Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Oct 9.
Published in final edited form as: Biol Psychiatry Cogn Neurosci Neuroimaging. 2025 Sep 24;11(8):963–975. doi: 10.1016/j.bpsc.2025.09.015

Longitudinal Associations of Structural and Functional Brain Connectivity With Dimensions of Psychopathology in Adolescence

Lucy Vanes 1,2, Divyangana Rakesh 1,2, Tobias Banaschewski 1,2, Arun LW Bokde 1,2, Sylvane Desrivières 1,2, Herta Flor 1,2, Hugh Garavan 1,2, Penny Gowland 1,2, Antoine Grigis 1,2, Andreas Heinz 1,2, Herve Lemaitre 1,2, Jean-Luc Martinot 1,2, Marie-Laure Paillère Martinot 1,2, Eric Artiges 1,2, Frauke Nees 1,2, Dimitri Papadopoulos Orfanos 1,2, Luise Poustka 1,2, Michael N Smolka 1,2, Sarah Hohmann 1,2, Nathalie Holz 1,2, Nilakshi Vaidya 1,2, Henrik Walter 1,2, Robert Whelan 1,2, Gunter Schumann 1,2, Gareth J Barker 1,2, the IMAGEN Consortium
PMCID: PMC13646494  NIHMSID: NIHMS2202782  PMID: 41005664

Abstract

BACKGROUND:

Adolescence is a critical period of neurodevelopment marked by ongoing maturation of structural and functional brain connectivity. Simultaneously, this period is associated with an increase in mental health problems, spanning from subclinical symptoms to diagnosable disorders.

METHODS:

This study investigated longitudinal associations between psychopathology dimensions and voxelwise brain measures related to connectivity across 3 time points (ages 14, 19, and 23) in more than 1500 participants using the IMAGEN dataset. White matter (WM) microstructure was indexed using diffusion metrics quantified along the WM skeleton (N = 1736), while functional connectivity was captured as voxelwise degree centrality (DC) derived from resting-state functional imaging (N = 1510).

RESULTS:

Development of WM microstructure was selectively linked to externalizing (but not internalizing) symptoms. Here, higher externalizing symptoms were associated with widespread reductions in fractional anisotropy (FA) across the WM skeleton as well as accelerated decreases in FA in the corticospinal tract over time. In contrast, functional DC was developmentally associated with general, rather than specific, psychopathology in frontal and temporal regions. An increase in total difficulties over time was associated with developmental decrease in DC in bilateral superior frontal gyri. In addition, a positive association between total difficulties and DC in left inferior temporal gyrus was observed in younger, but not older, adolescents or young adults.

CONCLUSIONS:

These findings highlight the dynamic interplay between brain connectivity development and psychopathology in adolescence, with potential implications for identifying neural markers of risk and resilience during sensitive windows of development.

Connectivity underpins the functional capacities of the brain, supported by both the structural integrity of white matter (WM) tracts and the fine-tuning of functional networks (1). While structural and functional brain connectivity undergo continued development throughout the lifespan (2,3), adolescence represents a particularly sensitive period of neural reorganization (4) driven by increasing environmental demands coupled with biological processes of maturation. As a result, the manner in which neurodevelopment unfolds during this stage can have profound effects on individual cognitive, emotional, and behavioral trajectories (5).

Adolescence is also marked by increased mental health problems, ranging from diagnosable disorders (6) to subclinical difficulties that signal vulnerability (7,8). Internalizing (broadly emotional) and externalizing (broadly behavioral) problems typically rise throughout adolescence, albeit with marked heterogeneity (9–11). This underscores the need to consider their dynamic nature in studies of neurobiological etiology. Furthermore, given the potential reciprocal influences between internalizing and externalizing dimensions (12,13), it is crucial to investigate the effects of one in the context of the other.

Accumulating evidence supports an association between alterations in brain connectivity and adolescent psychopathology (14–16). An emerging picture suggests that delayed or disrupted maturation of large-scale networks (including default mode, executive, and sensorimotor) may contribute (17), whereby disrupted structural connectivity might constrain the development of stable and flexibly interconnected functional networks (18). This highlights the importance of considering both structural and functional aspects of brain development when studying the neurodevelopmental correlates of psychopathology. Diffusion imaging sheds light on microstructural WM parameters that support interregional communication, while resting state functional imaging captures evolving patterns of functional connectivity. Studying both in the same individuals, especially longitudinally, enables a more complete understanding of how mechanisms supporting brain connectivity evolve in relation to emerging mental health symptoms.

Notably, most prior neuroimaging research has relied on cross-sectional designs, potentially overlooking dynamic associations evolving over time. Where longitudinal associations have been investigated [e.g. (19,20)], work frequently focuses on predefined regions, tracts, or networks. This approach poses challenges for capturing longitudinal associations, as structural and functional network configurations themselves may change over time (21,22). This highlights a need for fine-grained (i.e., voxelwise) analyses capturing coupled evolution of brain connectivity and mental health symptoms, embracing the interindividual variability inherent in both neural and behavioral development.

In this study, we investigated longitudinal associations of psychopathology dimensions in adolescence and early adulthood with structural and functional brain connectivity. We leveraged data from the large longitudinal IMAGEN dataset (23) to assess WM microstructure using diffusion imaging and functional connectivity using resting-state functional magnetic resonance imaging (MRI) in 1736 young people ages 14 to 23 years. IMAGEN work has previously illuminated adolescent processes such as reinforcement learning (24), emotion regulation (25), substance use (26), and specific clinical traits (27), but no study has jointly examined the longitudinal development of WM and functional connectivity at the voxelwise level or linked these trajectories to broad psychopathology dimensions.

Alongside voxelwise diffusivity measures, we derived voxelwise degree centrality (DC), a whole-brain index of functional coupling of each voxel with the rest of the brain that is well suited to longitudinal analyses in which network organization may shift over time. Combined, these metrics provide holistic insight into connectivity development, unconstrained by predefined network hypotheses. After examining longitudinal associations between overall psychopathology and brain measures, we investigated the respective contribution of internalizing and externalizing dimensions—each in the context of the other—to better understand which aspects drive observed associations. Our modeling statistically disentangled within-subject developmental effects from between-subject age differences, reducing the risk of conflating individual developmental trajectories with stable differences between participants.

METHODS AND MATERIALS

Participants and Procedure

Participants were drawn from the European imaging genetics project IMAGEN, a multicenter longitudinal study that recruited approximately 2000 typically developing adolescents across 4 assessment waves, of which 3 (baseline: age 14; follow-up 2: age 19; follow-up 3: age 23) included MRI. The current study focused on these 3 assessments. For a detailed description of recruitment and assessment, see Schumann et al. (23).

After quality control, the present study included data from 1736 adolescents with diffusion data from at least 1 time point (3278 scans) and 1510 adolescents with resting-state data from at least 1 time point (2434 scans) who also had complete psychopathology data at the time of each scan (see Supplement and Table 1 for more details on scan availability by time point and modality). The study was previously approved by the local ethics committees, and participants provided written informed consent or assent (for participants <18 years of age, their legal guardians provided written informed consent). Exclusion criteria included current bipolar disorder; schizophrenia; and major neurodevelopmental conditions such as autism, preterm birth, head trauma, or history of neurological or medical disorders.

Table 1.

Sample Characteristics by Assessment Time Point

Variable Baseline FU2 FU3
No. of Scans
 DTI 1411 943 924
 fMRI 331 988 1115
Age, Years
 DTI sample
  Mean (SD) 14.44 (0.40) 19.01 (0.71) 22.54 (0.65)
  Median 14.42 18.82 22.40
  Range 13.23–15.72 17.82–21.53 20.73–25.11
 fMRI sample
  Mean (SD) 14.48 (0.40) 19.01 (0.72) 22.55 (0.65)
  Median 14.47 18.82 22.42
  Range 13.52–15.48 17.82–21.53 20.89–25.13
Sex, Female, %
 DTI sample 52.7% 53.4% 53.7%
 fMRI sample 46.5% 53.0% 53.3%
SES, Mean (SD)
 DTI sample 0.68 (1.07)
 fMRI sample 0.60 (0.94)
SDQ Score, Mean (SD)
 DTI sample
  SDQ-TOT 17.98 (4.58) 18.24 (4.94) 17.73 (4.73)
  SDQ-INT 4.37 (2.89) 4.67 (3.14) 4.45 (3.09)
  SDQ-EXT 5.89 (3.09) 5.08 (2.81) 4.54 (2.62)
 fMRI sample
  SDQ-TOT 18.16 (4.73) 18.35 (4.97) 17.54 (4.68)
  SDQ-INT 4.57 (3.05) 4.78 (3.19) 4.40 (3.05)
  SDQ-EXT 5.99 (3.09) 5.07 (2.80) 4.56 (2.61)
Open in a new tab

DTI, diffusion tensor imaging; fMRI, functional magnetic resonance imaging; FU2, second follow-up of IMAGEN study; FU3, third follow-up of IMAGEN study; SDQ, Strengths and Difficulties Questionnaire; SDQ-EXT, SDQ externalizing; SDQ-INT, SDQ internalizing; SDQ-TOT, SDQ total; SES, socioeconomic status.

Psychopathology

Psychopathology was assessed by self-report on the Strengths and Difficulties Questionnaire (SDQ), a 25-item scale assessing behavioral and emotional difficulties (28), administered at each study time point. Total difficulties scores, internalizing difficulties scores, and externalizing difficulties scores were derived for all participants at each time point (ages 14, 19, and 23); see Supplement for details.

Imaging Acquisition and Preprocessing

MRI data were acquired on 3T scanners across IMAGEN sites (Siemens; Philips; General Electric) at ages 14, 19, and 23. Detailed acquisition parameters for structural T1-weighted, diffusion-weighted, and resting-state images are presented in the Supplement. Effects of site on mean fractional anisotropy (FA) and mean DC for each assessment time point are summarized in Table S7. In all analyses, site was included as a covariate of no interest to adjust for potential intersite differences. Preprocessing steps for the diffusion and resting-state data followed standard procedures of the IMAGEN Consortium; details are in the Supplement.

Calculation of Functional Voxelwise DC

To derive voxelwise relative DC maps for each participant at each time point, the time series from each voxel within a gray matter mask was extracted and a correlation matrix constructed from pairwise Pearson correlations between each pair of voxels. Correlations below 0.2 were excluded to avoid including spurious connections. Weighted DC was calculated for each voxel as the sum of correlations between this voxel and all other voxels within the mask, reflecting, in absolute terms, the centrality of each voxel within the gray matter mask. To enhance comparability across participants and time points, relative DC maps were generated by standardizing DC values to within-subject and within–time point z scores. Maps thus reflect the centrality of each voxel relative to all other gray matter voxels at the time of the scan, while reducing scanner-related global differences. Longitudinal analyses of DC therefore reflect changes in the relative centrality of each voxel over time. DC was chosen as our primary resting-state metric because it provides a whole-brain, voxelwise measure of global functional connectivity without constraining analyses to predefined network boundaries. This makes it particularly well suited for longitudinal analyses, in which functional networks may reorganize over time.

Statistical Analysis

Skeletonized diffusion metric maps (FA, trace [TR], radial diffusivity [RD], axial diffusivity [AD]) and functional connectivity maps (DC) underwent separate but identical voxelwise longitudinal analyses using nonparametric mixed-effects marginal modeling with the FSL Sandwich Estimator (SwE) tool (29) (see Supplement for more details), testing the longitudinal development of each brain metric and its association with psychopathology. Statistical inferences for effects of interest were achieved using a nonparametric wild bootstrap procedure with 999 iterations (30). Threshold-free cluster enhancement (31) was used for inference on diffusion maps to detect focal effects along the WM skeleton while controlling for familywise error rate (FWER) at p < .05 (2-tailed). Cluster-extent correction was used for inference on functional maps to identify spatially contiguous regions of significant effect, reflecting the typically broader, network-level changes observed in functional connectivity analyses, while also controlling for FWER. For these analyses, a cluster-forming parametric uncorrected p-value threshold of .001 was applied, with clusters retained at FWER-corrected p < .05.

For each modality, a first model including SDQ total (SDQ-TOT) scores was run, followed by a second model including both SDQ internalizing (SDQ-INT) and SDQ externalizing (SDQ-EXT) scores to clarify the relative contributions of these 2 symptom domains to any observed associations (described in more detail below). These 2 models were considered part of a planned, hierarchical analysis within each modality, with the second model (SDQ-INT and SDQ-EXT) serving as a decomposition of the first (SDQ-TOT). As they were not treated as independent tests, no additional correction was applied across them. Similarly, we did not correct across modalities, as these analyses were considered separate families of tests.

Longitudinal and Cross-Sectional Variable Decomposition.

The time-varying variables age at scan and SDQ-TOT, SDQ-INT, and SDQ-EXT scores were each decomposed into their cross-sectional and longitudinal components to distinguish between-subject and within-subject effects (32). This allows for a true separation of developmental effects occurring between individuals (cross-sectional) from those occurring within individuals over time (longitudinal), reducing the risk of confounding between the 2 sources of variation.

Cross-sectional age was each participant’s mean age across( time points centered) with respect to the whole sample (Agesubject¯−Agesample¯). A main effect of cross-sectional age reflects between-subject brain differences between (on average) younger and older participants. A histogram of mean subject age in Figure S1 shows that there was indeed meaningful variation in the average subject age, likely due to the fact that not all participants completed all scan time points.

Longitudinal age was the within-subject centered age at each time point (Agesubject,time point−Agesubject¯), henceforth referred to as time, as it represents the within-subject change over time while taking into account the temporal distance between scans. A main effect of time reflects longitudinal brain changes that occur within a subject.

Similarly, cross-sectional SDQ scores (SDQsubject¯−SDQsample¯) represent each subject’s average SDQ score across time points, while longitudinal SDQ (SDQsubject,time point−SDQsubject¯) represents the within-subject change in SDQ score over time. A main effect of cross-sectional SDQ reflects average brain difference between individuals experiencing different levels of psychopathology, while a main effect of longitudinal SDQ reflects within-subject coupled changes over time in psychopathology and brain metrics. SDQ-TOT, SDQ-INT, and SDQ-EXT scores were each decomposed in this manner.

SDQ-TOT Analysis.

The first model run for each brain modality (as a voxelwise time-varying outcome) included the following regressors: scan site, sex, cross-sectional age, time, cross-sectional SDQ-TOT, longitudinal SDQ-TOT, cross-sectional age × time (testing nonlinear developmental effects), cross-sectional age × cross-sectional SDQ-TOT (testing the interaction between average age and average SDQ-TOT score), and time × cross-sectional SDQ-TOT (testing whether longitudinal developmental effects differ as a function of average SDQ total score). For DC analyses, each participant’s mean FD was included as a further covariate.

SDQ-INT and SDQ-EXT Analysis.

A second model was run for each brain modality including the following regressors: scan site, sex, cross-sectional age, time, cross-sectional SDQ-INT, cross-sectional SDQ-EXT, longitudinal SDQ-INT, longitudinal SDQ-EXT, cross-sectional age × time, cross-sectional age × cross-sectional SDQ-INT, cross-sectional age × cross-sectional SDQ-EXT, time × cross-sectional SDQ-INT, and time × cross-sectional SDQ-EXT. Mean FD was included as an additional covariate in DC analyses only.

Sensitivity Analyses.

Additional sensitivity analyses were run to test the robustness of the observed effects. All reported effects remained significant across all sensitivity analyses; details are in the Supplement.

RESULTS

Sample characteristics are in Table 1.

Structural Connectivity Findings (FA, TR, RD, AD)

Developmental Effects.

Developmental effects were extracted from the main effects of cross-sectional age and time, as well as their interaction, within the first (SDQ-TOT) model run for each modality. Widespread developmental increases in FA were evident across the WM skeleton both cross-sectionally (positive effect of cross-sectional age), reflecting higher FA in older compared with younger participants (Figure 1A), and longitudinally (positive effect of time), reflecting a within-subject increase in FA over time (Figure 1B). In contrast, there were focal decreases in FA primarily in bilateral corticospinal tracts (CSTs), which were again reflected both cross-sectionally and longitudinally. Full cluster details are listed in Table S2. These developmental effects were largely mirrored by cross-sectional and longitudinal effects on TR, RD, and AD (Figure S2). There was no significant cross-sectional age × time interaction on any of the diffusion metrics, indicating largely linear developmental effects such that change over time did not differ as a function of age within the sample.

Figure 1.

Figure 1.

Open in a new tab

Positive (red) and negative (blue) developmental effects on fractional anisotropy (FA), captured cross-sectionally (A) and longitudinally (B). Scatterplots show mean FA across all voxels within the significant clusters displayed in the corresponding brain maps.

Psychopathology Effects.

FA analyses revealed a significant negative main effect of cross-sectional SDQ-TOT across large parts of the WM skeleton including corpus callosum, bilateral anterior thalamic radiation, bilateral CSTs, and bilateral inferior fronto-occipital fasciculi (Figure 2A; Table S3), indicating that a higher average SDQ-TOT score across time points was associated with widespread lower average FA across time points (Table S3). This effect was reflected in more focal effects for AD and, to a lesser extent, TR (Figure 2A). There was no effect of cross-sectional SDQ-TOT on RD.

Figure 2.

Figure 2.

Open in a new tab

Effect of average Strengths and Difficulties Questionnaire total (SDQ-TOT) score (A) and SDQ externalizing (SDQ-EXT) score (B) on fractional anisotropy (FA), trace (TR), and axial diffusivity (AD). Scatterplots show mean FA across all voxels within the significant clusters displayed in the corresponding brain maps.

Models including both SDQ-INT and SDQ-EXT scores revealed that the observed SDQ-TOT effects were mirrored in a significant negative main effect of cross-sectional SDQ-EXT on widespread FA, as well as TR and AD (Figure 2B; Table S4). In contrast, there was no main effect of cross-sectional SDQ-INT on any of the diffusion metrics. In addition, there was a significant negative time × cross-sectional SDQ-EXT interaction on FA in left CST, indicating that participants with higher externalizing symptoms demonstrated greater decreases in FA over time in this tract (Figure 3; Table S5). This effect was not reflected in TR, RD, or AD.

Figure 3.

Figure 3.

Open in a new tab

Change in fractional anisotropy (FA) over time as a function of average Strengths and Difficulties Questionnaire externalizing (SDQ-EXT) score. Scatterplot depicts mean FA across all voxels within the significant clusters displayed in the brain map. For visualization only, SDQ-EXT score is divided into quartiles (EXT-Q1 to EXT-Q4). Time reflects within-subject centered age, representing the deviation from each participant’s mean age across time points. FA in corticospinal tract decreases most rapidly in participants with highest average SDQ-EXT scores.

Degree Centrality

Developmental Effects.

There were no significant main effects or interaction of cross-sectional age or time on DC, indicating that there was no systematic (cross-sectional or longitudinal) developmental effect in terms of relative voxelwise connectivity.

Psychopathology Effects.

There was a significant negative effect of longitudinal SDQ-TOT, indicating that an increase in SDQ-TOT score over time was associated with a reduction of DC primarily in bilateral posterior superior frontal gyri (pSFG), extending to precentral gyri, over time (Figure 4; Table S6). Finally, there was a significant cross-sectional age × cross-sectional SDQ-TOT interaction in the left inferior temporal gyrus (ITG), indicating that, in younger participants, higher average SDQ-TOT scores were associated with increased average ITG connectivity. This association was not present in older participants (Figure 5; Table S7).

Figure 4.

Figure 4.

Open in a new tab

Effect of longitudinal Strengths and Difficulties Questionnaire total (SDQ-TOT) score on degree centrality (DC) in bilateral superior frontal gyri (SFG). Scatterplots depict mean DC within each significant cluster. Within-subject increase in total score is associated with within-subject DC decrease in SFG.

Figure 5.

Figure 5.

Open in a new tab

Age by Strengths and Difficulties Questionnaire total (SDQ-TOT) score interaction on degree centrality (DC) in left inferior temporal gyrus (ITG). Scatterplot depicts mean DC within the significant cluster. For visualization only, participants are grouped into age bands (13–17, 17–21, 21–25) based on cross-sectional age, reflecting each participant’s mean age across available time points. Age bands do not correspond directly to assessment waves. Higher total difficulties are associated with ITG hyperconnectivity in younger participants.

DISCUSSION

This study examined longitudinal associations between structural and functional brain development and psychopathology in adolescence and young adulthood using a large, population-based sample. We observed that WM microstructural development was selectively associated with externalizing symptoms, with higher symptom levels linked to both widespread lower FA and steeper declines of FA over time in the left CST. In contrast, functional brain connectivity, indexed by voxelwise DC, showed effects linked to overall psychopathology, without strong differentiation between internalizing and externalizing symptoms. Longitudinal increases in total difficulties were associated with lower superior frontal connectivity, and younger adolescents showed a selective association between total difficulties and inferior temporal connectivity. These results highlight distinct but complementary patterns across structural and functional domains, suggesting that specific features of brain development are differentially sensitive to distinct aspects of emerging psychopathology during adolescence.

Structural neuroimaging research on developmental mental health has previously highlighted broad WM involvement. Lower global WM microstructure has been associated with higher general psychopathology in children (33), while widespread alterations across multiple WM tracts have been linked to psychopathology in adolescents (16,34,35). Our findings in the IMAGEN cohort reflect this, showing widespread cross-sectional negative associations between FA and total difficulties. Corresponding patterns in TR and AD—though not RA—point to variation in axonal properties rather than differences in myelination as driving this effect (36,37). Notably, our follow-up analyses suggest that this brain-behavior relationship is primarily accounted for by externalizing rather than internalizing symptoms, aligning with previous work linking WM microstructure to disruptive behaviors in youth (38,39).

In addition to these widespread effects, we identified a more focal longitudinal association in the left CST, in which higher externalizing symptoms were linked to greater reductions in FA over time. This tract, traditionally associated with motor control (40), has increasingly been implicated in behavioral regulation (41) as well as higher-order processes (42). Similar longitudinal patterns have been reported in adolescents with attention-deficit/hyperactivity disorder (ADHD), in which atypical CST development was linked to persistent behavioral problems (43). Conversely, clinical improvement of hyperactivity-impulsivity symptoms in ADHD was associated with increased CST fiber density in young adulthood (44). Our work extends this by showing that even in a community sample, variation in CST maturation tracks externalizing symptoms, potentially reflecting delayed or disrupted integration of motor-executive pathways involved in impulse control.

Focusing on brain function, we found a longitudinal coupling between superior frontal connectivity and general psychopathology, in which reduced centrality of bilateral pSFG over time was associated with a symptom exacerbation from ages 14 to 22. DC reflects the extent to which a brain region is functionally connected to the rest of the brain, serving as a marker of its integrative role in large-scale brain communication. The SFG is a key hub within the frontoparietal and default mode networks, and its connectivity is thought to support cognitive control (45), emotion regulation (46), and broader executive functioning (47). pSFG more specifically has been implicated in sociocognitive processes such as mentalizing (48,49), and diminished default mode connectivity within this region has been linked to greater social dysfunction in individuals with major depressive disorder (50). Given the ongoing maturation of prefrontal circuits during adolescence, declining pSFG centrality may reflect an altered neurodevelopmental trajectory that compromises regulatory or socioemotional capacity, contributing to increased nonspecific psychiatric vulnerability. Previous research supports this, linking superior frontal dysconnectivity to affective symptoms (51,52), cognitive impairments (53), and psychotic-like experiences (54). These findings suggest that developmental changes in pSFG connectivity might serve as a sensitive marker for emerging mental health difficulties.

We further observed an age-specific effect linking general psychopathology to inferior temporal connectivity: Younger, but not older, adolescents showed a positive relationship between total difficulties and left ITG centrality. Left ITG has been linked to cognitive functions such as language (55), visual perception (56), and memory (57) and, more recently, emotion regulation (58,59)—all of which can be disrupted in youth with elevated psychopathology (60). Neuroimaging studies report altered ITG function and structure in relation to internalizing (61–63) or externalizing (64,65) symptoms, or both (66), though inconsistencies remain with respect to the timing and direction of effects. Our findings may help explain this variability, suggesting a compensatory mechanism at younger ages, whereby increased ITG connectivity supports emotional or sensory processing in the face of emerging difficulties. This interpretation aligns with evidence that enhanced left ITG connectivity may play a compensatory role that buffers against the impact of childhood trauma on risk for developing depression later in life (67). The reduction of the association between ITG connectivity and psychopathology in older adolescents could reflect a diminished reliance on, or capacity for, this compensatory recruitment as cognitive and neural systems mature. However, it is worth noting that this interaction was not reflected longitudinally (i.e., the ITG-psychopathology association did not decrease within subjects over time), suggesting the effect may reflect differences between age groups rather than true developmental change. Future work should test whether the observed effect is mediated by social or environmental factors that vary across age groups or reflects a time-limited neural strategy that supports regulation in the face of early symptoms.

Of note, sensitivity analyses showed that all effects remained significant after controlling for socioeconomic status and including interactions with sex. While this strengthens robustness, future work should directly test sex-specific and socioenvironmental influences given their importance in adolescent psychopathology. Our findings suggest that structural and functional brain changes show distinct but complementary associations with adolescent mental health, with WM more closely tied to externalizing symptoms and functional connectivity more closely tied to general psychopathology. This aligns with emerging models that emphasize both domain-specific and transdiagnostic neurobiological processes (68–70). However, it is also important to consider that structural and functional connectivity may show bidirectional influences over time, whereby WM architecture constrains functional brain network maturation (71), and functional activity shapes structural refinement through experience-dependent plasticity (72). Increased structure-function coupling has been linked to increased negative affectivity (73), suggesting that greater structural constraints on function may result in adverse psychological outcomes. At the same time, while bidirectional influences may exist at the biological level, our findings indicate that associations with psychopathology can be modality specific. Structure-function coupling is dynamic across development (74), with the developmental rate of this coupling itself associated with behavioral outcomes in adolescence (75). Moreover, structure and function may influence each other at different points in time, such that structural and functional pathways exert their effects on psychopathology at different stages of development, potentially explaining modality-specific associations observed in our study. Further work is needed to test how structure-function coupling contributes to emerging psychopathology and whether distinct structural and functional pathways converge or operate independently across symptom dimensions.

A separate analysis of structural and functional brain development in relation to psychopathology is also warranted given our observation of differing developmental patterns. While diffusion measures were highly sensitive to broad developmental effects, DC did not show comparable generalized age effects. This divergence may reflect differences in the developmental trajectories of structural and functional connectivity, with WM microstructure following more consistent age-related patterns, whereas functional DC may be more variable and context dependent (76). This contrasts with previous work showing normative age-related changes in functional connectivity (77,78), which were not reflected in our data. This discrepancy is likely due to differences in the specific measures of functional connectivity used across studies. Voxelwise relative DC has not, to our knowledge, been examined longitudinally in the age range studied here; the absence of normative developmental effects in our sample is therefore informative. It is possible that this developmental window represents a plateau phase in functional centrality, consistent with reports of increases in (parcellation-based) DC in younger cohorts (79) and decreases in DC in older cohorts (80). While DC did not show normative age effects in our study, it was sensitive to symptom-related changes as well as age-specific symptom associations, supporting its value for capturing clinically relevant variability. Together, these findings highlight the need to consider modality-specific developmental dynamics when interpreting brain-behavior associations. Here, gray matter morphometry may provide further complementary insights beyond brain connectivity, shedding light on the extent to which cortical and subcortical structural changes parallel or diverge from connectivity-based associations with psychopathology.

This study benefits from a large, longitudinal dataset and voxelwise analytic approaches that allow for fine-grained detection of developmental change. A key strength is the decomposition of within- and between-subject effects, helping to minimize conflation of cohort differences with true developmental change, a common limitation in adolescent research. However, there are also caveats to consider. The SDQ, while practical and widely validated, provides only a brief dimensional snapshot of psychopathology and may not fully capture clinical complexity (81). Our findings may have been impacted by the specific sample makeup of the IMAGEN cohort. For example, the relatively low average severity of psychopathology in this community-based sample may have limited sensitivity to detect associations that might be more pronounced in higher-risk clinical populations. DTI metrics such as FA are also inherently nonspecific and cannot unambiguously distinguish between biological processes (82). DC reflects the relative importance of a node within the network based on number of connections, but does not capture absolute connectivity strength and or account for known large-scale network organization, limiting interpretability in terms of specific functional systems. Despite these limitations, combining DTI and DC measures provides complementary insights into brain development, capturing microstructural and functional connectivity patterns linked to emerging mental health symptoms. Applied within a large, longitudinal cohort with voxelwise resolution and within-subject modeling, this approach offers a robust, data-driven characterization of neurodevelopmental risk.

Adolescence is a sensitive window for brain development during which connectivity patterns evolve alongside increasing psychopathology. Our findings suggest that externalizing symptoms are particularly linked to WM microstructure, while general psychopathology is associated with dynamic changes in functional connectivity. These results highlight the importance of distinguishing between domain-specific and transdiagnostic neurodevelopmental pathways, advancing our understanding of the neural mechanisms underlying diverse psychiatric outcomes.

Supplementary Material

1

ACKNOWLEDGMENTS AND DISCLOSURES

This work was supported by the European Union–funded FP6 Integrated Project IMAGEN (Reinforcement-Related Behaviour in Normal Brain Function and Psychopathology) (Grant No. LSHM-CT-2007–037286), the Horizon 2020–funded European Research Council Advanced Grant STRATIFY (Brain Network Based Stratification of Reinforcement-Related Disorders) (Grant No. 695313), Horizon Europe “environMENTAL” (Grant No. 101057429), UK Research and Innovation Horizon Europe funding guarantee (Grant Nos. 10041392 and 10038599), Human Brain Project (Grant Nos. HBP SGA 2, 785907, and HBP SGA 3, 945539), the Chinese government via the Ministry of Science and Technology, the German Center for Mental Health (DZPG), the Bundesministerium für Bildung und Forschung (Grant Nos. 01GS08152, 01EV0711, Forschungsnetz AERIAL 01EE1406A and 01EE1406B, and Forschungsnetz IMAC-Mind 01GL1745B), the Deutsche Forschungsgemeinschaft (Grant Nos. 458317126 [COPE], 186318919 [FOR 1617], 178833530 [SFB 940], 386691645 [NE 1383/14–1], 402170461 [TRR 265], and 454245598 [IRTG 2773]), the Medical Research Foundation and Medical Research Council (Grant Nos. MR/R00465X/1 and MR/S020306/1), the National Institutes of Health (NIH)–funded ENIGMA grants (Grant Nos. 5U54EB020403–05, 1R56AG058854–01, and U54 EB020403), NIH (Grant No. R01DA049238), Science Foundation Ireland (Grant No. 16/ERCD/3797), and National Natural Science Foundation of China (Grant No. 82150710554). Further support was provided by the Agence Nationale de la Recherche (Grant Nos. ANR-12-SAMA-0004 and AAPG2019—GeBra), ERA-NET NEURON (Grant Nos. AF12-NEUR0008–01—WM2NA and ANR-18-NEUR00002–01—ADORe), the Fondation de France (Grant No. 00081242), the Fondation pour la Recherche Médicale (Grant No. DPA20140629802), the Mission Interministérielle de Lutte Contre les Drogues et les Conduites Addictives, the L’Assistance Publique–Hôpitaux de Paris and Institut National de la Santé et de la Recherche Médicale (interface grant), Paris Sud University IDEX 2012, the Fondation de l’Avenir (Grant No. AP-RM-17–013), and the Fédération pour la Recherche sur le Cerveau. LV acknowledges funding from the Wellcome Trust (Wellcome Trust Institutional Strategic Support Fund; 204823/Z/16/Z).

For the purpose of open access, the authors have applied a Creative Commons Attribution (CC BY) license to any author accepted manuscript version that arises from this article.

TB served in an advisory or consultancy role for AGB Pharma, eye level, InfectoPharm, Medice, Neurim Pharmaceuticals, Oberberg, and Takeda; received conference support or speaker’s fee from Janssen-Cilag, Medice, and Takeda; and received royalties from Hogrefe, Kohlhammer, CIP-Medien, and Oxford University Press; the present work is unrelated to these relationships. GJB has received honoraria from GE HealthCare for teaching on scanner programming courses. LP served in an advisory or consultancy role for Roche and Vifor Pharm; received a speaker’s fee from Shire; and received royalties from Hogrefe, Kohlhammer, and Schattauer; the present work is unrelated to these relationships. All other authors report no biomedical financial interests or potential conflicts of interest.

Footnotes

Supplementary material cited in this article is available online at https://doi.org/10.1016/j.bpsc.2025.09.015.

REFERENCES

  • 1.Fox MD, Snyder AZ, Vincent JL, Corbetta M, Van Essen DC, Raichle ME (2005): The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc Natl Acad Sci U S A 102:9673–9678. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Lebel C, Deoni S (2018): The development of brain white matter microstructure. Neuroimage 182:207–218. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Edde M, Leroux G, Altena E, Chanraud S (2021): Functional brain connectivity changes across the human life span: From fetal development to old age. J Neurosci Res 99:236–262. [DOI] [PubMed] [Google Scholar]
  • 4.Vása F, Romero-Garcia R, Kitzbichler MG, Seidlitz J, Whitaker KJ, Vaghi MM, et al. (2020): Conservative and disruptive modes of adolescent change in human brain functional connectivity. Proc Natl Acad Sci U S A 117:3248–3253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Ferschmann L, Bos MG, Herting MM, Mills KL, Tamnes CK (2022): Contextualizing adolescent structural brain development: Environmental determinants and mental health outcomes. Curr Opin Psychol 44:170–176. [DOI] [PubMed] [Google Scholar]
  • 6.Dalsgaard S, Thorsteinsson E, Trabjerg BB, Schullehner J, Plana-Ripoll O, Brikell I, et al. (2020): Incidence rates and cumulative incidences of the full spectrum of diagnosed mental disorders in childhood and adolescence. JAMA Psychiatry 77:155–164. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Mennigen E, Bearden CE (2020): Psychosis risk and development: What do we know from population-based studies? Biol Psychiatry 88:315–325. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Noyes BK, Munoz DP, Khalid-Khan S, Brietzke E, Booij L (2022): Is subthreshold depression in adolescence clinically relevant? J Affect Disord 309:123–130. [DOI] [PubMed] [Google Scholar]
  • 9.Brieant A, Cai T, Ip K, Holt-Gosselin B, Gee D (2025): Heterogeneity in developmental trajectories of internalizing and externalizing symptomatology: Associations with risk and protective factors [published online Jan 9]. Child Psychiatry Hum Dev. [DOI] [PubMed] [Google Scholar]
  • 10.Bongers IL, Koot HM, Van Der Ende J, Verhulst FC (2004): Developmental trajectories of externalizing behaviors in childhood and adolescence. Child Dev 75:1523–1537. [DOI] [PubMed] [Google Scholar]
  • 11.Petersen IT, Lindhiem O, LeBeau B, Bates JE, Pettit GS, Lansford JE, et al. (2018): Development of internalizing problems from adolescence to emerging adulthood: Accounting for heterotypic continuity with vertical scaling. Dev Psychol 54:586–599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Lee EJ, Bukowski WM (2012): Co-development of internalizing and externalizing problem behaviors: Causal direction and common vulnerability. J Adolesc 35:713–729. [DOI] [PubMed] [Google Scholar]
  • 13.Katsantonis I (2024): Dynamic interplay of developing internalising and externalising mental health from early childhood to mid-adolescence: Teasing apart trait, state, and cross-cohort effects. PLoS One 19: e0306978. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Kaufmann T, Alnæs D, Doan NT, Brandt CL, Andreassen OA, Westlye LT (2017): Delayed stabilization and individualization in connectome development are related to psychiatric disorders. Nat Neurosci 20:513–515. [DOI] [PubMed] [Google Scholar]
  • 15.Xia CH, Ma Z, Ciric R, Gu S, Betzel RF, Kaczkurkin AN, et al. (2018): Linked dimensions of psychopathology and connectivity in functional brain networks. Nat Commun 9:3003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Alnæs D, Kaufmann T, Doan NT, Córdova-Palomera A, Wang Y, Bettella F, et al. (2018): Association of heritable cognitive ability and psychopathology with white matter properties in children and adolescents. JAMA Psychiatry 75:287–295. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Vanes L, Dolan R (2021): Transdiagnostic neuroimaging markers of psychiatric risk: A narrative review. Neuroimage Clin 30:102634. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Shakeel MK, Metzak PD, Lasby M, Long X, Souza R, Bray S, et al. (2025): Brain connectomes in youth at risk for serious mental illness: A longitudinal perspective. Brain Imaging Behav 19:82–98. [DOI] [PubMed] [Google Scholar]
  • 19.Whittle S, Vijayakumar N, Simmons JG, Allen NB (2020): Internalizing and externalizing symptoms are associated with different trajectories of cortical development during late childhood. J Am Acad Child Adolesc Psychiatry 59:177–185. [DOI] [PubMed] [Google Scholar]
  • 20.Muetzel RL, Blanken LM, van der Ende J, El Marroun H, Shaw P, Sudre G, et al. (2018): Tracking brain development and dimensional psychiatric symptoms in children: A longitudinal population-based neuroimaging study. Am J Psychiatry 175:54–62. [DOI] [PubMed] [Google Scholar]
  • 21.López-Vicente M, Agcaoglu O, Pérez-Crespo L, Estévez-López F, Heredia-Genestar JM, Mulder RH, et al. (2021): Developmental changes in dynamic functional connectivity from childhood into adolescence. Front Syst Neurosci 15:724805. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Vijayakumar N, Ball G, Seal ML, Mundy L, Whittle S, Silk T (2021): The development of structural covariance networks during the transition from childhood to adolescence. Sci Rep 11:9451. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Schumann G, Loth E, Banaschewski T, Barbot A, Barker G, Büchel C, et al. (2010): The IMAGEN study: Reinforcement-related behaviour in normal brain function and psychopathology. Mol Psychiatry 15:1128–1139. [DOI] [PubMed] [Google Scholar]
  • 24.Cao Z, Bennett M, Orr C, Icke I, Banaschewski T, Barker GJ, et al. (2019): Mapping adolescent reward anticipation, receipt, and prediction error during the monetary incentive delay task. Hum Brain Mapp 40:262–283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Albaugh MD, Hudziak JJ, Orr C, Spechler PA, Chaarani B, Mackey S, et al. (2019): Amygdalar reactivity is associated with prefrontal cortical thickness in a large population-based sample of adolescents. PLoS One 14:e0216152. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Baker TE, Castellanos-Ryan N, Schumann G, Cattrell A, Flor H, Nees F, et al. (2019): Modulation of orbitofrontal-striatal reward activity by dopaminergic functional polymorphisms contributes to a predisposition to alcohol misuse in early adolescence. Psychol Med 49:801–810. [DOI] [PubMed] [Google Scholar]
  • 27.Bourque J, Spechler PA, Potvin S, Whelan R, Banaschewski T, Bokde AL, et al. (2017): Functional neuroimaging predictors of self-reported psychotic symptoms in adolescents. Am Psychiatry 174:566–575. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Goodman R (1997): The Strengths and Difficulties Questionnaire: A research note. J Child Psychol Psychiatry 38:581–586. [DOI] [PubMed] [Google Scholar]
  • 29.Guillaume B, Hua X, Thomson P, Waldorp L, Nichols T (2014): The SwE Toolbox: A toolbox for the analysis of longitudinal and repeated measures neuroimaging data. Presented at the Organization for Human Brain Mapping (OHBM) Annual Meeting, June 8–12, Hamburg, Germany. [Google Scholar]
  • 30.Guillaume B, Nichols T (2015): Non-parametric inference for longitudinal and repeated-measures neuroimaging data with the wild bootstrap. Honolulu, Hawaii: Poster presented at the Organization for Human Brain Mapping (OHBM) Annual Meeting. June 14–18. [Google Scholar]
  • 31.Smith SM, Nichols TE (2009): Threshold-free cluster enhancement: Addressing problems of smoothing, threshold dependence and localisation in cluster inference. Neuroimage 44:83–98. [DOI] [PubMed] [Google Scholar]
  • 32.Neuhaus JM, Kalbfleisch JD (1998): Between-and within-cluster covariate effects in the analysis of clustered data. Biometrics 54:638–645. [PubMed] [Google Scholar]
  • 33.Neumann A, Muetzel RL, Lahey BB, Bakermans-Kranenburg MJ, van IJzendoorn MH, Jaddoe VW, et al. (2020): White matter microstructure and the general psychopathology factor in children. J Am Acad Child Adolesc Psychiatry 59:1285–1296. [DOI] [PubMed] [Google Scholar]
  • 34.Hinton KE, Lahey BB, Villalta-Gil V, Meyer FA, Burgess LL, Chodes LK, et al. (2019): White matter microstructure correlates of general and specific second-order factors of psychopathology. Neuroimage Clin 22:101705. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Riem MME, van Hoof MJ, Garrett AS, Rombouts SARB, van der Wee NJA, van IJzendoorn MH, Vermeiren RRJM (2019): General psychopathology factor and unresolved-disorganized attachment uniquely correlated to white matter integrity using diffusion tensor imaging. Behav Brain Res 359:1–8. [DOI] [PubMed] [Google Scholar]
  • 36.Tae WS, Ham BJ, Pyun SB, Kang SH, Kim BJ (2018): Current clinical applications of diffusion-tensor imaging in neurological disorders. J Clin Neurol 14:129–140. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Aung WY, Mar S, Benzinger TL (2013): Diffusion tensor MRI as a biomarker in axonal and myelin damage. Imaging Med 5:427–440. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Burley DT, Genc S, Silk TJ (2021): Childhood conduct problems are associated with reduced white matter fibre density and morphology. J Affect Disord 281:638–645. [DOI] [PubMed] [Google Scholar]
  • 39.Sarkar S, Dell’Acqua F, Froudist Walsh S, Blackwood N, Scott S, Craig MC, et al. (2016): A whole-brain investigation of white matter microstructure in adolescents with conduct disorder. PLoS One 11: e0155475. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Martin JH (2005): The corticospinal system: From development to motor control. Neuroscientist 11:161–173. [DOI] [PubMed] [Google Scholar]
  • 41.Ni HC, Lin HY, Tseng WI, Gau SS (2020): Association of self-regulation with white matter correlates in boys with and without autism spectrum disorder. Sci Rep 10:13811. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Pisner DA, Smith R, Alkozei A, Klimova A, Killgore WD (2017): Highways of the emotional intellect: White matter microstructural correlates of an ability-based measure of emotional intelligence. Soc Neurosci 12:253–267. [DOI] [PubMed] [Google Scholar]
  • 43.Hyde C, Fuelscher I, Efron D, Anderson VA, Silk TJ (2023): Adolescents with ADHD and co-occurring motor difficulties show a distinct pattern of maturation within the corticospinal tract from those without: A longitudinal fixel-based study. Hum Brain Mapp 44:5504–5513. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Damatac CG, Soheili-Nezhad S, Freches GB, Zwiers MP, De Bruijn S, Ikde S, et al. (2022): Longitudinal changes of ADHD symptoms in association with white matter microstructure: a tract-specific fixel-based analysis. Neuroimage Clin 35:103057. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Popov T, Westner BU, Silton RL, Sass SM, Spielberg JM, Rockstroh B, et al. (2018): Time course of brain network reconfiguration supporting inhibitory control. J Neurosci 38:4348–4356. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Frank D, Dewitt M, Hudgens-Haney M, Schaeffer D, Ball B, Schwarz N, et al. (2014): Emotion regulation: Quantitative meta-analysis of functional activation and deactivation. Neurosci Biobehav Rev 45:202–211. [DOI] [PubMed] [Google Scholar]
  • 47.Zhao J, Liu J, Jiang X, Zhou G, Chen G, Ding XP, et al. (2016): Linking resting-state networks in the prefrontal cortex to executive function: A functional near infrared spectroscopy study. Front Neurosci 10:452. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Laurita AC, Hazan C, Spreng RN (2017): Dissociable patterns of brain activity for mentalizing about known others: A role for attachment. Soc Cogn Affect Neurosci 12:1072–1082. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Dodell-Feder D, Koster-Hale J, Bedny M, Saxe R (2011): fMRI item analysis in a theory of mind task. Neuroimage 55:705–712. [DOI] [PubMed] [Google Scholar]
  • 50.Saris IMJ, Penninx BWJH, Dinga R, van Tol MJ, Veltman DJ, van der Wee NJA, Aghajani M (2020): Default mode network connectivity and social dysfunction in major depressive disorder. Sci Rep 10:194. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Hu YX, Shi JY, Xia GY, Liu LF, Li PF, Shan Q, Wang YM (2024): Analysis of functional connectivity changes in attention networks and default mode networks in patients with depression and insomnia. Sleep Breath 28:1731–1742. [DOI] [PubMed] [Google Scholar]
  • 52.Huang BK, Zhou JH, Deng Y, Li CH, Ning BL, Ye ZY, et al. (2024): Perceived stress and brain connectivity in subthreshold depression: Insights from eyes-closed and eyes-open states. Brain Res 1838: 148947. [DOI] [PubMed] [Google Scholar]
  • 53.Qiu X, Lu S, Zhou M, Yan W, Du J, Zhang A, et al. (2021): The relationship between abnormal resting-state functional connectivity of the left superior frontal gyrus and cognitive impairments in youth-onset drug-naive schizophrenia. Front Psychiatry 12:679642. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Debbané M, Lazouret M, Lagioia A, Schneider M, Van De Ville D, Eliez S (2012): Resting-state networks in adolescents with 22q11.2 deletion syndrome: Associations with prodromal symptoms and executive functions. Schizophr Res 139:33–39. [DOI] [PubMed] [Google Scholar]
  • 55.Mechelli A, Gorno-Tempini ML, Price CJ (2003): Neuroimaging studies of word and pseudoword reading: Consistencies, inconsistencies, and limitations. J Cogn Neurosci 15:260–271. [DOI] [PubMed] [Google Scholar]
  • 56.Ishai A, Ungerleider LG, Martin A, Schouten JL, Haxby JV (1999): Distributed representation of objects in the human ventral visual pathway. Proc Natl Acad Sci U S A 96:9379–9384. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Buckley M, Gaffan D, Murray E (1997): Functional double dissociation between two inferior temporal cortical areas: Perirhinal cortex versus middle temporal gyrus. J Neurophysiol 77:587–598. [DOI] [PubMed] [Google Scholar]
  • 58.De la Peña-Arteaga V, Berruga-Sánchez M, Steward T, Martínez-Zalacaín I, Goldberg X, Wainsztein A, et al. (2021): An fMRI study of cognitive reappraisal in major depressive disorder and borderline personality disorder. Eur Psychiatry 64:e56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Hu B, Rao J, Li X, Cao T, Li J, Majoe D, et al. (2017): Emotion regulating attentional control abnormalities in major depressive disorder: An event-related potential study. Sci Rep 7:13530. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Zelazo PD (2020): Executive function and psychopathology: A neurodevelopmental perspective. Annu Rev Clin Psychol 16:431–454. [DOI] [PubMed] [Google Scholar]
  • 61.Yang L, Jin C, Qi S, Teng Y, Li C, Yao Y, et al. (2022): Alterations of functional connectivity of the lateral habenula in subclinical depression and major depressive disorder. BMC Psychiatry 22:588. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Hu C, Jiang W, Huang J, Lin J, Huang J, Wang M, et al. (2023): The amplitude of low-frequency fluctuation characteristics in depressed adolescents with suicide attempts: A resting-state fMRI study. Front Psychiatry 14:1228260. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63.Strawn JR, Wehry AM, Chu WJ, Adler CM, Eliassen JC, Cerullo MA, et al. (2013): Neuroanatomic abnormalities in adolescents with generalized anxiety disorder: A voxel-based morphometry study. Depress Anxiety 30:842–848. [DOI] [PubMed] [Google Scholar]
  • 64.Ding WN, Sun JH, Sun YW, Chen X, Zhou Y, Zhuang ZG, et al. (2014): Trait impulsivity and impaired prefrontal impulse inhibition function in adolescents with internet gaming addiction revealed by a Go/No-Go fMRI study. Behav Brain Funct 10:20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65.Blok E, Geenjaar EP, Geenjaar EA, Calhoun VD, White T (2022): Neurodevelopmental trajectories in children with internalizing, externalizing and emotion dysregulation symptoms. Front Psychiatry 13: 846201. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.Letkiewicz AM, Wakschlag LS, Briggs-Gowan MJ, Cochran AL, Wang L, Norton ES, et al. (2024): Preadolescent externalizing and internalizing symptoms are differentially related to drift-diffusion model parameters and neural activation during a go/no-go task. J Psychiatr Res 178:405–413. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 67.Wang X, Liu Q, Fan J, Gao F, Xia J, Liu X, et al. (2022): Decreased functional coupling within default mode network in major depressive disorder with childhood trauma. J Psychiatr Res 154:61–70. [DOI] [PubMed] [Google Scholar]
  • 68.Dadds MR, Frick PJ (2019): Toward a transdiagnostic model of common and unique processes leading to the major disorders of childhood: The REAL model of attention, responsiveness and learning. Behav Res Ther 119:103410. [DOI] [PubMed] [Google Scholar]
  • 69.Förstner BR, Tschorn M, Reinoso-Schiller N, Maric ic LM, Röcher E, Kalman JL, et al. (2023): Mapping Research Domain Criteria using a transdiagnostic mini-RDoC assessment in mental disorders: A confirmatory factor analysis. Eur Arch Psychiatry Clin Neurosci 273:527–539. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 70.Dugré JR, Eickhoff SB, Potvin S (2022): Meta-analytical transdiagnostic neural correlates in common pediatric psychiatric disorders. Sci Rep 12:4909. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 71.Baum GL, Cui Z, Roalf DR, Ciric R, Betzel RF, Larsen B, et al. (2020): Development of structure–function coupling in human brain networks during youth. Proc Natl Acad Sci U S A 117:771–778. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 72.Larsen B, Sydnor VJ, Keller AS, Yeo BT, Satterthwaite TD (2023): A critical period plasticity framework for the sensorimotor–association axis of cortical neurodevelopment. Trends Neurosci 46:847–862. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 73.Kanel D, Zugman A, Stohr G, Scheinberg B, Cardinale E, Winkler AM, et al. (2025): Structure-function coupling in network connectivity and associations with negative affectivity in a group of transdiagnostic adolescents. J Mood Anxiety Disord 9:100094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 74.Zamani Esfahlani F, Faskowitz J, Slack J, Misic B, Betzel RF (2022): Local structure-function relationships in human brain networks across the lifespan. Nat Commun 13:2053. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 75.Soman SM, Vijayakumar N, Thomson P, Ball G, Hyde C, Silk TJ (2023): Cortical structural and functional coupling during development and implications for attention deficit hyperactivity disorder. Trans-lPsychiatry 13:252. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 76.Bottenhorn KL, Cardenas-Iniguez C, Mills KL, Laird AR, Herting MM (2023): Profiling intra-and inter-individual differences in brain development across early adolescence. Neuroimage 279:120287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 77.Rakesh D, Kelly C, Vijayakumar N, Zalesky A, Allen NB, Whittle S (2021): Unraveling the consequences of childhood maltreatment: deviations from typical functional neurodevelopment mediate the relationship between maltreatment history and depressive symptoms. Biol Psychiatry Cogn Neurosci Neuroimaging 6:329–342. [DOI] [PubMed] [Google Scholar]
  • 78.Sun L, Zhao T, Liang X, Xia M, Li Q, Liao X, et al. (2025): Human lifespan changes in the brain’s functional connectome. Nat Neurosci 28:891–901. [DOI] [PubMed] [Google Scholar]
  • 79.Tian Y, Xu G, Zhang J, Chen K, Liu S (2023): Nodal properties of the resting-state brain functional network in childhood and adolescence. J Neuroimaging 33:1015–1023. [DOI] [PubMed] [Google Scholar]
  • 80.Zuo XN, Ehmke R, Mennes M, Imperati D, Castellanos FX, Sporns O, et al. (2012): Network centrality in the human functional connectome. Cereb Cortex 22:1862–1875. [DOI] [PubMed] [Google Scholar]
  • 81.Wolpert M, Cheng H, Deighton J (2015): Measurement issues: Review of four patient reported outcome measures: SDQ, RCADS, C/ORS and GBO—their strengths and limitations for clinical use and service evaluation. Child Adolesc Ment Health 20:63–70. [DOI] [PubMed] [Google Scholar]
  • 82.Figley CR, Uddin MN, Wong K, Kornelsen J, Puig J, Figley TD (2022): Potential pitfalls of using fractional anisotropy, axial diffusivity, and radial diffusivity as biomarkers of cerebral white matter microstructure. Front Neurosci 15:799576. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2202782-supplement-1.pdf (1.4MB, pdf)

RESOURCES