Neural Timescale of Adolescents Major Depressive Disorder

Yujun Gao, Sanwang Wang, Hong Wei, Jiali He, Geoffrey Liu, Lang Liu, Xiaoqiang Liu, Xiao Li, Xianwei Guo, Xiaobo LiubioRxiv 2025 · DOI

TL;DR

Adolescents with Major Depressive Disorder (MDD) show widespread disruptions in the temporal stability of their brain activity, with faster processing in many areas but slower processing in key regions like the temporoparietal junction. These altered neural timescales are linked to symptom severity, early life trauma, and even predict clinical outcomes, suggesting they could be a valuable biomarker for youth depression.

What Problem Does This Paper Address?

Major Depressive Disorder (MDD) in adolescents is a complex condition with varied symptoms and underlying neurodevelopmental issues. While we know that brain activity changes with age, the specific temporal dynamics of neural activity in adolescent MDD and how these dynamics relate to clinical features and biological mechanisms remain largely unknown. This paper aims to fill that gap by investigating "intrinsic neural timescales" (INT) – a measure of how stable neural activity is over time – in adolescents with MDD compared to healthy controls. The researchers want to understand if these timescales are altered in MDD, and if so, how these alterations connect to symptoms, trauma, and the brain's underlying neurobiology.

Methods

The study employed a multi-faceted approach:

  • Resting-State fMRI: This was the primary tool to measure neural activity while participants were at rest.
  • Neural Timescale Calculation: Using the resting-state fMRI data, the researchers calculated the intrinsic neural timescale (INT) for different brain regions. This involved analyzing the autocorrelation of the blood-oxygen-level-dependent (BOLD) signal, essentially measuring how long neural activity patterns persist.
  • Network Attack Analysis: To understand the functional importance of regions with altered timescales, they simulated "lesions" by removing these regions from the brain's functional connectome and observed the impact on network modularity and clustering.
  • Machine Learning (Support Vector Regression - SVR): SVR models were trained to predict individual symptom severity (depression and anxiety) using the altered neural timescale patterns.
  • Biophysical Modeling: A parametric mean-field model (pMFM) was used to simulate neural population dynamics and investigate how local microcircuit properties (recurrent excitation and external input) relate to regional neural timescales.
  • Neurotransmitter Receptor Mapping: PET-derived receptor density maps were correlated with neural timescale alterations to explore neurochemical associations.
  • Transcriptomic Enrichment Analysis: Gene expression data was analyzed to identify molecular pathways associated with the observed timescale changes.

Datasets

  • Resting-state fMRI data: Collected from adolescents diagnosed with MDD and healthy controls (HCs).
  • Childhood Trauma Questionnaire (CTQ): Used to assess early-life adversity.
  • PET-derived neurotransmitter receptor maps: Data from the Allen Human Brain Atlas and other sources were used to map neurotransmitter systems.
  • Gene expression data: From the Allen Human Brain Atlas for transcriptomic analysis.

Key Insights and Ideas

  • Temporal Fingerprint of MDD: The study suggests that altered neural timescales are not just a byproduct of MDD but a core pathophysiological feature, acting as a "temporal fingerprint" of the disorder.
  • Bridging Scales: The findings elegantly bridge macroscale brain dynamics (network organization) with microcircuit and molecular architecture, offering a comprehensive view of MDD's neural underpinnings.
  • Biomarker Potential: The ability of neural timescales to predict symptom severity and associate with clinical outcomes highlights their potential as biomarkers for diagnosis and treatment monitoring.
  • Early Adversity Link: The strong association between altered timescales and childhood trauma underscores the lasting impact of early life experiences on brain development and mental health.

Key Results

  • Widespread Reductions in INT: Adolescents with MDD showed significantly reduced INT in frontal, parietal, and sensorimotor regions, indicating faster, less stable neural activity.
  • Prolongation in TPJ: In contrast, the left temporoparietal junction (TPJ) exhibited prolonged INT in MDD, suggesting region-specific functional abnormalities.
  • Network Disruption: Altered timescales significantly impaired network modularity and clustering, indicating a disruption in the brain's functional architecture.
  • Predictive Power: SVR models successfully predicted individual depression (HAMD: r = 0.32, p < 0.05) and anxiety (HAMA: r = 0.33, p < 0.05) severity based on neural timescale patterns.
  • Association with Trauma and Suicidality: Reduced timescales were significantly associated with suicidal ideation and childhood trauma, particularly emotional and physical neglect.
  • Circuit-Level Differences: In healthy controls, INT correlated with local recurrent excitation, while in MDD, it correlated more with external input strength, suggesting a shift towards bottom-up processing.
  • Neurochemical Links: Altered timescales colocalized with serotonergic, dopaminergic, and cholinergic systems.
  • Molecular Pathways: Transcriptomic analysis revealed associations with genes involved in mitochondrial function, synaptic signaling, and metabolic regulation.

Limitations and Future Work

  • Cross-sectional Design: The study's cross-sectional nature limits causal inferences. Longitudinal studies are needed to determine if timescale alterations precede symptom onset or are a consequence of chronic psychopathology.
  • Sample Heterogeneity: The focus on adolescents with MDD, while specific, means that variations in symptom profiles and medication histories might introduce variability.
  • Resolution Limitations: The resolution of PET receptor maps and gene expression atlases is coarse and may not capture individual-level variability.
  • Generalizability: The generalizability of findings to adult populations or other psychiatric conditions needs to be established.

Future work could explore whether specific regions with altered timescales (like the TPJ) are targets for cognitive training or neuromodulatory interventions. Further research into the molecular basis of temporal dysregulation, particularly concerning cellular bioenergetics and synaptic integrity, could lead to novel, metabolism-oriented therapeutic approaches.

Summary updated

Abstract

Abstract Adolescent major depressive disorder (MDD) is characterized by heterogeneous symptomatology and complex neurodevelopmental underpinnings. Here, we investigated whether cortical intrinsic neural timescales (INT)—a measure of temporal stability in neural activity—are altered in adolescents with MDD and whether these alterations relate to clinical symptoms, suicidality, early-life adversity, and underlying neurobiological mechanisms. Using resting-state fMRI in adolescents with MDD and healthy controls (HCs), we found widespread reductions in INT in frontal, parietal, and sensorimotor regions, with a notable prolongation in the left temporoparietal junction. Disrupted timescales significantly impaired network modularity and clustering, and SVR-based machine learning revealed that altered INT patterns predicted individual depression and anxiety severity. INT abnormalities were further associated with suicidal ideation and childhood trauma, particularly emotional and physical neglect. Biophysical modeling linked INT variations to local recurrent excitation and external input strength, differing between HCs and MDD. Spatial correlations with PET-derived neurotransmitter receptor maps demonstrated that INT alterations colocalize with serotonergic, dopaminergic, and cholinergic systems. Transcriptomic enrichment analysis revealed associations with genes involved in mitochondrial function, synaptic signaling, and metabolic regulation. Together, these findings identify neural timescale disruption as a core pathophysiological feature of adolescent MDD, bridging macroscale dynamics with microcircuit and molecular architecture. INT may serve as a promising biomarker and mechanistic target for precision psychiatry in youth depression.
BibTeX
@misc{gao2025neural,
  author = {Yujun Gao and Sanwang Wang and Hong Wei and Jiali He and Geoffrey Liu and Lang Liu and Xiaoqiang Liu and Xiao Li and Xianwei Guo and Xiaobo Liu and Xiaobo Liu and Xiaobo Liu},
  title = {{Neural Timescale of Adolescents Major Depressive Disorder}},
  year = {2025},
  doi = {10.1101/2025.07.01.662431},
  howpublished = {bioRxiv},
}
Figure