Treat reported confidence and actual correctness as different information.
Limited evidenceSupporting studies exist, but they are few and conditions matter.
Example
Editorial example: resolve an uncertain cost and assess comprehension separately from confidence.
Why it works
Lichtenstein and Fischhoff examine confidence against correctness on knowledge questions. Design should reveal uncertainty rather than merely raise confidence.
On the web
Offer “not sure” and clarification paths with evidence and support. Assess the gap between accuracy and confidence in learning tasks.
When it backfires
This knowledge-task research does not diagnose intelligence or personality. Hesitation over a purchase may be reasonable.
Sources
Examines calibration between correctness and confidence on general-knowledge tasks.
Lichtenstein, S., & Fischhoff, B. (1977). Do those who know more also know more about how much they know? Organizational Behavior and Human Performance, 20, 159–183. Open source