Related Experiment Videos
Comparing objective and subjective learning curves: judgments of learning exhibit increased underconfidence with
Asher Koriat1, Limor Sheffer, Hilit Ma'ayan
1Institute of Information Processing and Decision Making, University of Haifa, Israel. akoriat@research.haifa.ac.il
Journal of Experimental Psychology. General
|June 7, 2002
Summary
Judgments of learning (JOLs) show initial overconfidence, but a robust underconfidence-with-practice (UWP) effect emerges with repeated study cycles. This counterintuitive finding impacts memory research and understanding of learning confidence.
Area of Science:
- Cognitive Psychology
- Metacognition
- Learning Sciences
Background:
- Metacognition involves monitoring one's own cognitive processes.
- Judgments of learning (JOLs) are predictions of future memory performance.
- Previous research suggests JOLs are generally well-calibrated or slightly overconfident.
Purpose of the Study:
- To investigate the calibration of JOLs across repeated study-test cycles.
- To examine the robustness of the underconfidence-with-practice (UWP) effect.
- To explore theoretical explanations for the observed metacognitive phenomenon.
Main Methods:
- Participants studied paired associates over multiple cycles.
- Judgments of learning (JOLs) were collected after each study cycle.
- Experimental manipulations included feedback, presentation rate, incentives, and associative strength.
- JOLs were assessed both item-by-item and in aggregate.
Main Results:
- JOLs were well-calibrated with slight overconfidence on the first cycle.
- A significant shift to marked underconfidence occurred from the second cycle onwards (UWP effect).
- The UWP effect was robust across various experimental conditions and memory tasks.
Conclusions:
- The underconfidence-with-practice (UWP) effect is a reliable phenomenon in metacognition.
- Repeated exposure to learning material can lead to decreased confidence in memory prediction.
- Further theoretical work is needed to fully explain this counterintuitive metacognitive finding.