Related Experiment Videos
Exploring a mnemonic debiasing account of the underconfidence-with-practice effect
Asher Koriat1, Hilit Ma'ayan, Limor Sheffer
1Department of Psychology, University of Haifa, Haifa, Israel. akoriat@research.haifa.ac.il
Summary
Judgments of learning (JOLs) often underestimate learning gains from repeated study, known as the underconfidence-with-practice (UWP) effect. This bias is linked to foresight bias, but delayed JOLs can mitigate this effect.
Area of Science:
- Cognitive Psychology
- Educational Psychology
- Metacognition
Background:
- Judgments of learning (JOLs) are metacognitive assessments of future recall.
- The underconfidence-with-practice (UWP) effect describes JOLs underestimating actual recall improvement with repeated study.
- Existing theories suggest foresight bias contributes to the UWP effect.
Purpose of the Study:
- To investigate the role of foresight bias in the UWP effect.
- To examine how the timing and cueing of JOLs influence the UWP effect and practice's debiasing effects.
- To test whether delaying JOLs can moderate the UWP effect.
Main Methods:
- Three experiments were conducted involving repeated study of cue-target pairs.
- Participants made judgments of learning (JOLs) at different time points and with different cueing methods (cue alone vs. cue-target pair).
- Recall performance was measured after repeated study and JOLs were compared to actual recall.
Main Results:
- The UWP effect was most pronounced for items that induced a foresight bias.
- Delaying JOLs reduced the effectiveness of practice in debiasing judgments, thus moderating the UWP effect.
- This moderation occurred specifically when delayed JOLs were prompted by the cue alone, mimicking testing conditions.
Conclusions:
- Foresight bias plays a significant role in the underconfidence-with-practice effect.
- The timing and cueing of JOLs are critical factors influencing metacognitive accuracy.
- Prompting delayed JOLs with cues alone, similar to retrieval practice, can help moderate the UWP effect by reducing the impact of foresight bias.