nucleation

LIT-1187: The construction of preference.

2 min read

Properties4
StatusActive — read 2026-10-16 (NOTE-949) in full, from a scan of the published article on a University of Florida course page, compared with the author's manuscript in the University of Oregon's ScholarsBank; worth reading as the compact review of preference reversals: normatively equivalent procedures give different orderings, about 90% of reversals violate procedure invariance rather than transitivity, and scale compatibility and prominence explain them. Read it knowing that its evidence compares procedures and never follows a constructed preference into a later judgement, so it shows the elicitation entering the response, not changing the judge; that construction is claimed for important, complex and unfamiliar options; and that the only remedy it reports is market repetition with feedback and penalties.
Version2
Tagscognition, motivation
Cited as “paper” byNOTE-949 — Slovic — the construction of preference

Graph View

Slovic, P. (1995). American Psychologist 50(5):364–371. doi:10.1037/0003-066X.50.5.364.

Key takeaways

From the reading, NOTE-949.

  • Reversals. Choice and pricing, or choice and matching, order the same options differently. Las Vegas: equal choice, the $ bet priced higher about 88% of the time (p. 365). Highway safety: 67% choose the program that saves more lives, 4% of choices inferred from matching do (p. 368).
  • Not intransitivity. About 90% of reversal patterns violate procedure invariance, 10% are intransitive; overpricing the high-payoff bet is nearly two thirds (p. 367).
  • Mechanisms. Scale compatibility, an attribute weighing more when it shares the response’s scale (reversals 41% with money, 24% without), and prominence, the important attribute weighing more in choice than in matching (pp. 367–369). Both are effects of the response mode on how the answer is computed.
  • Construction. For “important, complex, and perhaps unfamiliar” options, preferences “are constructed on the spot” (p. 369).
  • Remedy. Reversals were eradicated only “in market settings characterized by repetition, feedback, and harsh penalties for being inconsistent” (p. 366).

Why this is here

Registered for CLAIM-341 as the decision-research form of “a measurement that generates interpretations”, with the open question whether the elicitation changes the judge and not only the response. The reading answers it for this paper: it shows only the response. Every result compares procedures, and the one sequential design it mentions, matching and then choosing in a later session (p. 368), is analysed for the discrepancy, not for carry-over. So it is evidence for the measurement context entering the response, the route TERM-039 attributes to A129’s model, and leaves the route into the interpreter’s state open. It supports the synchronic form, which CLAIM-341 sets aside, and the claim’s diachronic form only indirectly, by showing that a judgement is a construction. The mechanisms are classical. On the anchor of CLAIM-342, the one remedy it reports is outcome feedback with penalties, which is independent scoring arriving through the consequences of the respondent’s own choices. It carries no instruction for machine-learning practice.

Backlinks