High-confidence errors are more likely than low-confidence errors to be corrected on a later test after feedback.¶

Claim · All claims
Evidence · 2 studies · 1 causal r?, 1 review r? · q2–q3
Currency · newest study 2017 · 1 of 2 entries checked against their sources (automated) · not reviewed by a person

The "hypercorrection effect" describes the phenomenon where errors committed with high confidence are more likely to be corrected and remembered accurately later than low-confidence errors, likely due to increased attention triggered by the surprise of being wrong.

Subclaims¶

q2 i? Correcting high-confidence errors leads to superior long-term retention of the correct information compared to correcting low-confidence errors. → Metcalfe 2017 q3 i? Feedback that contradicts a high-confidence answer is better remembered, in both its appearance and its content, a pattern the authors read as consistent with surprise drawing attention to the feedback at encoding. → Fazio & Marsh 2009

Evidence¶

Primary evidence link: https://doi.org/10.1146/annurev-psych-010416-044022

Metcalfe 2017¶

Metcalfe, J. (2017). Learning from errors. Annual Review of Psychology, 68, 465–489. https://doi.org/10.1146/annurev-psych-010416-044022

q2 · narrative review · i? · the abstract prints no effect size; the full text may · n=N/A · review · r?

This comprehensive review synthesizes decades of research on learning from errors. It identifies the hypercorrection effect as a robust finding across multiple domains and age groups, highlighting the importance of generating answers (even if wrong) and receiving immediate corrective feedback.

Fazio & Marsh 2009¶

Fazio, L. K., & Marsh, E. J. (2009). Surprising feedback improves later memory. Psychonomic Bulletin & Review, 16(1), 88–92. https://doi.org/10.3758/PBR.16.1.88

q3 · peer-reviewed experiment · i? · no source text available to check; the entry prints no effect size · n=118 · causal · r?

Two experiments with undergraduates (46 in Experiment 1, 72 in Experiment 2) found better memory for the appearance of feedback (its font colour or voice) when it did not match their expectation (feedback on a high-confidence error or on a low-confidence correct answer), and Experiment 2 also showed the hypercorrection effect itself. The authors read this as consistent with the surprise hypothesis, using memory for the feedback's appearance as a measure of attention; attention was not manipulated directly.

Studies bearing on this claim¶

Which errors to correct first: are errors made with high confidence corrected more durably than low-confidence ones? This table gathers the studies behind this claim and the 0 narrower claims that bear on it: 2 distinct studies (1 review, 1 causal), q2–q3, 0 of 2 with an effect size coded. One row per study, codes as each entry records them; no effect is averaged and no verdict is drawn. Built by scripts/synthesize_claim.py from eval/synthesis/; a row that looks wrong is fixed on the claim it comes from.

Discussion¶

The effect is dependent on receiving immediate, non-punitive, and clear corrective feedback; without feedback, errors can be reinforced as misconceptions.

Related Claims¶

• Pretesting enhances learning — the general claim this finding bears on, with the studies behind it
• Erroneous examples improve conceptual understanding by forcing comparison with correct models.
• Prompting learners to self-explain improves understanding and problem solving on immediate tests by a moderate average amount, with little evidence yet on delayed or classroom gains.
• Emotion dynamics during problem-solving predict learning outcomes in a manner that depends on scaffolding design — related
• Feedback Enhances Retrieval Practice — related
• Retrieval practice improves long-term retention — a broader claim this one bears on
• Corrective feedback facilitates L2 grammatical development, with explicit feedback stronger immediately and implicit feedback better maintained over time — related