Second-Year University Physiology — Week 1 Cases
A university case session, rebuilt so students reason rather than recall
Who it’s for: Second-year undergraduate physiology students, and the instructor running the session
The work

- 15, across five parts
- ~95 minutes
- 12 — answer before the explanation opens
- Student, presentation, facilitator
- Second-year undergraduate
- Session design, at a professor’s request
Session design only. Designed at a professor’s request. The session design is mine and the course content is not, so there is no link; walkthrough on request.
In short
- Learners
- Second-year undergraduate physiology students, and the instructor running the session
- Problem
- Case material usually arrives as a slide deck. In that form a case is narrated: the lecturer poses a question, waits a moment, and supplies the answer, and the room follows along agreeing.
- My decision
- Commit before the reasoning opens. Twelve of the fifteen stages hold the explanation closed until the student has chosen an answer. This is the whole design.
- AI’s part
- AI sped up drafting and iteration. The reasoning structure, the stage gating and every call about what a second-year student can work out unaided were mine.
- Measured
- Design, structure and counts are checked against the built work. Learner outcomes have not been measured yet.
My key teaching decision
Commit before the reasoning opens
Twelve of the fifteen stages hold the explanation closed until the student has chosen an answer. This is the whole design. A student who has publicly committed to one explanation of the data reads the correction differently from one who was told the answer first — and remembers it differently.
How I used AI: AI sped up drafting and iteration. The reasoning structure, the stage gating and every call about what a second-year student can work out unaided were mine.
The thinking behind it
The whole case study: the learners, the problem, every design decision, what AI did and what it could not decide, and what has and has not been measured.
Read the thinking
Overview
A Week 1 case and help session for a second-year university physiology course, designed at a professor’s request: a session students work through rather than watch. Fifteen stages run across five parts and take about ninety-five minutes. Twelve of them ask the student to commit to an answer before any explanation opens. The session ends in a summary sheet the student can print and keep, and the same page carries a facilitator view with teaching notes and a presentation view for the lecture theatre.
The problem
Case material usually arrives as a slide deck. In that form a case is narrated: the lecturer poses a question, waits a moment, and supplies the answer, and the room follows along agreeing. Students leave believing they could have reached the same conclusion, and then cannot do it on an exam. The requirement was to convert a session that was delivered *to* students into one that cannot proceed until each student has committed to a position of their own.
Who it is for
Second-year undergraduates meeting physiological reasoning for the first time, working either in a help session or alone afterwards — and the instructor running it, who needs the same artefact to work on a projector and to carry notes only they can see.
My role
- Instructional design — a staged session with a defined reasoning method
- Learning design — the commit-then-reveal structure, the stage gating, and what each stage is allowed to assume
- Content design — structuring each case so it turns on a decision rather than a recall
- Interaction and visual design — the interface, the data displays and the three viewing modes
- Accessibility — every chart readable as a table, every state carrying a word as well as a colour
What it had to do
Objectives the design set out to meet. They are aims, and none is reported here as a result.
- Make the student state a position before the explanation is available to them
- Keep the course’s content and sequence intact — the design changes the form, not the substance
- Give the instructor one artefact for the projector, the help session and the student working alone
- Let a student stop partway through and come back without losing their reasoning
- End on something the student keeps, rather than on a completion message
How it works
One page holds the whole session. Stages unlock in order, so a student cannot read ahead to the explanation before committing — the gating is the pedagogy, not a restriction. Progress is held in the student’s own browser: nothing is submitted, nothing is scored, and nothing is reported to the course. The same page switches between a student view, a presentation view sized for a lecture theatre, and a facilitator view that adds teaching notes without a second document to keep in step.
Design decisions
Commit before the reasoning opens
Twelve of the fifteen stages hold the explanation closed until the student has chosen an answer. This is the whole design. A student who has publicly committed to one explanation of the data reads the correction differently from one who was told the answer first — and remembers it differently.
Five questions that keep returning
The session opens by naming the method rather than the content: what changed, what caused it, what determines the response, what limits it, what happens next. Every case afterwards is worked with those same five questions, so students leave with a procedure they can apply to a case they have never seen.
Real data, not illustrations
The cases use dose-response curves and uptake assays the student reads and interprets. Each chart is also available as a table, because a student who cannot read the chart is not a student who cannot read the physiology.
One artefact, three readings
Rather than a student handout, a lecturer deck and a facilitator guide that drift apart after the first revision, there is one page with three views. A change to a case is made once and is correct in all three.
What I knew from teaching
My judgement was about form. Which moments are genuinely decision points and which only look like them; where a student will guess rather than reason and how to close that off; how long ninety-five minutes actually is when every stage asks for a commitment. The hardest call was restraint: a case that has been made interactive everywhere stops being a case and becomes a quiz.
AI’s role
Drafting and iteration speed. The reasoning structure, the decision about what each stage may assume, and every judgement about what a second-year student can be expected to work out unaided were mine.
What was hard
Converting a narrated case without losing the narration. A lecturer builds tension across a case — here is a result, here is a second result that contradicts it, now what? That tension is easy to destroy by fragmenting the material into exercises. The structure had to preserve the arc while making the student the one who resolves it.
What changed along the way
Two earlier builds treated the case and the underlying principle as separate pages. Both failed the same way: students met the principle with no reason to care about it, then met the case with the principle already forgotten. The final version interleaves them — the principle arrives at the moment the case requires it.
Looking back
This is the clearest evidence I have that the design method transfers. The subject is undergraduate physiology, not primary curriculum, and the same structure held — commit first, open the reasoning second, end on something the learner keeps. The most useful thing a design like this contributes is not content; it is the decision about where a learner is made to think.
Skills
Let’s build better learning.
Hiring for a teaching, curriculum or learning-design role, or need curriculum writing, e-learning, a teacher workshop or tutoring? Tell me what you are trying to make possible.


