Second-order metacognitive judgments of subjective uncertainty show a small high-confidence error/correct difference, smaller than curiosity ratings show¶

Claim · All claims
Evidence · 1 study · causal r2 · q2
Currency · newest study 2023 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? When asked how surprised they would be to be wrong, participants showed slightly greater expected uncertainty for high-confidence errors than high-confidence correct answers, but the effect was small. → Metcalfe 2023

Evidence¶

Metcalfe 2023¶

Metcalfe, J., Vuorre, M., Towner, E., & Eich, T. S. (2023). Curiosity: The effects of feedback and confidence on the desire to know. Journal of Experimental Psychology: General, vol. 152, no. 2, pp. 464-482. https://doi.org/10.1037/xge0001284

q2 · i? · causal · r2

Experiment 3, 48 online adults made a second-order judgment (100 minus expected surprise at being wrong) after each confidence rating with no feedback. "high-confidence errors elicited slightly greater subjective uncertainty" (Difference = 8.02, pd > 99.9%), a smaller difference than curiosity ratings showed in Experiment 2.

"A post hoc comparison on subjec- tive uncertainty at con fidence = 90, between correct and incorrect answers indicated that high-confidence errors elicited slightly greater subjective uncertainty than did high-con fidence correct"

Discussion¶

Related Claims¶

• Curiosity ratings are functionally distinct from subjective uncertainty ratings: uncertainty declines near-linearly with confidence while curiosity is inverted-U shaped — related
• Without feedback, people are more curious about high-confidence errors than about equally high-confidence correct answers, replicated across experiments — related
• Yes/no feedback that one was wrong increased curiosity selectively for high-confidence errors, leaving curiosity about low-confidence errors largely unchanged — related
• Yes/no feedback that participants were correct nearly quenched curiosity for correct answers, whereas without feedback curiosity peaked at moderate-to-high confidence — related
• Social comparison information affects the magnitude of metacognitive judgments, with greater magnitude for higher peer performance — related
• Adults' confidence judgments track answer accuracy: confidence is higher on correct than incorrect general-knowledge trials (metacognitive monitoring) — related