Fig. 4

Modeling of the association between seasonal changes in neurobehavioral performance and changes in biomarker activities. Coefficients (β) and 95% confidence intervals (95% CI) were obtained from the multivariate quantile regression analysis performed on each standardized neurobehavioral variable as a function of the change in biomarker’s activities (AChE, BChE and APEH). Green circles represent EE group and golden circles OE group. All models were corrected by the sociodemographic variables (Fig. S3). As the modeled variables were ratios (i.e. they are unitless), significant coefficients (*) can be interpreted as the “percentage of change” in the response variable due to a 1% change in the referred predictor. Coefficient significance was determined by ANOVA analysis between alternative models with and without the respective biomarker activity as explanatory variable (see methods for details). The scoring of those neurobehavioral variables denoted with a psi symbol (ψ) must be considered inversely related to the individual performance, this is, higher scores denoted worse performances