---
title: "Irwin & Mandel's (2023) expert analysts inferred higher confidence from probability intervals located above 50% than from equally-wide intervals below 50%"
type: "claim"
status: "seedling"
audit_status: "capture-verified | 2026-08-13 cross-model audit (claude-opus-5, writer was claude-sonnet-5): source_quote found verbatim in the accepted manuscript (Results 2.2.4.1, expert sample), together with every statistic the note reports — experts Wilcoxon signed-rank Z = -4.69, p < .001; non-experts in Experiment 1 Z = -14.56, p < .001; the independent Experiment 2 sample of 624 Z = -15.17, p < .001. The paired-assessments setup is as the note describes: the 20%-40% and 60%-80% intervals, an identical 20-point spread, both equidistant from the 50% midpoint. This DISCHARGES the open NO MATCH flag raised against this note by seek_verify in 00-meta/reports/verify-2026-08-12.md, which classed the quote as 'artifact/fabrication class' — the cause was URL drift, not fabrication (see below). POINTER CAVEAT: source_url now 301s to https://osf.io/preprints/psyarxiv/hwp5r, whose primary file is a .docx, so the URL resolves to the preprint record but no longer serves a fetchable PDF, and the Wiley version of record returns HTTP 402 to automated fetch. An independent re-download was therefore not possible and the status is deliberately NOT upgraded to verified-verbatim; source_venue added so the citation reaches the version of record."
source_url: "https://psyarxiv.com/hwp5r"
source_author: "Daniel Irwin, David R. Mandel"
source_date: 2023
source_venue: "PsyArXiv preprint (accepted manuscript, marked 'In press: Risk Analysis'); version of record: Risk Analysis 43(5):943-957, 2023, doi:10.1111/risa.14009"
source_quote: "experts did not separate the location of the probabilities from the width of the intervals. Specifically, they judged the range below 50% as indicating mainly low or moderate confidence, whereas they judged the range above 50% as indicating mainly moderate or high confidence."
source_tier: 1
source_sha: "2848a23bbd421bf10eb93d5daa33781dadf8d515b7abd8a3f23c4c43eee89058"
provenance: "Promotion from 10-inbox/raw/2026-08-11-does-irwin-mandel-2023-actually-show-that-intelligence.md, 2026-08-11"
origin: "batch"
writer_model: "claude-sonnet-5"
derived_from: ["10-inbox/raw/2026-08-11-does-irwin-mandel-2023-actually-show-that-intelligence.md"]
date_created: "2026-08-11T00:00:00.000Z"
tags: ["intelligence-tradecraft","epistemics","forecasting","independence","probability-confidence","david-r-mandel"]
seek_code_commit: "b13747c"
---


In a second task within the same study, the 41 expert intelligence analysts were shown two numeric probability intervals with an *identical* 20-point spread (20%–40% and 60%–80%, equidistant from the scale's midpoint) and asked what confidence level — low, moderate, or high — each communicated. If confidence genuinely tracked only interval width, the two should read identically. They did not: "experts did not separate the location of the probabilities from the width of the intervals. Specifically, they judged the range below 50% as indicating mainly low or moderate confidence, whereas they judged the range above 50% as indicating mainly moderate or high confidence" (Wilcoxon signed-rank test, Z = −4.69, p < .001). Non-expert samples showed the identical pattern at larger scale (n = 440, Z = −14.56, p < .001; independent n = 624 sample, Z = −15.17, p < .001).

Together with [[claim-irwin-mandel-2023-confidence-shifts-inferred-probability|the forward-direction finding]] — that stated confidence shifts inferred probability — this establishes the conflation runs both ways: confidence functions as a de facto probability cue, and probability level functions as a de facto confidence cue. This is the specific mechanism behind [[claim-kelly-2025-odni-probability-confidence-guidance-incoherent|Kelly et al.'s summary]] that "analysts and nonexperts alike do treat probability and confidence as related constructs," and the second half of what resolves [[question-verify-irwin-mandel-2023-probability-confidence-cue]].

> [!note] Seek's commentary:
> The asymmetry is the interesting part: it isn't that confidence and probability blur into an undifferentiated mush, it's that they blur in a specific, directional way — high numbers read as more trustworthy numbers, on both sides of the manipulation. That's a bias with a sign, not just noise, which is a stronger claim than "independence fails" on its own and probably the more actionable one for anyone trying to write elicitation instructions that survive contact with an analyst.
> — Seek