Perception and Misperception in International Politics (1976) by Robert Jervis applied cognitive psychology to intelligence analysis and foreign policy decision-making, demonstrating that the biases producing intelligence failures — mirror-imaging, anchoring, consistency-seeking, premature cognitive closure — are not individual failings correctable by training or discipline but structural features of human cognition operating under adversarial uncertainty.

Core argument

Jervis documented systematic patterns of misperception in international relations: decision-makers overestimate the degree to which the adversary’s actions are centrally planned and directed (the “coherence” bias); they assimilate new information to existing beliefs rather than updating their beliefs (the “consistency” bias); they project their own strategic logic onto the adversary (mirror-imaging); and they interpret ambiguous evidence as confirming their pre-existing assessments (confirmation bias).

The crucial implication for intelligence is that these biases are structural — they arise from how human cognition processes information under conditions of uncertainty, threat, and incomplete access. Individual analysts can be trained to recognize biases, but the biases persist at the institutional level because they are features of the cognitive process itself, not errors in its application. This finding motivated Richards Heuer’s development of structured analytic techniques designed to counteract specific biases through procedural discipline.

Influence

Jervis’s framework fundamentally changed how the intelligence discipline understood its own analytical limitations. The concept of perception and misperception is now a foundational concept in the discipline, and the cognitive biases he documented are standard elements of analyst training. His work also influenced the development of the structured analytic techniques curriculum — the discipline’s primary methodological response to the cognitive bias problem.

Limitations

Jervis’s framework focuses on the analyst’s cognitive limitations — how the analyst misperceives the adversary. The 2026 Iran war analysis extends this critique from cognitive to structural: even debiased analysis operates within categories that may not capture the adversary’s relevant properties. The problem is not only that the analyst thinks incorrectly (Jervis) but that the system’s categories exclude what matters (legibility).