Year: 2026
Type: article
Abstract: The Alternate Uses Task (AUT) is the most widely used measure of divergent thinking, yet the robustness of significant AUT findings has not been systematically examined. This systematic review sought to quantify the statistical fragility of studies using the AUT by calculating the Continuous Fragility Index (CFI), a metric that estimates how many data points would need to change for a significant continuous outcome to become non-significant. We systematically searched Medline, Embase, PsycInfo, and ERIC from inception to July 21, 2025 for studies employing a two-group design and reporting a statistically significant AUT outcome. We summarized median CFI values across study-level subgroups, performed multivariable regression to examine predictors of CFI, and analyzed the effect of randomization on CFI. Sixty-nine studies met inclusion criteria were included in the quantitative analysis. The median CFI was 6.8 (IQR 3.4-14.6), with a median sample size of 67 (IQR 48-120). Our multivariable regression found that sample size was the only significant predictor of CFI ( p =0.001); journal impact factor, year of publication, citation count, and risk of bias were not associated with CFI. CFI values also did not differ significantly between randomized and non-randomized studies ( p =0.669). The fragility of AUT studies are comparable to other fields and could be made less fragile by ensuring adequate statistical power through larger sample sizes. Future AUT research should interpret statistical significance alongside robustness and the practical importance of between-group differences.
Source: Thinking Skills and Creativity
Language: English
Subfield: Experimental and Cognitive Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: hybrid
Estimated APC paid: $3,730
OpenAlex ID: W7224136111