A neuromarker of sustained attention was developed using whole-brain functional connectivity. Researchers used functional magnetic resonance imaging (fMRI) to identify brain networks whose strength during a sustained attention task predicted individual differences in performance. These networks generalized to previously unseen individuals, even predicting performance from resting-state connectivity alone. The model also predicted symptoms of attention deficit hyperactivity disorder (ADHD) in children and adolescents from resting-state connectivity. These results demonstrate that whole-brain functional network strength provides a broadly applicable neuromarker of sustained attention. The study used a 268-node functional brain atlas to define brain networks. Functional connectivity was measured using the blood oxygenation level-dependent (BOLD) signal, which reflects functional connectivity by revealing regions engaging in common or related processing. The strength of these networks was found to correlate with sustained attention performance, validated by its strong relationship with d', a measure of sensitivity in attention tasks. Network strength, calculated as the sum of edge strengths in the positive and negative tails, was shown to predict task performance and ADHD symptoms. Internal validation showed that network strength predicted task performance in novel individuals, with significant correlations between observed and predicted d' scores. External validation demonstrated that the model could predict ADHD symptoms in an independent dataset, indicating that attentional abilities are reflected in intrinsic connectivity. The model was also tested for generalizability across different data acquisition sites, age groups, and behavioral measures of attention. The study also examined the functional anatomy of attention networks, finding that high- and low-attention networks spanned numerous cortical, subcortical, and cerebellar nodes. These networks were found to involve regions such as the cerebellum, motor cortex, occipital lobes, and parietal regions. The results suggest that functional connectivity is a powerful predictor of attentional abilities, and that network models can serve as a holistic neural index of sustained attention. The study also compared functional connectivity with BOLD variance, finding that functional connectivity was a better predictor of attention than BOLD variance. The results suggest that functional brain networks can be used to predict cognitive abilities and clinical symptoms, and that these models are robust and generalizable across different datasets and populations. The study highlights the importance of data-driven analyses in understanding the neural mechanisms underlying sustained attention and ADHD.A neuromarker of sustained attention was developed using whole-brain functional connectivity. Researchers used functional magnetic resonance imaging (fMRI) to identify brain networks whose strength during a sustained attention task predicted individual differences in performance. These networks generalized to previously unseen individuals, even predicting performance from resting-state connectivity alone. The model also predicted symptoms of attention deficit hyperactivity disorder (ADHD) in children and adolescents from resting-state connectivity. These results demonstrate that whole-brain functional network strength provides a broadly applicable neuromarker of sustained attention. The study used a 268-node functional brain atlas to define brain networks. Functional connectivity was measured using the blood oxygenation level-dependent (BOLD) signal, which reflects functional connectivity by revealing regions engaging in common or related processing. The strength of these networks was found to correlate with sustained attention performance, validated by its strong relationship with d', a measure of sensitivity in attention tasks. Network strength, calculated as the sum of edge strengths in the positive and negative tails, was shown to predict task performance and ADHD symptoms. Internal validation showed that network strength predicted task performance in novel individuals, with significant correlations between observed and predicted d' scores. External validation demonstrated that the model could predict ADHD symptoms in an independent dataset, indicating that attentional abilities are reflected in intrinsic connectivity. The model was also tested for generalizability across different data acquisition sites, age groups, and behavioral measures of attention. The study also examined the functional anatomy of attention networks, finding that high- and low-attention networks spanned numerous cortical, subcortical, and cerebellar nodes. These networks were found to involve regions such as the cerebellum, motor cortex, occipital lobes, and parietal regions. The results suggest that functional connectivity is a powerful predictor of attentional abilities, and that network models can serve as a holistic neural index of sustained attention. The study also compared functional connectivity with BOLD variance, finding that functional connectivity was a better predictor of attention than BOLD variance. The results suggest that functional brain networks can be used to predict cognitive abilities and clinical symptoms, and that these models are robust and generalizable across different datasets and populations. The study highlights the importance of data-driven analyses in understanding the neural mechanisms underlying sustained attention and ADHD.