Instructor Notes — Chapter 31

Peptides in Veterinary Medicine — Where Some "Human" Peptides Were First Used Part 5 (closing chapter). Intermediate. Prerequisites: Chapters 5, 17, 18.

What this chapter is actually for

On the surface this is a chapter about veterinary medicine. It is not. It is the chapter where the book installs its inference machinery — the five mechanisms of translational failure and the four questions — and veterinary medicine is the case material because it is where students' bad reasoning is most confident.

If you have limited time, teach §31.5 and §31.6 and nothing else. Those two sections are the transferable content. Everything before them is setup and everything after is application.

The trap to avoid, and it is a real one

The chapter's argument is easy to teach badly. Taught badly, it becomes "animal evidence doesn't count," and students leave with a heuristic that is both wrong and useless — wrong because animal science produced insulin, exenatide, and leptin, and useless because it does not tell them what to do with an animal study when they meet one.

Three defenses are built into the chapter, and you should reinforce all three:

  1. §31.1 comes first, and it is not a courtesy. Veterinary medicine gets its full apparatus described before any criticism appears. Do not skip or compress it to get to the "good part."
  2. The ✅ rating exists structurally. If the chapter's four ratings were all ❌, the machinery would look like a species filter. Point this out explicitly — students who see the design choice understand the rating system better than students who just absorb the ratings.
  3. §31.9 is not a consolation section. Budget real time for it. The leptin example in particular — the model told the truth about the mechanism and misled about the application — is the single best encapsulation of the chapter's actual position.

If a student leaves able to say "animal evidence is real evidence about animals," you have succeeded. If they leave saying "mice lie," you have taught the wrong chapter.

Where students actually get stuck

The allometric scaling material (§31.5, Case Study 2). Expect two failure modes. Some students find the arithmetic trivially easy and stop there, missing that the second error survives correct arithmetic. Others get intimidated by the numbers and disengage. Both are addressed by leading with the direction of the error rather than with the magnitude: naive conversion always overestimates, never undershoots, and you know the sign before you check anything. That is a heuristic they can carry; the divisors are lookup.

The ban argument (§31.2). This one is emotionally sticky. "They wouldn't ban it if it didn't work" feels like common sense and students resist giving it up. The productive move is to grant the true part loudly — a ban tells you the compound is in circulation and that a rationale exists, which is genuine information — before dismantling the inference. Students who feel their point was heard let go of it. Students who feel dismissed argue for twenty minutes.

The "silent literature" concept (§31.3). The hardest thing for students to accept is that a large, rigorous literature can be no evidence rather than weak evidence. Case Study 1's gap table is built for this, and the "very precise thermometer is not weak evidence about your blood pressure" analogy from the chapter is the one that reliably lands.

Question 4 of the four questions. Students consistently under-rate "and collected how?" because it sounds procedural. It is the one that ends most real-world conversations. Spend disproportionate time on it.

Assessment notes

  • Exercises AE and AM are the two best written assignments. AE is instructive because it is the rare case where question 4 is satisfied and the claim still fails — good for showing that the four questions are conjunctive. AM asks students to argue against the chapter's thesis, which separates comprehension from agreement.
  • Quiz item 21 ("this chapter argues veterinary evidence is weaker than human evidence" — false) is the single best diagnostic in the set. A class that misses it in numbers has learned the wrong chapter, and you should re-teach §31.1 and §31.9 before moving on.
  • Case Study 2, question 3 is the discriminating item. The intended answer inverts most students' intuition.
  • Nothing in the chapter or its exercises requires or rewards knowledge of a dose. If a student answer contains one, that is a teaching moment about the book's own constraint, not a scoring question.

Timing

A full treatment runs two sessions. A one-session compression: §31.1 briefly (10 min), §31.5 in full (30 min), §31.6 in full (25 min), §31.9 (15 min), with §§31.2–31.4 assigned as reading and §§31.7–31.8 picked up in discussion. Do not compress §31.9 out; see above.