. Author manuscript; available in PMC: 2023 Nov 6.

Published in final edited form as: Dev Sci. 2022 Oct 27;26(3):e13330. doi: 10.1111/desc.13330

Table 2. Parameter estimates and confirmatory learning patterns.

Rank correlations (Spearman) between behavioural measures of confirmatory learning and two confirmation model parameter estimates: inverse temperature and confirmation bias. Confirmation bias is the difference between learning rates and is either raw: LR^Con – LR^Dis; or normalised: (LR^Con – LR^Dis) / (LR^Con + LR^Dis). ***p < .001, **p<0.01, *p<0.05, °p<0.1, FDR corrected (Benjamini-Hochberg method).

Behavioural pattern	Inverse temperature	Confirmation bias
Behavioural pattern	Inverse temperature	Raw	Normalised
Accuracy (Asymmetric and pre-reversal trials)	0.89***	0.06	0.05
Accuracy (Post-reversal trials)	-0.27*	0	-0.31*
Conservatism (Symmetric trials)	0.41***	0.3*	0.38**
Win-stay (All trials)	0.87***	0.21°	0.11
Lose-shift (All trials)	-0.79***	-0.26*	-0.34**