Extended Data Fig. 2: Association between age, home environment, country and path integration performance.

a, Path integration performance as a function of age for male and female participants who grew up in city and non-city environments. Path integration performance is averaged within 5-year windows, centre values correspond to the means. b, Difference of the effect of growing up outside cities on path integration performance across countries. We fit a logistic mixed model for path integration performance, with fixed effects for age, gender and education, and random environment slopes clustered by country, see Supplementary Methods. Positive values indicate an advantage for participants raised outside cities. c, SNE as a function of the environment effect size (random environment slope) in each country, as in Fig. 2b, see Supplementary Methods. All error bars correspond to standard errors, n = 182,122 participants.