Expectation Versus Perception

What it is

The conditions on the second claim come from the advertising literature. It holds when the evidence is ambiguous, it can reach the memory of an experience afterwards, and it does neither reliably when the evidence is clear or the person is expert. Walter Lippmann's sentence from 1922, that for the most part we do not first see and then define but define first and then see, is usually quoted as though it covered both claims at once, and it names no condition under which the inversion fails. Stephen Hoch and Young-Won Ha supplied that condition sixty-four years later, and it is a severe one.

In effect

The cleanest experimental separation is the beer with balsamic vinegar in it. Leonard Lee, Shane Frederick and Dan Ariely had patrons of a pub evaluate regular beer and a doctored version in one of three conditions: blind, told before tasting, or told immediately after tasting but before indicating a preference. Disclosure of the secret ingredient significantly reduced preference only when the disclosure preceded tasting, which the authors read as the disclosure influencing the experience itself rather than acting as an independent negative input or modifying the retrospective interpretation. Set against that, the advertising work shows expectation acting on what a person concludes from ambiguous evidence, and on what they later remember concluding. Both are real, they are not the same finding, and neither licenses the other.

What it does not say

It does not say information arriving after an experience is powerless, because Kathryn Braun's post-experience work shows advertising reaching the memory of a trial. What the beer study shows is narrower and harder: in that design, with that disclosure, the verdict did not move once the tasting was done. The limits on the evidence are real. The beer paper's sample size and preference figures are not available to us, the advertising papers are closed and print no sample sizes, and the two bodies of work have never been tested against each other in a single design, so the boundary drawn here is an editorial rule inferred from separate studies rather than a measured dividing line.

---