Dynamic sources of evidence supporting confidence judgments and error detection

Area of Science:

• Cognitive Neuroscience
• Decision Science

Background:

• Subjective confidence accompanies human decisions, reflecting perceived accuracy.
• Understanding the basis of confidence judgments is crucial for metacognitive research.

Purpose of the Study:

• Investigate the informational basis of confidence judgments.
• Examine the relationship between perceptual evidence, decisions, and metacognitive evaluations.
• Explore how post-decision perceptual information influences confidence.

Main Methods:

• Human participants performed a visual discrimination task (identifying stimuli with more dots).
• Stimuli remained visible for one second post-decision to assess confidence.
• Experimental conditions manipulated emphasis on speed versus accuracy.

Main Results:

• Confidence and error detection were influenced by stimulus evidence from both pre- and post-decision periods.
• This effect persisted irrespective of whether speed or accuracy was prioritized.
• Findings suggest a shared computational mechanism for confidence and error monitoring.

Conclusions:

• Confidence judgments integrate perceptual evidence beyond the immediate decision point.
• A shared computational basis for error detection and confidence evaluation is proposed.
• Results have implications for computational models of metacognition and decision-making.