This CIA Manual Trains the World’s Sharpest Analytical Minds

Stephen Petro · 13:33 · Five thinking traps from Richards Heuer’s Psychology of Intelligence Analysis — once-classified CIA analyst training — with the counter-technique for each.

Trap 1: Mirror imaging

• 1998: India detonated five nuclear devices; CIA missed it despite satellite evidence captured 6h before and open BJP campaign promises to test. Postmortem (Adm. David Jeremiah): “everybody thinks like us.”
• Analysts ran US cost-benefit logic (sanctions too costly); BJP ran prestige/nationalist logic.
• Counter: before judging what another person/org will do, write “If I were them, I would…” then cross it out. Ask instead: what actual evidence do I have about how they think, their values, their constraints?
• Business version: predicting competitors by your own strategic logic; calling difficult people irrational when they’re rational by different priorities.

Trap 2: Satisficing (confirmation bias)

• Aug 1978: CIA assessed Iran “not in a revolutionary or even pre-revolutionary situation.” Locked onto “regimes with loyal militaries don’t fall,” then only gathered confirming evidence while mosques organized, bazaar merchants funded revolution, oil strikes spread.
• Wason experiment: given 2-4-6, nearly everyone tests only sequences fitting their hypothesis (“ascending even numbers”) and never tries disconfirming ones like 3-5-7 that reveal the real rule (“any ascending numbers”).
• Counter: before finalizing a judgment answer “What evidence would cause me to change my mind?” If you can’t generate an answer, your mind is already closed.

Trap 3: Analysis of Competing Hypotheses (ACH)

• Heuer’s core tool, still standard in Western intel. Reverses normal reasoning: list all plausible hypotheses at once; score each piece of evidence consistent/inconsistent against each; pick the survivor.
• Winner = fewest disconfirmations, not most confirmations. Confirming evidence is usually weak (consistent with many hypotheses); one solid disconfirmation can kill a hypothesis — the fever-vs-no-fever diagnostic asymmetry.
• Practical form: 3 hypotheses across the top × 3 key pieces of evidence down the side; mark cells; work the inconsistencies.

Trap 4: The vividness criterion

• Vivid/concrete/dramatic beats abstract statistical evidence, even when less reliable.
• Same surgeon-general data on smoking reached all US doctors — but radiologists and oncologists (who see the consequences daily) quit most; psychiatrists and dermatologists kept smoking longest.
• Vietnam: Westmoreland’s vivid “we’re winning” frame absorbed every contradicting signal (200% Ho Chi Minh Trail truck traffic spike, captured offensive plans) until Tet, Jan 30 1968.
• Counter habit: when moved by anecdote or narrative ask — (1) what’s the base rate? (2) is this case representative or a vivid outlier?

Trap 5: The information paradox

• More information raises confidence without raising accuracy. Horse-handicapper study: gave handicappers 5/10/20/40 variables — accuracy flat (three got worse), confidence climbed steadily. At 40 variables they were badly overconfident while no more correct.
• Iraq WMD 2003: enormous data, extremely high confidence, catastrophically wrong model — extra data reinforced the wrong frame instead of correcting it.
• Counter: before gathering more, ask if you already have the minimum data needed. Then the real question shifts from “what else do I need to know?” to “is my way of thinking about this problem the right way?”

Common thread

All five traps are invisible and unconscious — the analysts who failed India and Iran weren’t careless or lazy, they were trying hard inside flawed processes. Heuer offers no guarantee of being right; he offers disciplines that systematically reduce the probability of being wrong in the ways that matter.