Sleep Problems in Children Leave Detectable Brain Signatures That Track Mental Health

Childhood sleep problems are among the most common complaints that parents bring to pediatricians, and for decades clinicians have suspected that troubled sleep and troubled minds travel together. What has been far less clear is whether the brain itself carries a measurable imprint of those sleep difficulties, and whether that imprint can predict how a child’s mental health will unfold over the years that follow. A new study published in BMC Medicine suggests that it can. Drawing on the enormous Adolescent Brain Cognitive Development Study, or ABCD, an international team of researchers led by Yulin Wang and Debo Dong of Southwest University in Chongqing, working with colleagues at the Research Centre Jülich and Heinrich Heine University Düsseldorf in Germany, has identified distinct patterns of brain structure and function that are associated with different kinds of sleep problems in preadolescent children, and has shown that one of those patterns foreshadows the trajectory of emotional and behavioral difficulties over the following two years.

The research team capitalized on the scale and richness of the ABCD cohort, a landmark longitudinal study supported by the National Institutes of Health that recruited more than ten thousand children aged nine to ten and is following them for a decade into early adulthood. For this analysis, the researchers used data from children who had undergone both structural and functional magnetic resonance imaging at baseline, along with detailed parent-reported questionnaires about their children’s sleep. Sleep was measured with the Sleep Disturbance Scale for Children, a widely used instrument that captures a broad range of difficulties, from trouble falling asleep and night waking to snoring, sleepwalking, and excessive daytime sleepiness. Mental health was tracked with the Child Behavior Checklist across three assessment points: baseline, a one-year follow-up, and a two-year follow-up, allowing the team to model not just snapshots of symptoms but their developmental trajectories over time.

A central methodological choice set this study apart from much of the previous literature. Rather than treating sleep problems as a single lump-sum score, the researchers took a dimensional approach, using multivariate statistical techniques including partial least squares and principal component analysis to discover how different facets of sleep disturbance relate to different aspects of brain organization. This analysis revealed two significant dimensions of sleep-related problems. The first was a general sleep disturbance dimension, reflecting broad difficulties initiating and maintaining sleep. The second was a hypersomnolence-parasomnia dimension, capturing excessive sleepiness and unusual sleep behaviors such as sleepwalking and night terrors. The distinction matters because it implies that what clinicians often collapse into a single category of poor sleep may in fact represent neurobiologically separable phenomena.

When the team mapped each dimension onto the brain, they found that each was associated with partially distinct patterns of cortical morphology and resting-state functional connectivity, the synchronization of activity between brain regions measured while a person lies quietly in the scanner. Notably, the two dimensions showed differential alignment along the hierarchical organization of cortical neurodevelopment, the well-described gradient along which the cortex matures, with primary sensory areas developing earliest and higher-order association areas maturing latest. This suggests that different flavors of sleep disturbance may relate to different stages or levels of the brain’s maturational hierarchy, a finding that could eventually help explain why certain sleep complaints cluster with certain psychiatric vulnerabilities.

Despite their differences, the two dimensions also shared common ground. Both were linked to disruptions spanning the somatosensory network, which processes bodily sensations; the attention networks, including the dorsal and ventral attention systems that govern how the brain allocates focus; and the default mode network, the set of regions most active during inward reflection and mind-wandering. The involvement of these particular networks is intriguing because each has been repeatedly implicated in psychiatric conditions. The default mode network has been tied to rumination and depressive thinking, the attention networks to distractibility and impulsivity, and the somatosensory cortex to the bodily dimension of emotional experience. The overlap hints at a common neural substrate through which diverse sleep problems might exert their effects on mental health.

The most consequential finding, however, concerned prediction. The researchers fitted latent growth curve models to the mental health data, a statistical framework that estimates each child’s baseline level of symptoms and their rate of change across the three assessments. Only the multimodal neural pattern associated with the general sleep disturbance dimension was prospectively associated with mental health trajectories. Children whose brains carried this signature showed higher baseline levels of both internalizing symptoms, such as anxiety and depression, and externalizing symptoms, such as aggression and rule-breaking behavior. More strikingly, the same signature predicted slower model-estimated reductions in these symptoms across the two-year follow-up window, meaning that children with this neural profile not only started with more difficulties but appeared to recover from them more slowly.

It is worth being precise about what this does and does not mean. The hypersomnolence-parasomnia dimension, despite having its own distinct neural correlates, did not predict mental health trajectories in the same way. That asymmetry suggests that the general sleep disturbance signature is doing particular predictive work, and it sharpens the study’s central claim: a specific, measurable brain pattern linked to a specific, common form of childhood sleep disturbance captures information about future mental health that questionnaires alone may miss. The authors are careful to frame these neural signatures as correlates that may inform future studies of the mechanisms linking sleep and mental health, rather than as deterministic markers. The associations are statistical patterns at the group level, not diagnoses for any individual child.

Even with those caveats, the implications are considerable. Sleep is one of the few health behaviors that families can actually change, and a growing body of work suggests that treating childhood sleep problems can improve daytime functioning and mood. If the neural signature identified here reflects a pathway through which poor sleep shapes the developing brain, then early identification and intervention for sleep disturbance could become a concrete strategy for bending mental health trajectories before problems crystallize. The study’s dimensional framework also offers a template for future research: instead of asking whether children sleep badly, researchers can ask which dimension of sleep is disturbed, and look for correspondingly specific neural and clinical signatures.

The technical achievement underlying these conclusions should not be overlooked. Multimodal neuroimaging of thousands of children is a formidable undertaking, requiring careful harmonization of structural measurements of cortical thickness and morphology with functional connectivity data across many scanning sites, and rigorous statistical control to avoid false positives. By applying multivariate mapping with false discovery rate correction and validating the findings against longitudinal outcomes, the team has produced one of the most comprehensive pictures to date of how childhood sleep problems are written into the brain. As the ABCD cohort continues into adolescence, the next question is whether these early signatures deepen, fade, or transform, and whether interventions that improve sleep can rewrite them. For now, the message for parents and pediatricians is a familiar one with new scientific weight: a child’s sleepless nights may matter far more, and far longer, than they appear to on the surface.

Subject of Research: Neural correlates of childhood sleep problems and their association with preadolescent mental health trajectories

Article Title: Multimodal neuroimaging signature of sleep problems captures preadolescent mental health trajectories

Article References: Wang, Y., Tahmasian, M., Genon, S., Samea, F., He, Z., Liu, X., Lei, X., Eickhoff, S. B., & Dong, D. (2026). Multimodal neuroimaging signature of sleep problems captures preadolescent mental health trajectories. BMC Medicine. https://doi.org/10.1186/s12916-026-05273-1

Image Credits: AI Generated

DOI: 10.1186/s12916-026-05273-1

Keywords: sleep problems, preadolescence, neuroimaging, ABCD study, functional connectivity, default mode network, internalizing symptoms, externalizing symptoms, mental health trajectories, BMC Medicine, Multimodal, signature