Low-confidence info-seeking trials show inverted decision weights, unlike explicit reports and post-decision wagering¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r2 · q3 · i2 medium
Currency · newest study 2022 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i2 A significant three-way interaction of condition, selection and confidence level occurred because decision weights were reversed for low-confidence info-seeking trials (generalized eta squared = 0.08). → Lorenz Weise 2022 q3 i2 On low-confidence info-seeking trials the non-selected option was positively associated with choice (Cohen's d = 0.56) and the selected option negatively (Cohen's d = -0.87). → Lorenz Weise 2022 (2)

Evidence¶

Lorenz Weise 2022¶

Lorenz Weise, Saskia D. Forster, Siegfried Gauggel. (2022). Reverse-correlation reveals internal error-corrections during information-seeking. Metacognition and Learning (2022) 17:321–335. https://doi.org/10.1007/s11409-021-09286-4

q3 · i2 · causal · r2

Repeated-measures ANOVA (3 condition × 2 selection × 2 confidence) on decision weights from 27 participants. The article reports the three-way interaction with F(2, 52) = 24.79 and effect size generalized eta squared (ɳG2) = 0.08, driven by reversed weights in low-confidence info-seeking trials.

"the significant three way interaction between condition, selection and confidence level ( F(2, 52) = 24.79, p < 0.001, ɳG2 = 0.08). This effect occur red because the decision weights were reversed for low-confidence, info-seeking trials"

Lorenz Weise 2022 (2)¶

Lorenz Weise, Saskia D. Forster, Siegfried Gauggel. (2022). Reverse-correlation reveals internal error-corrections during information-seeking. Metacognition and Learning (2022) 17:321–335. https://doi.org/10.1007/s11409-021-09286-4

q3 · i2 · causal · r2

Post-hoc t-tests on the low-confidence info-seeking trials of the reverse correlation analysis. The non-selected option showed a positive association (M = 0.07, Cohen's d = 0.56) and the selected option a negative association (M = -0.16, Cohen's d = -0.87).

"for low-confidence, info-seeking trials, the non-selected stimulus option was indeed positively associated with choice (M = 0.07, t(26) = 2.89, p = 0.008, Cohen's d = 0.56) and t he selected option was negatively associated (M = -0.16, t(26) = -4.53, p < 0.001, Cohen's d = -0.87)."

Discussion¶

Related Claims¶

• Reverse-correlation decision weights are positive for the chosen stimulus dimension and weaker on low-confidence trials across confidence-assessment methods — related
• Internal error-corrections, not correct-to-incorrect changes-of-mind, produce the inverted low-confidence decision weights — related
• Primary task accuracy is higher in the info-seeking condition than in the explicit reports and wagering conditions — related
• The inverted low-confidence info-seeking pattern persists when only each participant's first condition is analyzed — related
• A single decision-and-confidence model with no condition-specific parameters fits decision weights across all three methods — related