Skip to main content
NCBI home page
Neuropsychiatric Disease and Treatment logoLink to Neuropsychiatric Disease and Treatment
. 2026 Sep 24;22:636580. doi: 10.2147/NDT.S636580

Associations Between Thalamic Gray Matter Volume and Psychomotor Symptoms in Major Depressive Disorder: A Voxel-Based Morphometry Study

Yanxi Gu 1,2,*, Xinlin Wang 1,2,*, Fei Liu 1,2, Jiaxin Wu 1, Qiong Zhang 1, Haonan Ma 1,2, Jingyi Bai 1,2, Xiangyun Long 3,✉, Zheng Lu 1,2,4,✉
PMCID: PMC13618326  PMID: 42807528

Abstract

Background

Major depressive disorder (MDD) is a heterogeneous disorder with diverse symptom profiles and treatment responses. Psychomotor retardation and agitation represent clinically relevant dimensions of MDD heterogeneity, but their neurobiological basis remains unclear. Given the role of the thalamus in motor regulation and affective processing, this study examined thalamic gray matter volume (GMV) differences among patients with MDD with predominant psychomotor retardation (MDD-R), patients with MDD with predominant psychomotor agitation (MDD-A), and healthy controls (HCs).

Methods

Structural MRI data from the REST-meta-MDD consortium included 54 patients with MDD-A, 123 with MDD-R, and 122 HCs. Total, hemispheric, and 30 subregional thalamic GMV measures were extracted using voxel-based morphometry and the AAL3v1 atlas. ComBat harmonization was applied before analysis. Group differences were tested with analysis of covariance controlling for sex, age, education, and total gray matter volume. Exploratory partial Spearman correlations examined associations with psychomotor symptoms, HAMD(adj), and anxiety symptoms.

Results

After FDR correction, significant group effects were limited to the bilateral ventral anterior nucleus (VA). Patients with MDD-R had greater bilateral VA GMV than HCs, whereas MDD-A did not differ from HCs. No direct MDD-A vs MDD-R difference survived FDR correction; only the left inferior pulvinar showed a nominal difference. Exploratory correlations linked several thalamic nuclei, including the reuniens nucleus, ventral posterolateral nucleus, lateral geniculate nucleus, and anteroventral nucleus, to psychomotor, depressive, or anxiety symptoms.

Conclusion

MDD-R was associated with selective bilateral VA structural differences rather than widespread thalamic abnormalities. No direct thalamic GMV difference between MDD-A and MDD-R survived FDR correction. The exploratory symptom associations suggest that thalamic structure may relate to multidimensional clinical heterogeneity in MDD. These findings require replication given the cross-sectional design, unequal subgroup sizes, and incomplete medication data.

Keywords: major depressive disorder, psychomotor retardation, psychomotor agitation, thalamus, gray matter volume, voxel-based morphometry, REST-meta-MDD

Background

Major depressive disorder (MDD) is a major public health problem worldwide. Data from the World Health Organization (WHO) and the Global Burden of Disease (GBD) study show that depressive disorders remain a major source of global disease burden and are associated with substantial health loss and social impairment.1,2 MDD is clinically heterogeneous, with marked variation in symptom profiles, treatment response, and prognosis.3 This heterogeneity complicates precise diagnosis and individualized treatment. Subtyping depression according to core clinical features has therefore become an important research direction.4

Psychomotor symptoms provide one clinically relevant basis for subtyping MDD. Patients with MDD-R commonly show reduced movement, slowed speech, impaired cognitive processing, diminished motivation, and reduced behavioral initiation.5,6 Patients with MDD-A more often present with inner tension, restlessness, irritability, and motor agitation.7 Psychomotor agitation has also been associated with suicidal behavior.8 Psychomotor disturbance may also be associated with differential response to antidepressant medication and ECT.9,10 Studying the neural basis of these psychomotor phenotypes may therefore help clarify clinical heterogeneity in MDD and support more targeted treatment approaches.11

The thalamus has traditionally been viewed as a relay for sensory information to the cerebral cortex, but it is now recognized as an integrative structure involved in sensory, motor, cognitive, and affective functions.12–14 Thalamic nuclei have distinct anatomical connections and functional roles.15 The pulvinar complex regulates communication between cortical regions and modulates sensory processing according to behavioral demands.16 Among thalamic nuclei, the ventral anterior (VA) nucleus is a component of motor thalamic circuitry and participates in basal ganglia-thalamocortical pathways involved in movement preparation and initiation.17,18 In contrast, the lateral geniculate nucleus (LGN) is a primary visual thalamic nucleus involved in visual information processing and thalamocortical transmission.19 The thalamus is also embedded within cortico-striatal-pallidal-thalamic and limbic circuits implicated in mood regulation, motivation, and the pathophysiology of depression.20,21 Structural alterations across distinct thalamic subfields have been reported in MDD.22 However, whether specific thalamic subregions are differentially associated with psychomotor symptoms remains unclear.

Previous structural MRI studies have reported thalamic abnormalities in MDD, including reductions in whole-thalamic volume and alterations involving individual thalamic nuclei, although the direction and distribution of findings have varied across samples.22,23 Functional imaging studies have also demonstrated region-specific thalamocortical dysconnectivity in MDD, involving prefrontal, motor, parietal, default-mode, and cerebellar regions.24 Few studies have directly compared patients with predominant psychomotor agitation and those with predominant psychomotor retardation. It therefore remains unclear whether thalamic structural alterations exhibit subtype-specific patterns, particularly whether specific changes are associated with psychomotor agitation or retardation. Furthermore, many neuroimaging studies have investigated the thalamus as a whole, overlooking the potentially distinct anatomical and functional roles of individual thalamic subregions.13,22 Consequently, it remains difficult to determine whether thalamic abnormalities represent a general feature of MDD or are specifically related to distinct psychomotor dimensions.

This study compared total, hemispheric, and subregional thalamic gray matter volume (GMV) among patients with MDD-A, patients with MDD-R, and healthy controls (HCs), and examined associations between thalamic GMV and psychomotor symptom dimensions. Based on the role of the thalamus in motor initiation and regulation and on prior structural findings in MDD, we hypothesized that patients with MDD-R would show more prominent thalamic structural alterations, particularly in motor-related thalamic regions, and that subregional thalamic GMV would be associated with psychomotor symptom severity. These analyses were intended to identify candidate imaging features for future longitudinal and multimodal studies.

Materials and Methods

Participants

Data were obtained from the REST-meta-MDD multisite neuroimaging consortium.25,26 All participants were screened using standardized inclusion and exclusion criteria. The initial sample included 1300 patients who met DSM-IV or ICD-10 diagnostic criteria for MDD and 1128 HCs. Healthy controls were recruited from the community and had no personal or family history of psychiatric disorders. After excluding patients younger than 18 years or older than 65 years and those with missing key clinical information, including illness duration and HAMD Items 8 or 9, 512 patients remained eligible. Application of the psychomotor subtype criteria yielded 177 patients with MDD, including 54 with MDD-A and 123 with MDD-R. One HC from a site with no other eligible participant was excluded before ComBat harmonization because site-specific scale parameters could not be estimated reliably. The final analytic sample therefore included 299 participants: 177 patients with MDD and 122 HCs. The present secondary analysis was approved by the Ethics Committee of Shanghai Tongji Hospital (Approval No. 2022–058). The original studies contributing data to the REST-meta-MDD consortium were approved by the respective local institutional review boards or ethics committees, and written informed consent was obtained from all participants.

Clinical Assessment and Subtype Definition

Depressive symptom severity was assessed using the 17-item Hamilton Depression Rating Scale (HAMD-17).27 The Chinese version of the HAMD-17 has shown good reliability and validity.28

Psychomotor subgroups were defined using the HAMD-17 items that assess psychomotor symptoms. Item 8 assesses psychomotor retardation and Item 9 assesses psychomotor agitation; both are scored from 0 to 4. Item 8 ranges from no retardation to increasing motor and verbal slowing, difficulty during the interview, and stupor, whereas Item 9 ranges from no agitation to restlessness and marked motor agitation. Previous studies have used scores of ≥2 on these items to indicate clinically evident psychomotor disturbance. A HAMD Item 8 score ≥2 has been used to define psychomotor retardation,29 and an Item 9 score ≥2 has been used to define psychomotor agitation.30 We therefore applied mutually exclusive criteria: MDD-A was defined as Item 9 ≥2 and Item 8 <2, whereas MDD-R was defined as Item 8 ≥2 and Item 9 <2. Clinical assessments were completed by psychiatrists or research assistants who had received standardized rater training.

MRI Acquisition, Preprocessing, and Quality Control

All participants underwent high-resolution structural T1-weighted MRI. Structural images were preprocessed using the Data Processing Assistant for Resting-State fMRI (DPARSF)31 in conjunction with SPM8 and the VBM8 toolbox,32 following the standardized structural MRI preprocessing pipeline of the REST-meta-MDD consortium.26,33

Following correction for image intensity inhomogeneity, T1-weighted images were segmented into gray matter (GM), white matter, and cerebrospinal fluid. The segmented GM images were then spatially normalized to Montreal Neurological Institute (MNI) space using DARTEL (Diffeomorphic Anatomical Registration Through Exponentiated Lie Algebra) nonlinear registration, followed by modulation to preserve local tissue volume.34 Total thalamic, hemispheric, and subregional GMV measures were extracted from modulated, normalized gray matter images (mwc1 images). Thalamic subregions were defined using the AAL3v1 atlas,35 and ROI-based volume extraction was performed before spatial smoothing. The normalized and modulated GM images were subsequently smoothed with an 8-mm full width at half maximum (FWHM) Gaussian kernel in accordance with the REST-meta-MDD structural imaging preprocessing pipeline.

Quality control was performed before and after preprocessing, including checks of image completeness, tissue segmentation quality, and homogeneity of the normalized GM maps. Images with marked artifacts, structural distortion, or segmentation errors were excluded.

Because the data were acquired across multiple sites, thalamic GMV measures were harmonized using ComBat before statistical analysis to reduce nonbiological variability associated with scanner and site differences.36,37

Statistical Analysis

Demographic and clinical analyses were conducted in SPSS version 27.0. Continuous variables are presented as mean ± standard deviation, and categorical variables as number (percentage). Continuous demographic variables were compared across MDD-A, MDD-R, and HC groups using one-way analysis of variance (ANOVA), and categorical variables were compared using chi-square tests. Clinical variables available only in the patient groups, including illness duration, HAMD-17, HAMD(adj), and HAMA scores, were compared between MDD-A and MDD-R using independent-samples t-tests.

Before group and correlation analyses, ComBat was applied to harmonize thalamic GMV measures across sites. Site was entered as the batch variable, while group, sex, age, education, and total gray matter volume were preserved as biological covariates. All 299 participants (MDD-A, n=54; MDD-R, n=123; HC, n=122) were included in the demographic and imaging analyses, and all subsequent imaging analyses used ComBat-harmonized thalamic GMV measures.

Three-group differences in thalamic GMV were tested using analysis of covariance (ANCOVA). Total thalamic volume, left and right hemispheric thalamic volumes, and 30 thalamic subregional volumes were entered as dependent variables, with group as the fixed factor and sex, age, years of education, and total gray matter volume as covariates. Multiple comparisons were controlled using the Benjamini-Hochberg false discovery rate (FDR). The three total/hemispheric measures were corrected as one family, and the 30 subregional measures as a separate family. For subregions with an FDR-corrected omnibus group effect, two prespecified post hoc contrasts (MDD-A vs HC and MDD-R vs HC) were tested and corrected within each region using the Benjamini-Hochberg procedure.

To directly compare the two psychomotor subtypes, ANCOVA was repeated in the MDD-A and MDD-R groups only. In addition to sex, age, education, and total gray matter volume, these models included illness duration and HAMD(adj) as covariates. HAMD(adj) was defined as the HAMD-17 total score minus Items 8 (psychomotor retardation) and 9 (psychomotor agitation). FDR correction was applied separately to the three total/hemispheric measures and the 30 subregional measures.

Exploratory partial Spearman correlations were used to examine associations between thalamic GMV and HAMD-8, HAMD-9, HAMD(adj), and HAMA scores in the full MDD sample and separately in the MDD-A and MDD-R groups. Analyses of HAMD-8, HAMD-9, and HAMA controlled for sex, age, years of education, illness duration, total gray matter volume, site, and HAMD(adj). Analyses involving HAMD(adj) controlled for sex, age, years of education, illness duration, total gray matter volume, and site. Symptom scores, thalamic GMV, and continuous covariates were rank-transformed; sex and site were coded as categorical variables. Residuals from separate covariate-adjusted models for symptoms and GMV were then correlated. Because these analyses were exploratory, two-sided P values are reported without correction for multiple comparisons and are interpreted cautiously. The significance threshold was alpha=0.05.

Results

Participant Characteristics

The HC, MDD-A, and MDD-R groups did not differ in sex distribution (χ2 = 0.613, P = 0.736), age (F = 1.192, P = 0.305), or total gray matter volume (TGMV; F = 1.026, P = 0.360). Years of education differed across groups (F = 4.475, P = 0.012); Tukey post hoc testing showed fewer years of education in MDD-R than in HCs (P = 0.012). Education was therefore included as a covariate in subsequent imaging analyses. MDD-A and MDD-R did not differ in illness duration (t = −0.150, P = 0.881), HAMD-17 total score (t=0.141, P=0.888), or HAMD(adj) (t = −0.070, P = 0.945). HAMA data were available for 20 patients with MDD-A and 34 with MDD-R. Mean HAMA scores were 22.45 ± 8.44 in MDD-A and 17.91 ± 10.66 in MDD-R, with no significant group difference (t = 1.626, P = 0.110). Participant characteristics are shown in Table 1.

Table 1.

Demographic and Clinical Characteristics of Participants

MDD Healthy Controls (N=122) F/χ2/t P
MDD-A (N=54) MDD-R(N=123)
Gender (Female, %) 35 (64.8%) 87 (70.7%) 84 (68.9%) 0.613b 0.736
Age (Mean±SD, years) 32.57 ± 9.39 35.35 ± 12.14 33.82 ± 12.00 1.192a 0.305
Education (Mean±SD, years) 12.48 ± 3.09 11.33±4.19 12.73±3.70 4.475a 0.012
Illness duration (Mean±SD, months) 34.07 ± 80.83 35.68±58.01 – −0.150c 0.881
HAMD-17 (Mean±SD) 24.26 ± 4.80 24.14±5.42 – 0.141c 0.888
HAMD(adj) (Mean±SD) 21.26 ± 4.74 21.32 ± 5.24 – −0.070c 0.945
HAMA (Mean±SD)d 22.45 ± 8.44 (n=20) 17.91±10.66 (n=34) – 1.626c 0.110
TGMV (Mean±SD, cm3) 670.67 ± 73.60 655.29 ± 64.31 655.70±75.03 1.026a 0.360
Open in a new tab

Notes: Continuous variables are presented as mean ± standard deviation (Mean±SD), and categorical variables as number (percentage, %). One-way ANOVA was used to compare age, education, and TGMV across the three groups. The chi-square test was used to compare gender distribution. Independent-samples t-tests were used to compare illness duration, HAMD-17, HAMD(adj), and HAMA scores between the MDD-A and MDD-R groups. HAMA data were not available for healthy controls. Statistical notation: aOne-way analysis of variance (ANOVA); bχ2 test; cindependent-samplest-test; dHAMA data were available for 20 participants in the MDD-A group and 34 participants in the MDD-R group.

Abbreviations: MDD, major depressive disorder; MDD-A, major depressive disorder with agitation; MDD-R, major depressive disorder with retardation; HCs, healthy controls; TGMV, total gray matter volume; HAMD-17, 17-item Hamilton Depression Rating Scale; HAMD(adj), HAMD-17 total score excluding psychomotor retardation (Item 8) and psychomotor agitation (Item 9); HAMA, Hamilton Anxiety Rating Scale.

Group Differences in Total, Hemispheric, and Subregional Thalamic GMV

After ComBat harmonization, the three-group analysis included 299 participants (MDD-A, n=54; MDD-R, n=123; HC, n=122). ANCOVA models adjusted for sex, age, years of education, and total gray matter volume. After FDR correction, significant omnibus group effects were confined to the bilateral ventral anterior nucleus (VA): left VA (F = 7.198, PFDR = 0.021, partial η2 = 0.047) and right VA (F = 6.740, PFDR = 0.021, partial η2 = 0.044). Post hoc analyses showed greater bilateral VA GMV in MDD-R than in HCs. The adjusted mean difference was 1.998 for the left VA (95% CI: [0.951, 3.045]; PFDR < 0.001) and 1.970 for the right VA (95% CI: [0.910, 3.030]; PFDR < 0.001). Differences between MDD-A and HCs did not reach significance for the left VA (PFDR = 0.051) or right VA (PFDR = 0.256). The FDR-significant group effects and corresponding post hoc comparisons are summarized in Table 2, while the complete ANCOVA results for all 33 thalamic measures are provided in Supplementary Table S1.

Table 2.

Differences in ComBat-Harmonized Thalamic Subregional Gray Matter Volume Between Depressive Subtype Groups and Healthy Controls

Thalamic Subregions P (FDR) F Partial η2 Adjusted Mean
(MDD-A/MDD-R/HC)		 Pairwise Comparison Mean Difference 95% CI Post Hoc
P (FDR)
Thal VA L 0.021 7.198 0.047 20.383/21.062/19.064 MDD-A vs HC 1.318 [−0.007, 2.644] 0.051
MDD-R vs HC 1.998 [0.951, 3.045] <0.001
Thal VA R 0.021 6.740 0.044 26.715/27.909/25.939 MDD-A vs HC 0.776 [−0.567, 2.118] 0.256
MDD-R vs HC 1.970 [0.910, 3.030] <0.001
Open in a new tab

Notes: Thalamic GMV measures were harmonized across sites using ComBat prior to statistical analysis. The final analytic sample included 299 participants (MDD-A, n=54; MDD-R, n=123; HC, n=122); the single HC participant from a site containing no other eligible participants was excluded before demographic and imaging analyses because within-site scale parameters could not be stably estimated for ComBat. Group differences were assessed using analysis of covariance (ANCOVA), adjusting for sex, age, years of education, and total gray matter volume. Estimated marginal means are presented in the order MDD-A/MDD-R/HC. FDR correction was applied separately across the three global/hemispheric thalamic measures and across the 30 thalamic subregions. For subregions showing significant FDR-corrected omnibus group effects, the two prespecified comparisons with healthy controls (MDD-A vs HC and MDD-R vs HC) were further corrected using the Benjamini-Hochberg FDR procedure within each region. Direct comparisons between the MDD-A and MDD-R groups were conducted separately with additional adjustment for illness duration and HAMD(adj).

Abbreviations: MDD-A, major depressive disorder with agitation; MDD-R, major depressive disorder with retardation; HC, healthy control; VA, ventral anterior thalamic nucleus; GMV, gray matter volume; TGMV, total gray matter volume; CI, confidence interval; FDR, false discovery rate; partial η2, partial eta-squared.

These group differences are illustrated in Figure 1.

Figure 1.

Violin-box plots and MRI slices highlight ventral anterior thalamus volume effects.

Open in a new tab

Group differences in thalamic gray matter volume among MDD-A, MDD-R, and healthy controls.

Notes: (A) Covariate-adjusted gray matter volume in the right and left ventral anterior thalamic nuclei (VA). Significant omnibus group effects were observed in the right VA (F = 6.740, PFDR = 0.021) and left VA (F = 7.198, PFDR = 0.021). Post hoc comparisons showed greater GMV in the MDD-R group than in HCs in both regions (both PFDR < 0.001). ***indicates PFDR < 0.001 for the MDD-R vs HC comparison. (B) Spatial distribution of omnibus F statistics across the 30 thalamic subregions. Warmer colors indicate larger F statistics but do not necessarily indicate statistical significance; only the bilateral VA survived FDR correction. All ANCOVA models were adjusted for sex, age, years of education, and total gray matter volume.

Direct Comparison Between MDD-A and MDD-R

After ComBat harmonization, the direct comparison between MDD-A and MDD-R identified a nominal difference in left inferior pulvinar (PuI) GMV, with lower volume in MDD-R than in MDD-A (F = 3.921, P = 0.049, partial η2 = 0.023). This difference did not survive FDR correction (PFDR = 0.623). Complete results for the direct MDD-A vs MDD-R comparison across all 33 thalamic measures are provided in Supplementary Table S2.

Exploratory Correlations with Psychomotor Symptoms and Overall Depressive Severity

Exploratory partial Spearman correlations were performed using ComBat-harmonized thalamic GMV. HAMD-8 and HAMD-9 analyses adjusted for sex, age, years of education, illness duration, total gray matter volume, site, and HAMD(adj); HAMD(adj) analyses adjusted for sex, age, years of education, illness duration, total gray matter volume, and site. P values were not corrected for multiple comparisons.

In the full MDD sample (n=177), HAMD-8 psychomotor retardation scores were positively associated with right reuniens nucleus (Re) GMV (ρ = 0.168, P = 0.034). In subgroup analyses, HAMD-8 scores were positively associated with right lateral geniculate nucleus (LGN) GMV in MDD-R (ρ=0.192, P = 0.04997), while no association with P<0.05 was observed in MDD-A.

For HAMD-9 psychomotor agitation, scores were positively associated with left ventral posterolateral nucleus (VPL) GMV in the full MDD sample (ρ=0.156, P = 0.0496). In MDD-A, HAMD-9 scores were positively associated with right LGN GMV (ρ=0.384, P = 0.019); no association with P<0.05 was observed in MDD-R.

For HAMD(adj), which excludes HAMD Items 8 and 9, higher scores were associated with lower right LGN GMV in the full MDD sample (ρ= −0.165, P = 0.037). In MDD-A, HAMD(adj) was negatively associated with right anteroventral nucleus (AV) GMV (ρ=-0.365, P =0.024). No association with P<0.05 was observed in MDD-R. HAMD-8, HAMD-9, and HAMD (adj) were not associated with total, left, or right hemispheric thalamic GMV at P < 0.05 (Figure 2). Complete exploratory correlation results for HAMD-8, HAMD-9, and HAMD(adj) across all thalamic measures are provided in Supplementary Table S3.

Figure 2.

Six scatter plots: gray matter volume vs. HAMD measures in major depressive disorder. Six scatter plots with a legend listing MDD-A, MDD-R and All MDD. Each plot shows points and a fitted line with a shaded band. The vertical axis label is Residualized GMV with tick labels at minus 2, minus 1, 0, 1, 2 in all plots. The horizontal axis label varies by plot: Residualized HAMD-8, Residualized HAMD-9, or Residualized HAMD(adj). Thal Re R: x-axis Residualized HAMD-8 with tick labels at minus 3, minus 2, minus 1, 0, 1, 2, 3. y-axis Residualized GMV with tick labels at minus 2, minus 1, 0, 1, 2. The fitted line slopes upward. Points are densest around x near 0 to 1 and y near minus 1 to 1, with additional points around x near minus 2 to minus 1. Thal LGN R with Residualized HAMD-8: x-axis Residualized HAMD-8 with tick labels at minus 2, minus 1, 0, 1, 2. y-axis Residualized GMV with tick labels at minus 2, minus 1, 0, 1, 2. The fitted line slopes upward. Points are spread across the full x range, with many points concentrated near x around minus 1 to 1 and y around minus 1 to 1. Thal VPL L: x-axis Residualized HAMD-9 with tick labels at minus 2, minus 1, 0, 1, 2. y-axis Residualized GMV with tick labels at minus 2, minus 1, 0, 1, 2. The fitted line slopes upward. Points cluster around x near 0 and y near minus 1 to 1, with additional points extending to x near 2. Thal LGN R with Residualized HAMD-9: x-axis Residualized HAMD-9 with tick labels at minus 2, minus 1, 0, 1, 2. y-axis Residualized GMV with tick labels at minus 3, minus 2, minus 1, 0, 1, 2. The fitted line slopes upward. Points are scattered broadly, with several points near y below minus 2 and several near y above 1. Thal LGN R with Residualized HAMD(adj): x-axis Residualized HAMD(adj) with tick labels at minus 2, minus 1, 0, 1, 2. y-axis Residualized GMV with tick labels at minus 2, minus 1, 0, 1, 2. The fitted line slopes slightly downward. Points are densest around x near 0 to 1 and y near minus 1 to 1, with overlap of the plotted groups. Thal AV R: x-axis Residualized HAMD(adj) with tick labels at minus 1.5, minus 1.0, minus 0.5, 0, 0.5, 1.0, 1.5, 2.0. y-axis Residualized GMV with tick labels at minus 2.0, minus 1.5, minus 1.0, minus 0.5, 0, 0.5, 1.0, 1.5, 2.0. The fitted line slopes downward. Points are spread across the x range, with many points around y near 0 and several points near y close to minus 2 and near y close to 2.

Open in a new tab

Exploratory associations between psychomotor symptoms, overall depressive symptom severity, and thalamic gray matter volume in patients with major depressive disorder.

Notes: Partial Spearman correlation analyses were performed using ComBat-harmonized thalamic gray matter volume (GMV). Scatterplots show covariate-residualized symptom scores and GMV values, with regression lines and 95% confidence intervals. From left to right, the upper panels show associations between HAMD-8 and the right reuniens nucleus (Re) in the combined MDD sample (ρ = 0.168, P = 0.034), HAMD-8 and the right lateral geniculate nucleus (LGN) in the MDD-R group (ρ = 0.192, P = 0.04997), and HAMD-9 and the left ventral posterolateral nucleus (VPL) in the combined MDD sample (ρ = 0.156, P = 0.04960). The lower panels show associations between HAMD-9 and the right LGN in the MDD-A group (ρ = 0.384, P = 0.019), HAMD(adj) and the right LGN in the combined MDD sample (ρ = -0.165, P = 0.037), and HAMD(adj) and the right anteroventral nucleus (AV) in the MDD-A group (ρ = -0.365, P = 0.024). For HAMD-8 and HAMD-9 analyses, correlations were adjusted for sex, age, years of education, illness duration, total gray matter volume, site, and HAMD(adj); analyses involving HAMD(adj) were adjusted for sex, age, years of education, illness duration, total gray matter volume, and site. Red and blue points/lines denote MDD-A and MDD-R participants, respectively, whereas black lines indicate associations in the combined MDD sample. Given the exploratory nature of these analyses, P values were not corrected for multiple comparisons. HAMD-8, psychomotor retardation item of the 17-item Hamilton Depression Rating Scale; HAMD-9, psychomotor agitation item; HAMD(adj), HAMD-17 total score excluding Items 8 and 9; MDD-A, major depressive disorder with agitation; MDD-R, major depressive disorder with retardation.

Exploratory Correlations Between Anxiety Symptoms and Thalamic GMV

Among patients with available HAMA data (n=54; MDD-A, n=20; MDD-R, n=34), exploratory partial Spearman correlations examined associations between anxiety symptoms and thalamic GMV, adjusting for sex, age, years of education, illness duration, total gray matter volume, site, and HAMD(adj). P values were not corrected for multiple comparisons.

In the combined MDD sample, HAMA scores were positively associated with total thalamic GMV (ρ =0.430, P = 0.004), right thalamic GMV (ρ =0.503, P<0.001), and left thalamic GMV (ρ =0.306, P = 0.043). At the subregional level, HAMA scores were positively associated with the right lateral posterior nucleus (LP; ρ =0.434, P = 0.003), right ventral lateral nucleus (VL; ρ =0.454, P = 0.002), right VPL (ρ=0.372, P = 0.013), right intralaminar nuclei (IL; ρ=0.551, P < 0.001), left medial mediodorsal nucleus (MDm; ρ =0.419, P=0.005), right MDm (ρ =0.379, P = 0.011), left lateral mediodorsal nucleus (MDl; ρ =0.358, P = 0.017), right MDl (ρ= 0.503, P < 0.001), right medial pulvinar (PuM; ρ =0.449, P = 0.002), left lateral pulvinar (PuL; ρ =0.302, P = 0.046), and right PuL (ρ =0.515, P < 0.001).

In MDD-R, HAMA scores were positively associated with right hemispheric thalamic GMV (ρ =0.446, P = 0.029), as well as right VL (ρ =0.436, P = 0.033), right VPL (ρ = 0.466, P = 0.022), right IL (ρ =0.517, P = 0.010), right PuM (ρ =0.419, P = 0.041), and right PuL (ρ =0.652, P < 0.001). No HAMA-thalamic association with P<0.05 was observed in MDD-A. Nominally significant HAMA–thalamic GMV associations are summarized in Table 3, and complete exploratory correlation results across all thalamic measures are provided in Supplementary Table S4.

Table 3.

Exploratory Partial Correlations Between HAMA Scores and Thalamic Gray Matter Volume

Analysis Sample Brain Region Partial ρ P
All MDD(n=54) Total thalamus 0.430 0.004
Right thalamus 0.503 <0.001
Left thalamus 0.306 0.043
Thal LP R 0.434 0.003
Thal VL R 0.454 0.002
Thal VPL R 0.372 0.013
Thal IL R 0.551 <0.001
Thal MDm L 0.419 0.005
Thal MDm R 0.379 0.011
Thal MDl L 0.358 0.017
Thal MDl R 0.503 <0.001
Thal PuM R 0.449 0.002
Thal PuL L 0.302 0.046
Thal PuL R 0.515 <0.001
MDD-R(n=34) Right thalamus 0.446 0.029
Thal VL R 0.436 0.033
Thal VPL R 0.466 0.022
Thal IL R 0.517 0.010
Thal PuM R 0.419 0.041
Thal PuL R 0.652 <0.001
Open in a new tab

Notes: Exploratory partial Spearman correlation analyses were performed between HAMA scores and ComBat-harmonized thalamic GMV, adjusting for sex, age, years of education, illness duration, total gray matter volume, site, and HAMD(adj). Only associations with uncorrected P < 0.05 are shown. Given the exploratory nature of these analyses, P values were not corrected for multiple comparisons and should be interpreted cautiously. HAMA data were available for 54 patients with MDD (MDD-A, n=20; MDD-R, n=34). No nominal associations with P < 0.05 were observed in the MDD-A subgroup.

Abbreviations: MDD, major depressive disorder; MDD-R, major depressive disorder with retardation; LP, lateral posterior nucleus; VL, ventral lateral nucleus; VPL, ventral posterolateral nucleus; IL, intralaminar nuclei; MDm, medial mediodorsal nucleus; MDl, lateral mediodorsal nucleus; PuM, medial pulvinar; PuL, lateral pulvinar; HAMA, Hamilton Anxiety Rating Scale; GMV, gray matter volume.

Discussion

This multisite structural MRI study compared total and subregional thalamic GMV among patients with MDD-A, patients with MDD-R, and HCs, and examined associations with psychomotor symptoms, overall depressive severity, and anxiety symptoms. After ComBat harmonization, the main group effects were confined to the bilateral ventral anterior nucleus (VA), with greater VA GMV in MDD-R than in HCs. MDD-A did not differ from HCs after FDR correction. Direct comparison between MDD-A and MDD-R showed no FDR-corrected thalamic difference; only the left inferior pulvinar showed a nominal difference. Exploratory analyses identified associations between different thalamic regions and psychomotor and depressive symptom dimensions, while anxiety symptoms were associated with a broader set of thalamic measures. Taken together, the findings support a role for thalamic structure in the clinical heterogeneity of MDD but do not support thalamic GMV as a subtype-specific biomarker.

The clearest group-level finding was greater bilateral VA GMV in MDD-R. The VA is a component of motor thalamic circuitry and receives prominent input from the basal ganglia, with projections to premotor and other frontal motor-related cortical regions. Together with neighboring motor thalamic territories, it participates in circuits involved in movement preparation and initiation.17,18 This anatomical organization provides a plausible link between VA morphology and reduced drive, difficulty initiating behavior, and motor slowing in MDD-R. The direction of the effect is noteworthy. Although many structural imaging studies of MDD have reported regional gray matter reductions,38 increases in GMV have also been described, including increased bilateral thalamic GMV in first-episode, medication-naive patients.39 Increased volume should not be interpreted as preserved or enhanced function. It may reflect disease-related structural adaptation, but other explanations, including illness stage, treatment exposure, and individual differences, remain possible. Structural MRI studies indicate that regional gray matter measures can change with experience and treatment.40,41 Because the present study is cross-sectional, the observed increase cannot be attributed to compensation, vulnerability, or a secondary disease-related process.

In MDD-A, HAMD (adj) was negatively associated with right anteroventral nucleus (AV) GMV (Figure 2). This finding suggests that thalamic morphology may relate not only to psychomotor phenotype but also to the burden of non-psychomotor depressive symptoms. The dimensional association does not mirror the categorical VA group effect, which argues against a simple linear interpretation of thalamic volume. Different thalamic nuclei may relate to different components of the clinical phenotype.

Direct structural differences between MDD-A and MDD-R were limited. Only the left inferior pulvinar showed a nominal difference, and this finding did not survive FDR correction. The present data therefore do not support a thalamic subregion as a reliable structural marker for distinguishing the two psychomotor subtypes. The MDD-A group was smaller (n=54) than the MDD-R (n=123) and HC (n=122) groups, which may have reduced power to detect small or moderate effects. The absence of a significant MDD-A finding should therefore be interpreted as a null result under the current sample and threshold, not as evidence that thalamic structure is normal in MDD-A.

The exploratory correlations suggest that thalamic structure may track continuous symptom dimensions as well as categorical subtypes. In the full MDD sample, HAMD-8 psychomotor retardation was positively associated with right Re GMV, while HAMD-9 psychomotor agitation was positively associated with left VPL GMV. The Re is a key relay within prefrontal–hippocampal circuitry, and experimental work has implicated this nucleus in stress- and depression-related behavioral regulation.42 Right LGN GMV was associated with psychomotor scores in both subtype analyses, and HAMD (adj) was negatively associated with right LGN GMV in the full MDD sample (Figure 2). Although the LGN is classically regarded as a visual relay nucleus, its activity is modulated by attention, arousal and behavioral state, as well as by corticothalamic feedback.16,19,43 Previous work has also suggested that LGN activity can be influenced by motor planning.44 These properties provide a possible framework for interpreting the observed LGN associations in relation to sensory processing, attention, and response preparation. However, these correlations were exploratory and were not corrected for multiple comparisons, and therefore require replication in independent samples.

Anxiety symptoms also showed positive associations with several thalamic measures. Several of these nominal associations involved mediodorsal thalamic subdivisions, which are strongly interconnected with prefrontal and limbic regions and contribute to cognitive and affective regulation.45 Because HAMA data were available only in a subset of patients and these analyses were not corrected for multiple comparisons, the findings do not establish a specific anxiety-related thalamic mechanism (Table 3). They do, however, support the broader possibility that thalamic morphology is related to multiple clinical dimensions in MDD rather than to psychomotor subtype alone.

The current findings do not support the use of thalamic GMV for individual subtype diagnosis or treatment selection. The bilateral VA and LGN may serve as candidate nodes for future studies of thalamocortical circuits, but this will require replication with functional connectivity, structural connectivity, objective psychomotor measures, and longitudinal treatment data. Multimodal imaging may help determine whether thalamic structure adds predictive value when combined with clinical measures. Future studies could also examine whether thalamocortical network features involving these regions have relevance for network-informed neuromodulation strategies, including transcranial magnetic stimulation, although the present structural findings do not identify a stimulation target.

Several limitations should be considered. First, the cross-sectional design precludes causal inference about thalamic structural changes. Second, although ComBat reduced site effects, residual differences in scanners, acquisition parameters, and clinical procedures may remain. Third, the database lacked complete information on medication exposure, number of prior episodes, and some clinical variables, so treatment- and illness-related effects could not be fully controlled. Fourth, psychomotor subtypes were defined using HAMD Items 8 and 9. Single-item ratings may not capture the full range of psychomotor disturbance, and the strict mutually exclusive subtype criteria may restrict symptom variability.6 Fifth, the MDD-A group was relatively small, and HAMA data were available only for a subset of participants, limiting the stability of subgroup and anxiety-related analyses. Finally, although thalamic volumes were extracted before spatial smoothing, small thalamic nuclei remain susceptible to partial-volume effects and registration error.

In conclusion, patients with MDD-R showed greater bilateral VA GMV than HCs, whereas no direct thalamic difference between MDD-A and MDD-R survived FDR correction. Exploratory analyses further linked the AV, LGN, and other thalamic regions to depressive, psychomotor, and anxiety symptom dimensions. These findings suggest that thalamic structure may reflect multidimensional clinical heterogeneity within MDD rather than a biomarker specific to a single psychomotor subtype.

Acknowledgments

The authors thank all members of the REST-meta-MDD consortium who contributed clinical and neuroimaging data to this study. The authors are also grateful to all participants and staff at the participating research centers.

Funding Statement

This work was supported by the Academic Research Project of Beijing Union University (ZK10202301). The funder had no role in the design or conduct of the study; the collection, management, analysis, or interpretation of the data; the preparation, review, or approval of the manuscript; or the decision to submit the manuscript for publication.

Data Sharing Statement

The data used in this study are governed by the REST-meta-MDD consortium/Psychological Science Data Bank and cannot be directly distributed by the authors. Access to the data is subject to the application and approval procedures of the data provider. Correspondence regarding data availability may be addressed to Zheng Lu (LuZheng2501@163.com).

Ethics Approval and Consent to Participate

This study was a secondary analysis of de-identified data from the REST-meta-MDD database. Access to the REST-meta-MDD database was approved by the REST-meta-MDD consortium/Psychological Science Data Bank, with data-use approval ID U0434. The data were used in accordance with the REST-meta-MDD Data Use Agreement.

The present secondary analysis was approved by the Ethics Committee of Shanghai Tongji Hospital (approval No. 2022-058). The REST-meta-MDD database was assembled from previously collected datasets contributed by multiple participating hospitals and research centers in China. According to the REST-meta-MDD/DIRECT project information and previous publications based on this database, the original studies were approved by the local institutional review boards or ethics committees of the participating institutions, and written informed consent was obtained from all participants.

The present study did not involve new participant recruitment, new data collection, or access to identifiable personal information. All procedures involving human participants were conducted in accordance with the Declaration of Helsinki and its later amendments.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors declare that they have no competing interests.

References

  • 1.World Health Organization. World Health Statistics 2024: Monitoring Health for the SDGs, Sustainable Development Goals. Geneva: World Health Organization; 2024. [Google Scholar]
  • 2.Gbd DAIC, Lim SS, Abbafati C, et al. Global burden of 369 diseases and injuries in 204 countries and territories, 1990-2019. A systematic analysis for the global burden of disease study 2019. Lancet. 2020;396(10258):1204–12. doi: 10.1016/S0140-6736(20)30925-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Monroe SM, Harkness KL. Major depression and its recurrences: life course matters. Annu Rev Clin Psychol. 2022;18(1):329–357. doi: 10.1146/annurev-clinpsy-072220-021440 [DOI] [PubMed] [Google Scholar]
  • 4.Dunlop K, Grosenick L, Downar J, et al. Dimensional and categorical solutions to parsing depression heterogeneity in a large single-site sample. Biol Psychiatry. 2024;96(6):422–434. doi: 10.1016/j.biopsych.2024.01.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Sobin C, Sackeim HA. Psychomotor symptoms of depression. Am J Psychiatry. 1997;154:4–17. [DOI] [PubMed] [Google Scholar]
  • 6.Buyukdura JS, Mcclintock SM, Croarkin PE. Psychomotor retardation in depression: biological underpinnings, measurement, and treatment. Prog Neuropsychopharmacol Biol Psychiatry. 2011;35(2):395–409. doi: 10.1016/j.pnpbp.2010.10.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Sampogna G, Del Vecchio V, Giallonardo V, Luciano M, Fiorillo A. Diagnosis, clinical features, and therapeutic implications of agitated depression. The Psychiatric Clinics N Am. 2020;43:47–57. doi: 10.1016/j.psc.2019.10.011 [DOI] [PubMed] [Google Scholar]
  • 8.Rogers ML, Ringer FB, Joiner TE. A meta-analytic review of the association between agitation and suicide attempts. Clin Psychol Rev. 2016;48:1–6. doi: 10.1016/j.cpr.2016.06.002 [DOI] [PubMed] [Google Scholar]
  • 9.Singh AB, Bousman CA, Ng CH, Byron K, Berk M. Psychomotor depressive symptoms may differentially respond to venlafaxine. Int Clin Psychopharmacol. 2013;28(3):121–126. doi: 10.1097/YIC.0b013e32835f1b9f [DOI] [PubMed] [Google Scholar]
  • 10.Jelovac A, Landau S, Beeley P, et al. Why is electroconvulsive therapy for depression more effective in older age? A causal mediation analysis. Psychol Med. 2025;55:e110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Wüthrich F, Lefebvre S, Mittal VA, et al. The neural signature of psychomotor disturbance in depression. Mol Psychiatry. 2023;29(2):317–326. doi: 10.1038/s41380-023-02327-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Shine JM, Lewis LD, Garrett DD, Hwang K. The impact of the human thalamus on brain-wide information processing. Nat Rev Neurosci. 2023;24:416–430. doi: 10.1038/s41583-023-00701-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Biesbroek JM, Verhagen MG, van der Stigchel S, Biessels GJ. When the central integrator disintegrates: a review of the role of the thalamus in cognition and dementia. Alzheimers Dement. 2023;20(3):2209–2222. doi: 10.1002/alz.13563 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Rikhye RV, Wimmer RD, Halassa MM. Toward an integrative theory of thalamic function. Annu Rev Neurosci. 2018;41(1):163–183. doi: 10.1146/annurev-neuro-080317-062144 [DOI] [PubMed] [Google Scholar]
  • 15.Sherman SM, Guillery RW. The role of the thalamus in the flow of information to the cortex. Philos Trans R Soc Lond B Biol Sci. 2002;357(1428):1695–1708. doi: 10.1098/rstb.2002.1161 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Saalmann YB, Kastner S. Gain control in the visual thalamus during perception and cognition. Curr Opin Neurobiol. 2009;19(4):408–414. doi: 10.1016/j.conb.2009.05.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Haber SN, Calzavara R. The cortico-basal ganglia integrative network: the role of the thalamus. Brain Res Bull. 2009;78(2–3):69–74. doi: 10.1016/j.brainresbull.2008.09.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Bosch-Bouju C, Hyland BI, Parr-Brownlie LC. Motor thalamus integration of cortical, cerebellar and basal ganglia information: implications for normal and Parkinsonian conditions. Front Comput Neurosci. 2013;7:163. doi: 10.3389/fncom.2013.00163 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Usrey WM, Alitto HJ. Visual functions of the thalamus. Annu Rev Vis Sci. 2015;1(1):351–371. doi: 10.1146/annurev-vision-082114-035920 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Price JL, Drevets WC. Neurocircuitry of mood disorders. Neuropsychopharmacol. 2010;35:192–216. doi: 10.1038/npp.2009.104 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Bora E, Harrison BJ, Davey CG, Yücel M, Pantelis C. Meta-analysis of volumetric abnormalities in cortico-striatal-pallidal-thalamic circuits in major depressive disorder. Psychol Med. 2012;42(4):671–681. doi: 10.1017/S0033291711001668 [DOI] [PubMed] [Google Scholar]
  • 22.Hong W, Li M, Liu Z, et al. Heterogeneous alterations in thalamic subfields in major depression disorder. J Affect Disord. 2021;295:1079–1086. [DOI] [PubMed] [Google Scholar]
  • 23.Nugent AC, Davis RM, Zarate CAJ, Drevets WC. Reduced thalamic volumes in major depressive disorder. Psychiatry Res. 2013;213(3):179–185. doi: 10.1016/j.pscychresns.2013.05.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Kong QM, Qiao H, Liu CZ, et al. Aberrant intrinsic functional connectivity in thalamo-cortical networks in major depressive disorder. CNS Neurosci Ther. 2018;24(11):1063–1072. doi: 10.1111/cns.12831 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Yan CG, Chen X, Li L, et al. Reduced default mode network functional connectivity in patients with recurrent major depressive disorder. Proc Natl Acad Sci U S A. 2019;116(18):9078–9083. doi: 10.1073/pnas.1900390116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Wang K, Hu Y, Yan C, et al. Brain structural abnormalities in adult major depressive disorder revealed by voxel- and source-based morphometry: evidence from the REST-meta-MDD consortium. Psychol Med. 2023;53(8):3672–3682. doi: 10.1017/S0033291722000320 [DOI] [PubMed] [Google Scholar]
  • 27.Hamilton M. A rating scale for depression. J Neurol, Neurosurg Psychiatry. 1960;23(1):56–62. doi: 10.1136/jnnp.23.1.56 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Sun XY, Li YX, Yu CQ, Li LM. Reliability and validity of depression scales of Chinese version: a systematic review. Zhonghua Liu Xing Bing Xue Za Zhi. 2017;38(1):110–116. doi: 10.3760/cma.j.issn.0254-6450.2017.01.021 [DOI] [PubMed] [Google Scholar]
  • 29.Janzing JGE, Birkenhäger TK, van den Broek WW, Breteler LMT, Nolen WA, Verkes RJ. Psychomotor retardation and the prognosis of antidepressant treatment in patients with unipolar psychotic depression. J Psychiatr Res. 2020;130:321–326. doi: 10.1016/j.jpsychires.2020.07.020 [DOI] [PubMed] [Google Scholar]
  • 30.Lane MF. Acute antidepressant response to fluoxetine and sertraline in psychiatric outpatients with psychomotor agitation. Int J Psychiatry Clin Pract. 2001;5(2):103–109. doi: 10.1080/136515001300374830 [DOI] [PubMed] [Google Scholar]
  • 31.Yan CG, Zang YF. DPARSF: a MATLAB toolbox for “pipeline” data analysis of resting-state fMRI. Front Syst Neurosci. 2010;4:13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Ashburner J, Friston KJ. Unified segmentation. Neuroimage. 2005;26:839–851. doi: 10.1016/j.neuroimage.2005.02.018 [DOI] [PubMed] [Google Scholar]
  • 33.Long JY, Qin K, Pan N, Fan WL, Li Y. Impaired topology and connectivity of grey matter structural networks in major depressive disorder: evidence from a multi-site neuroimaging data-set. Br J Psychiatry. 2024;224(5):170–178. doi: 10.1192/bjp.2024.41 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Ashburner J. A fast diffeomorphic image registration algorithm. Neuroimage. 2007;38:95–113. doi: 10.1016/j.neuroimage.2007.07.007 [DOI] [PubMed] [Google Scholar]
  • 35.Rolls ET, Huang CC, Lin CP, Feng J, Joliot M. Automated anatomical labelling atlas 3. Neuroimage. 2020;206:116189. doi: 10.1016/j.neuroimage.2019.116189 [DOI] [PubMed] [Google Scholar]
  • 36.Johnson WE, Li C, Rabinovic A. Adjusting batch effects in microarray expression data using empirical bayes methods. Biostatistics. 2007;8:118–127. doi: 10.1093/biostatistics/kxj037 [DOI] [PubMed] [Google Scholar]
  • 37.Fortin JP, Cullen N, Sheline YI, et al. Harmonization of cortical thickness measurements across scanners and sites. Neuroimage. 2018;167:104–120. doi: 10.1016/j.neuroimage.2017.11.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Koolschijn PCMP, van Haren NEM, Lensvelt-Mulders GJLM, Hulshoff Pol HE, Kahn RS. Brain volume abnormalities in major depressive disorder: a meta-analysis of magnetic resonance imaging studies. Hum Brain Mapp. 2009;30(11):3719–3735. doi: 10.1002/hbm.20801 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Peng W, Chen Z, Yin L, Jia Z, Gong Q. Essential brain structural alterations in major depressive disorder: a voxel-wise meta-analysis on first episode, medication-naive patients. J Affect Disord. 2016;199:114–123. doi: 10.1016/j.jad.2016.04.001 [DOI] [PubMed] [Google Scholar]
  • 40.May A. Experience-dependent structural plasticity in the adult human brain. Trends Cogn Sci. 2011;15(10):475–482. doi: 10.1016/j.tics.2011.08.002 [DOI] [PubMed] [Google Scholar]
  • 41.Sartorius A, Demirakca T, Böhringer A, et al. Electroconvulsive therapy increases temporal gray matter volume and cortical thickness. Eur Neuropsychopharmacol. 2016;26(3):506–517. doi: 10.1016/j.euroneuro.2015.12.036 [DOI] [PubMed] [Google Scholar]
  • 42.Kafetzopoulos V, Kokras N, Sotiropoulos I, et al. The nucleus reuniens: a key node in the neurocircuitry of stress and depression. Mol Psychiatry. 2018;23(3):579–586. doi: 10.1038/mp.2017.55 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.O’connor DH, Fukui MM, Pinsk MA, Kastner S. Attention modulates responses in the human lateral geniculate nucleus. Nat Neurosci. 2002;5:1203–1209. doi: 10.1038/nn957 [DOI] [PubMed] [Google Scholar]
  • 44.Casagrande VA, Sáry G, Royal D, Ruiz O. On the impact of attention and motor planning on the lateral geniculate nucleus. Prog Brain Res. 2005;149:11–29. [DOI] [PubMed] [Google Scholar]
  • 45.Georgescu IA, Popa D, Zagrean L. The anatomical and functional heterogeneity of the mediodorsal thalamus. Brain Sci. 2020;10(9):624. doi: 10.3390/brainsci10090624 [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.

Data Availability Statement

The data used in this study are governed by the REST-meta-MDD consortium/Psychological Science Data Bank and cannot be directly distributed by the authors. Access to the data is subject to the application and approval procedures of the data provider. Correspondence regarding data availability may be addressed to Zheng Lu (LuZheng2501@163.com).

Articles from Neuropsychiatric Disease and Treatment are provided here courtesy of Dove Press

RESOURCES