Correction to: Nature Human Behaviour https://doi.org/10.1038/s41562-022-01474-9. Published online 21 November 2022.

In the version of this article initially published, some survey data in Study 4 were accidentally duplicated, and the duplicates have now been removed and the analysis revised. The impact is that numbers indicating acrophily have slightly changed from the old model (while keeping the significance and direction), b = 0.12, t(427) = 3.62, P < 0.001, R² = 0.02, 95% confidence interval [0.05, 0.18], to the revised model, b = 0.10, t(1,542) = 3.11, P = 0.001, R² = 0.02, 95% confidence interval [0.04, 0.17]. Differences between political groups were slightly different as well, but were still non-significant. In the original model, b = 0.01, t(415) = 0.19, P = 0.84, R² = 0.02, 95% confidence interval [−0.08, 0.10], while in the new model, b = 0.03, t(1,542) = 0.74, P = 0.45, R² = 0.02, 95% confidence interval [−0.06, 0.14]. Patterns of statistical (non)significance have remained unaffected in all relevant analyses. Text in the Results, Study 4 section has been revised to the new model in the HTML and PDF versions of the article.