Facilitator Notes & Model Answers — Lesson 13
For instructors and self-study · Model walk-through and rationale
Model decision walk-through
| Step | Model reasoning |
|---|---|
| Evidence vs inference | Known: an alert fired; two analysts disagree. Inferred: it is/ isn't real; the sector campaign applies. Keep the two apart |
| Base rate | Acknowledge many such alerts are false, but the cost of a missed real one is high — asymmetric |
| First move | A reversible protective step (e.g., increase monitoring, restrict the suspect access) while verifying in parallel |
| One-way doors | Defer public statements and attribution until certainty improves |
Bias checklist (model)
- Confirmation — am I only collecting what fits my first theory?
- Anchoring — am I over-weighting the first claim or number?
- Sunk cost — am I persisting because of past investment?
- Groupthink — is the room agreeing to avoid friction?
- Normalcy — am I assuming it can't be serious?
Why the poll answer is C (reversible action + verify in parallel)
- A — wait for certainty: paralysis; damage spreads while you seek proof.
- B — announce a major breach: panic; an irreversible move on unverified information.
- C — reversible protective step + verify: matches speed to reversibility; no regret under either outcome.
- D — assume false alarm: normalcy bias; the most dangerous early error.
Teaching point: the two failure modes are paralysis and panic. Discipline lives in the reversible action taken while investigating.
Homework — what good looks like
Task 1
Strong checklists specify a no-regret first action and a self-known bias, not generic "stay calm."
Task 2
The best thresholds are testable in the moment ("two independent confirmations of X"), decided while calm.
Task 3
Good answers pair an honestly identified bias with a concrete ritual to counter it next time.