Foundations of Decision Making
What this domain tests
Candidates must distinguish a good process from a good outcome. A disciplined decision process (define the problem, clarify objectives, generate alternatives, gather information, evaluate trade-offs, decide, then review) can produce a bad outcome by chance, and a sloppy process can get lucky. Questions often reward the process answer and punish reasoning that judges a decision only by how it turned out.
Know the vocabulary: bounded rationality (people satisfice because search and computation are costly), satisficing (choosing the first option that clears an acceptable threshold), anchoring (over-weighting an initial number), confirmation bias (seeking supportive evidence), sunk cost (irrelevant past spending), overconfidence (too-narrow ranges), and framing (wording shifts preferences).
Traps in questions
- Outcome bias: "The project failed, so the manager decided badly." Outcome alone does not establish process quality.
- Sunk-cost trap: continued investment justified by money already spent.
- Hindsight: calling a decision obviously wrong once the result is known.
- False dichotomy: presenting two options when a third exists.
Worked example
A team spent 500,000 on a product that now forecasts only 200,000 in future revenue with no further spending. Continuing requires another 100,000 and would yield 120,000. The sunk 500,000 is irrelevant. Marginal test: 120,000 minus 100,000 equals +20,000, so continue. A question that says "stop because we already lost 500,000" is testing the sunk-cost trap.
Rule of thumb
Ask three questions before answering: What was the decision-maker's information at the time? Was the process systematic? Would the same logic work on a fresh case? If yes to the last two, choose the process answer even when the described outcome is poor.