← Writing · Nestor G Pestelos Jr

Five Checks Against Motivated Reasoning

Published August 24, 2026

TL;DR

Wanting a thing to be true and it actually being true feel identical from the inside. Five checks catch the gap: name the preference first, ask what would falsify the claim, use a confidence number, apply the same skepticism to sources you like and dislike, and treat certainty itself as a cost. Three of the five trace to specific, citable sources; two are my own synthesis, stated as such.

The trap

A fact you carefully checked and a fact you're just hoping for both arrive in your head wearing the same badge: "I'm sure." You can't tell them apart from the inside, in the moment. Your preferences are strongest exactly on the questions you care most about, which is exactly where they distort your reasoning the most. The fix isn't "trust yourself less in general." It's five specific checks that don't rely on the feeling of certainty at all.

1. Name the preference before it decides the conclusion

Write down what you want to be true before you go check. Wanting a thing is not evidence for it, and the only way to catch a preference smuggling itself into your conclusion is to have written the preference down first, separately, where you can compare the two afterward. Naming the want doesn't erase it. It stops the want from hiding inside the conclusion.

2. Ask what would falsify the claim

Karl Popper proposed falsifiability as the line between a scientific claim and an unfalsifiable one: a theory only counts as testable if some observation could, in principle, prove it false. No number of confirmations proves a theory, he argued, but a single genuine counterexample can refute one. Applied to an everyday belief, the same test works: if nothing you could observe would change your mind, the belief is functioning as an axiom, not a finding, no matter how much evidence its holder cites for it.

3. Assign a number, not a feeling

"I believe it" collapses a 60%-confident claim and a 95%-confident claim into the same sentence. Research from the Good Judgment Project, led by Barbara Mellers and Philip Tetlock, found that assigning explicit probabilities and tracking them against outcomes over time is what separates calibrated judgment from confident guessing. Ability still correlated with accuracy, but thinking style and update frequency mattered more: the most accurate forecasters revised their numbers often, in small increments, rather than making rare, dramatic calls — the same pattern I found in a related forecasting study while writing "The Addiction to Being Right Is Rarer Than You Think."

4. Check whether the skepticism is symmetric

Ziva Kunda's 1990 paper gave this pattern its mechanism. Motivation, she found, doesn't shut off critical reasoning; it changes which evidence gets scrutinized. The motivated reasoner applies heavy scrutiny to evidence against their preferred conclusion and light scrutiny to evidence for it, while still feeling like they're being appropriately rigorous the whole time. Uneven scrutiny is the signature, not an absent one: would the same piece of evidence, from a source you liked, pass your bar?

5. Hold certainty as a cost, not a virtue

The distinction that survives scrutiny runs between commitment (held firm, on purpose, because you've decided to act) and certainty about reality (held loosely, because it can turn out wrong). Treating certainty as free is what makes updating feel like a defeat instead of routine maintenance. This check is a working synthesis, not a citation, matching the first one.

Sources

Back to top