No open-access location is known for this record yet.
No open-access location is known for this record yet.
In computer-based assessments, immediate performance feedback on incorrect responses may have major cognitive and metacognitive benefits but it may also be demotivating for students by mirroring task-related failure. This experiment investigated the impact of three different error clarification types. We systematically varied the presence and complexity of immediate error messages (1 × 4 between-subjects design) and analyzed the effects of feedback with different levels of error clarification complexity on students' cognition, metacognition, and motivational state. Our sample included 439 university students who worked on 12 computer-based constructed-response geometry tasks. The students received either no feedback or feedback with different error message complexities after incorrect responses (i.e., Knowledge of Results [KR], Knowledge of Correct Response [KCR], or Elaborated Feedback [EF]) paired with confirmatory KCR feedback after correct responses. All feedback types, namely KR, KCR, and EF, significantly enhanced error correction and equally fostered students' metacognitive calibration accuracy. Moreover, error correction increased with the time students' spent on the error-related feedback (even for simple KR messages), which suggests that students actively used the additional time to reflect on their mistakes. Nevertheless, KCR and EF increased error correction more effectively than KR. Students' motivational state after correct responses did not significantly differ after receiving feedback compared to no feedback. After incorrect responses, we found a negative effect on students’ motivational state for KR feedback. This detrimental effect was in contrast to EF, which mitigated the negative effect compared to KR feedback and showed a similar motivational state level as the KCR and no feedback group.
Loading…
Stored only when both papers are in the corpus — 50 citations exist upstream.
Papers that cite the same references — the relational fallback while embeddings are still backfilling.
Imported from OpenAlex (CC0) — nothing here is generated.