posted on 2025-10-09, 09:42authored byGhassani Swaryandini, Jessica Graham, Shantell Griffith, Vasco Grilo, Federica Ruzzante, Xingruo Zhang, Siu Kit Yeung, Marta MangiaruloMarta Mangiarulo, Geetanjali Basarkod, Clarence Ng, Philip Parker, Jason Tangen, Alexander Saeri, Emily Grundy, Peter Slattery, Michael Noetel
Resistance to cognitive biases is a crucial element of rationality that influences judgement and decision-making. Here we synthesized the effects of debiasing training in educational settings. Our systematic review found 54 randomized controlled trials consisting of 383 effect sizes and 10,941 participants. Our meta-analysis of educational interventions showed a small, yet significant, improvement in reducing the likelihood of committing biases compared with control conditions (g = 0.26, 95% confidence interval 0.14 to 0.39), 160 effects from 41 studies, P < 0.001). Most studies focused on reducing the likelihood of committing biases (for example, confirmation bias) using cognitive strategies. Some biases seemed difficult to overcome (for example, representativeness heuristic), and questions remain about the depth and transferability of learning beyond classroom settings. All studies had unclear or high risk of bias and there was some risk of publication bias. While evidence suggests that educational interventions can reduce bias on targeted tasks, more research is needed to determine whether these improvements translate to meaningful changes in real-world decision-making and to identify which paedagogical approaches are most effective for reducing the influence of cognitive biases.<p></p>
History
Author affiliation
College of Life Sciences
Psychology & Vision Sciences
The data supporting the findings of this systematic review and meta-analysis are openly available via the OSF repository at https://osf.io/7x5z6. This repository contains the complete dataset extracted from the included studies, including effect sizes, moderator variables and risk of bias assessments. The minimum dataset necessary to interpret, verify and extend the research includes the coded effect sizes from each study, sample sizes and moderator variables used in our analyses. All data are provided in accessible formats.