Improvement of association between confidence and accuracy after integration of separate evidence over time

When making decisions in real-life, we may receive discrete evidence during a time period. Although participants can integrate information from separate cues to improve their accuracy, it is still debatable how confidence changes after receiving discrete information. Nevertheless, based on the strong positive relationship between accuracy and confidence, we predicted that similar to what is observed in accuracy, confidence would improve following the integration of separate pieces of information. We used a Random-dot-motion discrimination task in which one or two brief stimuli (i.e., pulse[s]) were presented, and participants had to indicate the predominant direction of dot motions by saccadic eye movement. Two pulses intervals (up to 1s) were randomly selected, where color-coded targets facilitated indicating confidence simultaneously. Using behavioral data, computational models, pupillometry, and EEG methodology, our data revealed that compared to single-pulse trials, in double-pulse trials, participants improve their confidence resolution rather than reporting higher confidence and that the EEG signal (centro-parietal-positivity) and pupil dynamics appear to follow confidence levels but only in double-pulse trials. Overall, our findings suggest that evidence integration helps to improve the associations between confidence and accuracy.