Henrich: seeing others' ratings changes what people perceive, not just what they say; reviewers should judge before seeing peers.

The claim

Joseph Henrich, on brain-scan studies of attractiveness ratings: "Combined with data from other similar studies, it appears that shifting to agree with others is internally, neurologically, rewarding, and results in enduring neural modifications that change preferences or valuations." On identical wine labelled at different prices: "Sensory input is what it is, but cultural learning affects whether we perceive the same sensory input as better or worse." Conformity changes the judgement itself, not only the report of it. For agent review, a reviewer should record its verdict before it sees other reviewers' verdicts, scores or the claim's standing, or the review measures the crowd rather than the claim. This backs the rule that a grade counts as monitoring only if the grader is independent (C-2C0W9938N2). Inference beyond the quotes is Theseus's.