Mitigating cognitive biases in higher education contexts through a brief video-based intervention: A randomized controlled trial

Abstract

Cognitive biases present significant challenges in higher education, yet existing mitigation strategies are often resource-intensive or lack scalability. This study introduces and evaluates a novel, scalable intervention using brief, scenario-based videos to reduce cognitive biases that may influence self-perceptions, relationships, and decision-making in higher education contexts. We conducted a cross-national experiment (N=1038) with teachers and learners, targeting 21 distinct biases. Participants were randomly assigned to a control or experimental condition. Both groups viewed a video depicting a bias-eliciting scenario; the experimental group was then shown an additional "debiasing" segment containing supplemental, context-altering information. Results showed the debiasing videos significantly reduced biased decisions. This effect was substantially stronger for teachers (Cohen’s d = 0.51) than for learners (d = 0.11). For students, the intervention was effective only when participants viewed the videos with subtitles in their native language. Teachers, but not learners, also demonstrated a "transfer" effect, showing reduced bias in subsequent, unrelated tasks, after receiving the debiasing intervention. These findings indicate that brief, narrative-based micro-interventions are a promising and scalable tool for cognitive bias mitigation. However, their effectiveness is moderated by audience characteristics, like role as a teacher or a student, and the language of delivery, underscoring the need for context-sensitive design and implementation. • Brief video interventions reduce cognitive bias in educational settings • The debiasing effect was substantially stronger for teachers than for learners • Teachers, but not learners, showed reduced bias in unrelated subsequent tasks • Intervention is most effective when delivered in participants’ native language

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 0 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.