2.8. Relations between brain age gap and cognitive profiles

Finally, we explored the possibility of using the estimated brain age of each individual as a predictor of their cognitive profiles, including the in‐scanner task performance and arithmetic skills (Figure 1e). This investigation is conducted using linear regression analysis. To account for the potential confounding effects of actual age, which is inherently correlated with the estimated brain age and task performance, we incorporated participants' chronological age as a nuisance regressor in our analysis. This adjustment is crucial as models estimating brain age can also lead to overestimation for younger individuals and underestimation of older individuals (Le et al., 2018). We aimed to discern the specific contribution of the estimated brain age in predicting cognitive outcomes, offering a more accurate assessment of its significance in relation to participants' cognitive abilities.