Fig. 2

Metacognitive judgments of accuracy, computation of reminder bias using optimal indifference point (OIP) and actual indifference point (AIP), and relationship between reminder bias and metacognitive bias or self-judged accuracy (the confidence judgment) across participants. (A) Mean confidence judgments (self-reported accuracy estimates) and actual accuracy under forced-internal memory versus forced‐external reminder conditions. (B) Illustration of reminder bias computation: the probability of choosing reminders across target values (2–8) is plotted with the optimal indifference point (OIP) and the actual indifference point (AIP). Reminder bias (shorten as |bias| in the plot) is quantified as the difference between OIP and AIP. (C) Scatterplot of all participants’ AIP versus OIP. The diagonal line denotes perfect calibration (AIP = OIP); points below it indicate over‐reliance on reminders (AIP < OIP), points above indicate under‐use of reminders (AIP > OIP). The group-level deviation from this line reveals a significant reminder bias. (D) Correlation plot of the relationship between reminder bias and metacognitive bias, showing that participants who underestimate their memory accuracy (negative values on metacognitive bias) tend to rely more on reminders (positive values on reminder bias). (E) Correlation plot showing the relationship between reminder bias and the self-judged accuracy (the confidence judgment) in executing delayed intentions with internal memory.