TL;DR
This study used normative modeling to map typical brain development trajectories in children and adolescents. It found that deviations from these typical trajectories, particularly in cortical volume, are associated with internalizing, externalizing, and emotion dysregulation symptoms, with emotion dysregulation symptoms showing the most widespread brain differences.
What Problem Does This Paper Address?
Understanding how the developing brain relates to psychopathology in children and adolescents is crucial, but previous research has yielded inconsistent findings. Studies often report differing brain regions and even opposite directions of association between brain morphology and psychopathology. This inconsistency may stem from non-linear brain development across age and variations in study samples. This paper aims to address this by establishing normative developmental curves for brain morphology and then examining how deviations from these typical curves relate to different domains of psychopathology.
Methods
The researchers employed normative modeling using Gaussian Process Regression (GPR) to create typical developmental curves for (sub-)cortical volume and cortical thickness across 37 and 32 brain regions, respectively. This technique models the expected brain measure at a given age, along with its variability. By comparing individual children's brain measures to these normative curves, they derived z-scores representing deviations from typical development. These z-scores were then statistically analyzed to determine their association with three domains of psychopathology: internalizing, externalizing, and emotion dysregulation (DP) symptoms. Sensitivity analyses were conducted to assess the robustness of findings, including adjustments for total intracranial volume (ICV) and exploring domain-specific relationships.
Datasets
The study utilized data from the Generation R Study, a large, longitudinal, population-based cohort. Neuroimaging and behavioral data were collected from children at three time points, spanning ages 6 to 16 years.
- T1: 842 children (ages 6-10)
- T2: 2,708 children (ages 8-12)
- T3: 1,904 children (ages 13-16)
This provided a substantial sample for establishing normative curves and examining developmental trajectories.
Key Insights and Ideas
- Brain Development is Highly Regional and Age-Dependent: The study clearly illustrates that different brain regions follow distinct developmental trajectories, with some increasing in volume, others decreasing, and some showing inverted U-shaped patterns. This highlights the complexity of neurodevelopment.
- Deviations, Not Absolute Values, Matter for Psychopathology: The core insight is that it's not just the absolute size or thickness of a brain region, but rather how it deviates from the typical developmental path for its age, that is linked to psychopathology.
- Emotion Dysregulation Symptoms Show Broadest Brain Associations: Children with higher emotion dysregulation symptoms exhibited the most widespread negative deviations from typical brain development across various cortical and subcortical regions. This suggests a more pervasive impact of these symptoms on brain development.
- Global vs. Region-Specific Effects: While many associations between psychopathology and brain deviations were widespread, a significant portion of these effects attenuated after controlling for total intracranial volume. This suggests a potential global influence of psychopathology on brain development, with some region-specific effects remaining.
- Cortical Thickness Shows Different Patterns: Unlike cortical volume, which showed widespread negative deviations associated with psychopathology, cortical thickness showed more nuanced relationships, with some positive deviations linked to specific symptoms.
Key Results
- Normative modeling successfully charted developmental curves for (sub-)cortical volume and cortical thickness across the cerebrum.
- Psychopathology symptoms were associated with negative deviations from typical development for cortical volume in widespread regions.
- Positive deviations from typical development for cortical thickness were observed in specific regions (orbitofrontal, frontal pole, pericalcarine, posterior cingulate) in relation to psychopathology.
- Children with emotion dysregulation (DP) symptoms were most heavily affected, showing the broadest range of significant associations with deviations in brain development.
- The anterior cingulate cortex (ACC) showed negative deviations related to all three psychopathology domains for cortical volume.
- For cortical thickness, the orbitofrontal cortex showed positive deviations related to externalizing and DP symptoms.
- A significant positive interaction effect between age and deviations in fusiform and parahippocampal volume was observed for externalizing symptoms, indicating that these deviations become larger with age.
Limitations and Future Work
- Sample Distribution: While the Generation R Study is large, the majority of participants had relatively low levels of psychopathology, which may limit the power to detect associations in individuals with clinical psychopathology.
- Age Distribution: The data was not equally distributed across all ages, which could influence results around ages with fewer data points.
- MRI Scanner Differences: Data from T1 was acquired on a different MRI scanner than T2 and T3, which could introduce variability.
- Small Effect Sizes: The study acknowledges that small effect sizes are common in this field, and the inter-individual variability in brain measures contributes to this.
- Future Work: The authors suggest extending this work by assessing the relationship between the general psychopathology factor and brain morphology, and by using longitudinal MRI data to better understand temporal changes in deviations and their relationship to psychopathology. Investigating the bidirectional relationship between brain and behavioral development is also highlighted as a future direction.
Summary updated