim targets Understanding and Applying Ohm’s Law by allowing learners to vary resistance and observe non-linear power effects.”
Anti-pattern to avoid: “Fun but conceptually orphaned” simulations.
4. Design MicroSims as Experiments, Not Animations
Encourage teams to treat MicroSims as interactive experiments:
- At least one controllable variable
- One observable outcome
- One learner question the sim helps answer
Ask teams to explicitly state:
- What should the learner predict before running the sim?
- What should surprise them?
This aligns naturally with PRIMM-style learning without naming it explicitly.
5. Require Accessibility as a First-Class Constraint
Before coding begins, teams should answer:
-
How can this MicroSim be explored with:
-
Keyboard only?
- Reduced motion?
- Screen reader narration?
- What textual feedback mirrors the visual state?
This reframes accessibility as design intelligence, not compliance.
6. Limit the Scope on Purpose
Give teams explicit limits:
- 5–7 chapters maximum
- 1–2 MicroSims per chapter
- One “hero” MicroSim they polish deeply
Why: Depth beats coverage. A well-integrated MicroSim is more impressive than five shallow ones.
Note
You will not receive a lower grade for an incomplete textbook. We encourage teams to push themselves for depth, not breath.
7. Encourage Clear Team Roles (but Shared Ownership)
Suggest—but do not rigidly enforce—roles:
- Learning Architect (concepts, flow, assessments)
- MicroSim Designer (interaction, experimentation)
- Content Curator (writing, examples, clarity)
- Integrator (MkDocs structure, navigation, polish)
Make it clear:
- Everyone must contribute to at least one MicroSim
- Everyone must contribute to concept modeling
8. Require an Explicit “AI Usage Disclosure”
Since this is an intelligent textbook, ask teams to document:
- Where AI was used (drafting, concept extraction, sim ideation, feedback)
- Where human judgment overrode AI output
- One mistake the AI made and how it was corrected
This reinforces critical AI literacy, not passive use.
9. Define Success in Terms of Learner Evidence
Ask teams to include:
- 3–5 learner tasks or challenges
- What success looks like for each
- How the MicroSims support those tasks
Optional but powerful:
- A short “If we had more time” reflection identifying the next MicroSim they would build.
10. End With a Narrative, Not a Demo
For final presentations, encourage teams to:
- Walk the audience through a learner journey
- Show confusion → exploration → insight
- Use one MicroSim live, not many slides
The goal is to demonstrate:
“We understand how people learn with interactive systems.”
A Strong Framing Sentence You Can Give Students
“Your capstone is not a book with simulations attached—it is a learning system where text, interaction, and feedback work together to change how someone thinks.”
If you’d like, I can turn this into:
- A one-page capstone brief for students
- A grading rubric aligned to learning impact
- A starter MkDocs + MicroSim template that bakes these constraints in