Festinger's Cognitive Dissonance and Kunda's Motivated Reasoning
Leon Festinger's cognitive dissonance theory, introduced in his 1957 book, proposed that humans experience psychological discomfort when they hold two contradictory beliefs simultaneously, and that the primary response is not honest self-examination but rationalisation. His field study — infiltrating a UFO cult whose prophecies had failed to materialise — found that rather than abandoning the belief, members doubled down and became more fervent proselytisers after the predicted event did not occur. Ziva Kunda, whose 1990 paper The Case for Motivated Reasoning in Psychological Bulletin is one of the most cited in social psychology, described the phenomenon precisely: people do not reason toward conclusions they want; they reason from conclusions they want.
This explains why highly intelligent people are often more effective at self-deception than less intelligent ones. Intelligence improves the quality of the post-hoc rationalisation. The smart person constructs a more elaborate and internally consistent justification for the belief they wanted to hold. The mechanism is the same; the execution is more sophisticated. The desire to believe something triggers a search for evidence that supports it, and the search stops when sufficient evidence has been found — not when a comprehensive assessment has been made. The conclusion was reached first; the reasoning was assembled afterward.
The Pre-Mortem and Adversarial Collaboration
Pre-mortems — introduced by Gary Klein and popularised by Daniel Kahneman — are the single most evidence-supported technique for disrupting motivated reasoning in advance. Before a major decision, imagine that one year has passed and the decision has failed catastrophically. Ask: what most plausibly caused the failure? The exercise forces generation of reasons against the preferred option, which the mind would otherwise suppress. Research by Klein shows pre-mortems increase the number of potential failure modes identified by 30 percent compared with standard prospective analysis. The adversarial collaboration — structuring decision review to include a genuine advocate for the opposing position with incentives to argue it effectively — is the institutional version that Amazon has formalised as disagree and commit.