People with diabetes have increased risk of depression that further worsens clinical outcome. However, occurrence of depression among patients with diabetes exposed to antidiabetes agents is under-studied. We investigated whether antidiabetes agents would decrease the risk of depression in people with diabetes.

This population-based cohort study identified 686,522 patients with incident type-2 diabetes between 2002-2021 in Hong-Kong who were exposed to any antidiabetes agents, using territory-wide medical-record database of public-healthcare services. Associations of exposure to antidiabetes agents and risk of first-diagnosed depression were examined by Cox proportional-hazards models for each individual agent. An array of covariates, including age, sex, calendar-year period, catchment-area, pre-existing physical/psychiatric-comorbidities, average HbA1c-level, diabetic-complications, cardiovascular/lipid-lowering medications, and presence of other antidiabetes agents under investigation was adjusted. Three sets of sensitivity-analyses were conducted by (a) restricting to patients with cumulative drug-exposure ≥90 days and ≥180 days, (b) monotherapy, and (c) incorporating commonly-used non-antidiabetic medications as exposure variables.

Lower risk of new-onset depression was associated with exposure to any antidiabetes agents (HR 0.42, 95%CI 0.39-0.45) compared with no exposure in patients with incident type-2 diabetes. Lower depression risk was associated with exposure to metformin (0.61[0.57-0.66]) compared with no antidiabetes agents, and to pioglitazone (0.25[0.10-0.63]), linagliptin (0.26[0.13-0.52]), and insulin (0.63[0.55-0.72]). Sensitivity-analyses affirmed that lower depression risk was associated with metformin and insulin.

Patients with type-2 diabetes with exposure to several antidiabetes agents is at decreased risk of new-onset depression. Further research is warranted to clarify the depression-reducing mechanisms of antidiabetes agents in this vulnerable population.

Negative symptoms are a core feature of psychosis and a strong predictor of functional outcome, yet they remain difficult to assess due to conceptual and methodological challenges. Although abnormalities in emotional expressivity and emotional reactivity are documented in individuals at clinical high-risk (CHR) for psychosis, these domains are typically examined independently, and their relationship remains unclear.

Facial expressions were quantified using automated facial analysis (FaceReader) during clinical interviews in 101 CHR individuals and 41 healthy controls (HCs). Emotional reactivity was assessed using the International Affective Picture System (IAPS). Principal component analyses were conducted on facial expression and emotional reactivity variables within the CHR group. Associations with negative symptom domains, positive symptoms, and social functioning were examined using correlational and two-step regression analyses.

CHR participants showed greater disgust expression than HCs (g = 0.40, uncorrected p = .0025, FDR-corrected p = .023). Facial expression and emotional reactivity components showed minimal associations (p > .20). Reduced high-arousal facial expressions were associated with greater emotional expressivity deficits (r = −.22, p = .027), whereas greater happy facial expression was associated with more motivation and pleasure impairment (r = .21, p = .044). Happy facial expression explained additional variance in motivation symptoms beyond emotional reactivity (ΔR2 = .089, p = .008).

Automated facial expression captured variance in some negative symptom domains that was largely independent of emotional reactivity. These findings support the use of multimodal, objective assessments to improve characterization of negative symptoms in psychosis risk.

Adverse childhood experiences (ACEs) are associated with increased long-term mental and physical health risk, yet inter-individual variation indicates that resilience is possible. Distinct trajectories of sustained mental well-being have been observed among individuals exposed to ACEs, predicting favorable long-term outcomes. However, the biopsychosocial markers underpinning sustained trajectories of resilience remain poorly understood. This study aimed to identify factors that predict higher well-being (‘resilient’) trajectories over time in individuals with and without ACE exposure.

Previously identified 12-year well-being trajectory classes (higher vs. lower well-being) were examined in participants with no ACE history (n = 779) and those exposed to ACEs (n = 889). Logistic regression models were used to identify factors differentiating higher versus lower well-being trajectories in the non-ACE sample, and resilience versus risk trajectories in the ACE sample. Predictors included childhood adversity characteristics, genetic predisposition, demographic and environmental factors, social and occupational factors, health and lifestyle factors, and psychological functioning.

Across both samples, higher well-being was predicted by higher education, better subjective physical health, greater social engagement, stronger work performance, and adaptive personality traits (lower neuroticism and higher extraversion). In the non-ACE sample, polygenic well-being scores and family history of mental illness further differentiated trajectories. In contrast, resilience following ACEs was additionally characterized by modifiable factors, including parenting style, relationship status, BMI, and conscientiousness.

Resilience in the context of childhood adversity is defined by identifiable, largely modifiable social, health, and personality factors, highlighting potential targets for interventions to enhance long-term well-being.

Functional connectivity (FC) is consistently altered in patients with schizophrenia. The brain’s primary inhibitory neurotransmitter, gamma-aminobutyric acid (GABA), and its relationship to FC in psychosis spectrum disorders are under-investigated. The anterior cingulate cortex (ACC) has been implicated in many cognitive functions impaired in psychosis. We hypothesize that the relationships between ACC GABA and FC in key brain networks will be altered in first-episode psychosis (FEP) patients as compared to healthy controls (HC).

We used magnetic resonance spectroscopy (MRS) with a MEGA-PRESS sequence to quantify ACC GABA levels in 67 antipsychotic medication-naïve FEP patients and 110 HC. Resting state functional magnetic resonance imaging (fMRI) was used to assess positive and negative FC within the default mode (DMN), salience (SN), dorsal attention (DAN), and executive control (ECN) networks. We used linear regressions to test GABA–FC relationships in each network between groups.

FEP patients had significantly lower GABA levels compared to HC. We also found several clusters in the ECN, DAN, and DMN where FC differed between groups. Ultimately, we found significant GABA–FC group interactions in two ECN clusters and one SN cluster, where GABA and FC were positively correlated in HC but negatively correlated in FEP.

Our data add to the growing literature supporting GABA’s significant role in psychosis spectrum disorders, especially as it relates to FC in key brain networks. Our findings call for further investigation of the mechanisms underlying altered neurometabolic activity and connectivity in psychosis spectrum disorders.

Antipsychotic medications are recommended for schizophrenia spectrum disorders, yet their long-term effects on functional recovery remain unclear, with conflicting evidence often derived from between-subject comparisons vulnerable to confounding by indication.

We conducted a nationwide register-based cohort study of 65,630 individuals with incident schizophrenia spectrum disorders in Denmark (1998–2023). We modeled antipsychotic exposure against ‘productive engagement’ (employment or education). We used two analytical approaches: (1) within-subject stratified Cox models with time-varying covariates to eliminate time-invariant confounding; and (2) Fine–Gray competing risks models with baseline exposure, accounting for mortality and emigration.

Over 26.9 million person-weeks, the overall productive engagement rate was 48.2%. Integration of hospital pharmacy data revealed 6.1% exposure misclassification in studies relying solely on community records. The primary within-subject analysis revealed significant temporal heterogeneity: medication use was associated with reduced engagement rates in the acute (0–2 years; HR = 0.908) and consolidation phases (2–5 years; HR = 0.946), but reversed to a small positive association in the maintenance phase (5+ years; HR = 1.019). The between-subject Fine–Gray model, which estimates cumulative engagement probabilities, yielded an SHR of 1.002 (95% CI = 0.988–1.015), a population-level average that obscured these phase-specific dynamics.

Antipsychotic pharmacotherapy exerts a time-dependent, biphasic impact on vocational recovery. We identified a window of vulnerability during the post-acute ‘consolidation’ phase (years 2–5) where treatment is associated with a transient reduction in productive engagement, before becoming protective after 5 years. These findings challenge the assumption that symptomatic stability automatically facilitates functional reintegration.

Childhood traumas increase the risk of psychosis and voice-hearing. While trauma profiles have been identified in voice-hearers, pathways linking these to voice-related distress remain unclear. This study examined between-group differences in mediation by psychological and behavioral variables in profile–distress associations, with potential moderation by gender.

This cross-sectional study derived childhood trauma profiles via latent class analysis of Childhood Trauma Questionnaire (CTQ) scores from 266 voice-hearing Challenge trial participants with schizophrenia-spectrum diagnoses. Mediation analyses (structural equation modeling with bootstrapped 95% confidence intervals for indirect effects) tested between-group differences in indirect effects of negative voice content, persecutory beliefs about voices, voice power, voice relating style, negative self/other beliefs, emotion regulation, depression, and sleep disturbances in the association between childhood trauma profiles (exposure) and voice-related distress (outcome), with gender as a moderator. Hypotheses were preregistered on the Open Science Framework. Reporting followed AGReMA-guidelines.

Three childhood trauma profiles were established: (1) ‘variable severity’ (n = 160), (2) ‘severe neglect and emotional abuse’ (n = 84), and (3) ‘severe poly-trauma’ (n = 22). Significant between-group differences in indirect effects were observed for persecutory beliefs about voices (1 < 3), voice power (1 < 3), and sleep disturbances (1 < 2). Age-adjustment revealed a between-group difference in indirect effect of negative self-beliefs (1 < 3). No moderation by gender was found.

This is the first investigation of mediators and moderators of childhood trauma profiles and voice-related distress in clinical voice-hearers. Findings suggest that trauma profiles may provide indicators of mediators potentially relevant to inform individualized formulation and therapy planning.

Gaming disorder (GD) is increasingly recognized for its adverse impact on social functioning, yet the underlying neural mechanisms remain unclear, particularly during real-time social interactions. This study examined neural differences in the prefrontal cortex (PFC) and temporoparietal junction (TPJ) during cooperation and competition tasks in adolescents and young adults with GD.

Using functional near-infrared spectroscopy (fNIRS), we examined 175 male participants, classified into Gaming Disorder (n = 42), Hazardous Gaming (n = 69), and Healthy Control (n = 64) groups. Participants performed cooperation and competition tasks while activation, functional connectivity, and network topology were assessed within the PFC and TPJ.

In the current study, GD participants exhibited significantly greater impairments in social and emotional functioning compared to the HG and HC groups. The GD group showed increased right TPJ activation during cooperation and decreased activation during competition. Both GD and HG groups demonstrated heightened medial PFC activation and functional connectivity, indicating compensatory engagement. Graph theory analysis revealed disrupted network topology in GD, including reduced nodal efficiency and centrality within the PFC and TPJ. Neural alterations were significantly correlated with clinical measures, including gaming duration and social-emotional deficits.

This study identifies critical neural disruptions underlying social dysfunction in GD, particularly within mentalizing and executive control networks. The neural markers observed in the GD group have important implications for clinical diagnosis and targeted intervention, whereas similar patterns in the HG group highlight opportunities for early detection and preventive strategies.

Anxiety disorders show striking sex differences in prevalence, symptoms, and clinical characteristics, shaping how they manifest and are experienced.

Here, we report the first sex-specific meta-analysis of genome-wide association studies (GWAS) of anxiety, leveraging two of the largest biobank datasets, UK Biobank and All of Us, comprising 85,042 female cases with 196,789 controls and 36,732 male cases with 136,924 controls. Functional annotation, sex-specific polygenic scores (PGS), and genetic correlations were performed to assess genetic differences and functional implications.

In females, 21 lead SNPs were significantly associated with anxiety, compared to five in males. Although the genetic correlation between sexes was high, it was significantly different from one, indicating partially distinct genetic architectures. In addition, both the SNP-based observed and liability-scale heritabilities (assuming a 2:1 female-to-male prevalence ratio) were significantly higher in females. Gene-based tests and functional prioritization identified different genes associated with anxiety in females and males. Moreover, genetic correlation analyses revealed stronger associations of female anxiety with attention-deficit/hyperactivity disorder (ADHD) and body mass index (BMI), whereas male anxiety showed stronger correlations with waist-hip-ratio-adjusted BMI.

While the overall genetic architecture of anxiety is largely shared, our findings reveal distinct sex-specific genetic associations and correlations, highlighting the value of analyzing the sexes separately to uncover genetic signals that may be masked in sex-combined samples.

Neuropsychiatric symptoms (NPSs) are prevalent in Alzheimer’s disease (AD), yet their neurobiological etiology remains elusive. We investigated relationships between NPS subsyndromes, plasma neurofilament light chain (NFL), AD-vulnerable brain atrophy, and cognition.

We included 146 participants from a Chinese cohort. NPSs were assessed using the Neuropsychiatric Inventory Questionnaire and clustered into four subsyndromes (hyperactivity, psychosis, affective, apathy), each graded by severity (none, mild, severe). Sequential mediation analyses examined whether NPSs influence cognition through NFL and atrophy. Additionally, 1534 ADNI participants were enrolled to (1) replicate mediation effects; (2) examine longitudinal relationships of NPSs with incident cognitive decline; and (3) evaluate cognitive discrimination of NPSs and NFL and onset hazards of NPS subsyndromes.

In the discovery cohort, global NPSs burden and three subsyndromes (hyperactivity, affective, apathy) were associated with elevated plasma NFL, poorer global cognition and memory, and reduced brain volumes (all P < 0.05). Sequential mediation revealed that plasma NFL and atrophy mediated the cross-sectional NPSs–cognition relationship, replicated in ADNI. Adding NPSs and NFL improved cognitive discrimination (AUC: 0.6754 to 0.8210, P < 0.001). Additionally, apathy and psychosis showed lower onset hazards than affective and hyperactivity (both P < 0.001).

Baseline NPSs were cross-sectionally associated with elevated NFL, brain atrophy, and poorer cognition. Sequential mediation models supported a pathway linking NPSs to cognition via NFL and atrophy, though longitudinal evidence did not fully confirm temporal directionality. These hypothesis-generating findings require prospective validation.

Linguistic abnormalities in schizophrenia (SCZ) span morphological, syntactic, semantic, and discourse levels. Converging cross-linguistic evidence suggests that SCZ may involve semantic narrowing alongside reduced syntactic differentiation, yet how these changes co-occur across linguistic domains and whether they represent core, task-general disturbances remains unclear. We applied a multilevel NLP framework to a large Japanese dataset to identify structurally related linguistic markers of SCZ across elicitation contexts.

Speech from 104 patients with SCZ and 101 healthy controls was collected through semi-structured interviews. Transcripts from free conversation, storytelling, and picture description were analyzed using GiNZA, Word2Vec, TF-IDF, and SentenceBERT to extract 76 morphosyntactic, semantic, and discourse features. Factor analysis identified representative features independent of diagnosis, which were tested using generalized estimating equations and validated with bootstrap and permutation procedures. Cross-task stability was examined to determine core linguistic markers.

In free conversation, reduced Case-particle (Kakujoshi) and Adverb use and increased Mean Pairwise Word Similarity were strongly associated with SCZ (AUC = 0.87, 95% CI: 0.74–0.97). Adverbial, case-particle, and semantic-network measures functioned as cross-task markers.

SCZ involves multidimensional language disturbances characterized by a tripartite linguistic phenotype of diminished morphosyntactic explicitness, semantic narrowing, and reduced modification-based contextual modulation in spontaneous discourse. Extending cross-linguistic evidence, our results indicate that lexical-semantic contraction co-occurs with reduced overt marking of argument relations in Japanese, alongside weakened adverbial elaboration and framing – suggesting convergent, largely task-general dimensions of SCZ language pathology, most evident in free conversation.

Prenatal depression is associated with offspring behavioral problems, but heterogeneity in the strength of this association is not well understood. Maternal vitamin D concentration during pregnancy is important for fetal brain development and may help explain this variability, with potential differences by timing of exposure and maternal race.

Using data from 1,451 mother–child pairs in the Environmental influences on Child Health Outcomes cohort, linear mixed-effects models examined associations between prenatal depressive symptom severity, gestational 25-hydroxy-vitamin D (25[OH]D) concentrations, and internalizing and externalizing behaviors in preschool-aged children. Analyses were stratified by common 25(OH)D deficiency thresholds, prenatal timing, and race.

Prenatal depressive symptom severity was associated with greater child internalizing (β = 0.18, 95% CI = 0.11, 0.25) and externalizing (β = 0.21, 95% CI = 0.14, 0.28) behaviors. Gestational 25(OH)D concentration did not moderate depression effect estimates in adjusted models. In stratified analyses, the association between prenatal depressive symptoms and child externalizing behaviors persisted regardless of 25(OH)D threshold levels, but the association with internalizing behaviors attenuated at 25(OH)D < 20 ng/mL. Timing of 25(OH)D measurement (early/late pregnancy) did not modify relationships. Higher gestational 25(OH)D was associated with fewer externalizing problems among offspring of Black mothers only.

Prenatal depressive symptoms showed robust associations with child behavioral problems, largely independent of gestational 25(OH)D. However, attenuated risk for internalizing behaviors with low vitamin D levels warrants investigation of social–environmental factors.

Post-traumatic stress disorder (PTSD) is a growing health problem whose neurobiology remains incompletely understood. Neuroimaging is useful in probing PTSD-related brain dysfunction, and techniques continue to evolve. Structural–functional coupling (SFC) offers a novel integrated perspective on PTSD neurobiology. We sought to define unique SFC alterations in PTSD and explore their associations with clinical symptoms, brain molecular architecture, and gene expression.

We studied 61 PTSD patients and 62 trauma-exposed non-PTSD controls (TENC) recruited from earthquake survivors. We compared SFC constructed from multimodal MRI data by an eigendecomposition method between the two groups. We explored the spatial correlation of SFC with molecular maps, used partial least squares (PLS) regression to associate them with Allen Human Brain Atlas gene data, and conducted enrichment analysis on the identified genes.

PTSD patients showed significant regional SFC alterations in multiple regions: lower SFC in PTSD versus TENC in the default mode network (DMN), frontoparietal network (FPN), dorsal attention network, sensorimotor network, visual network, and thalamus, and higher SFC in PTSD versus TENC in the DMN, FPN, and ventral attention network. Some changes were correlated with clinical symptom severity. In both groups, the spatial distribution of SFC was similarly correlated with molecular architectures. The second component of the PLS regression genes were linked to PTSD-specific SFC variations, enriched mainly in molecular functions and pathways related to synapses and neurotransmitter signaling.

This study yields new insights into PTSD pathophysiology by connecting macroscale SFC changes with their microscale molecular and transcriptional basis.

Bipolar disorders (BD) rank among the most disabling conditions, affecting millions worldwide. Cognitive impairment in BD is linked with brain changes and worse functional outcomes. Obesity and metabolic syndrome (MetSy) are overrepresented in BD and are associated with more pronounced brain structural and cognitive alterations in the general population. Here, we studied for the first time the contribution of metabolic dysfunction to both brain and cognitive alterations in BD.

We recruited 163 participants (76 individuals with BD, 87 controls). We used principal component analysis (PCA) to derive a composite measure of metabolic health from WHR, BMI, HOMA-IR, HbA1c, TGC, HDL, LDL, systolic, diastolic BP, and to derive a composite cognitive measure from the California Verbal Learning Test and Digit Span. Brain structure was indexed using machine-learning-predicted BrainAGE derived from T1-weighted MRI.

Obesity, hypertension, insulin resistance, and dyslipidemia contributed most strongly to variance in metabolic health in this sample. This MetSy-associated risk cluster predicted higher BrainAGE (β = 0.75 ± 0.29, p = 0.011) and lower cognitive performance (β = −0.19 ± 0.09, p = 0.037), accounting for 30% of the association between BD and higher BrainAGE and 25% of the association between BD and worse cognitive performance. These effects were independent of symptoms, medications, illness course, and duration.

MetSy, particularly obesity, was closely linked with brain/cognitive impairments in BD. While the need for metabolic monitoring should be informed by the diagnosis of BD, screening for MetSy can also help track brain health and cognitive functioning.

Attention-deficit/hyperactivity disorder (ADHD) involves altered neurodevelopment, yet the underlying mechanisms remain elusive. Aperiodic EEG components may reflect neural functions like excitatory/inhibitory balance, but their age-related differences in ADHD and link to default mode network (DMN) dysfunction are unexplored.

We included 110 medication-naïve children/adolescents with ADHD and 100 matched typically developing peers aged 6–14 years. Aperiodic parameters (exponent, offset) were derived, and source-localized alpha-band DMN coherence was computed. The sample was stratified into middle childhood (6–9 years) and early adolescence (10–14 years) subgroups to delineate age-dependent patterns.

ADHD showed globally increased exponent and offset versus controls. Normative age-related decreases were significantly attenuated in ADHD, indicating divergence from typical development. Age-stratified analyses revealed distinct patterns: in middle childhood, increased frontal offset correlated positively with inattention (right hemisphere) and hyperactivity/impulsivity (left hemisphere); in early adolescence, it associated with reduced coherence in two DMN pathways (right mSFG–left hippocampus and left mSFG–right MTG).

Aperiodic activity differences in ADHD are age-dependent. Younger children exhibit focal, symptom-linked frontal abnormalities, whereas adolescents show pervasive network-level dysregulation. Aperiodic measures may capture age-varying ADHD pathophysiology, informing developmentally targeted biomarkers.

Working memory (WM) deficits are frequently observed in patients with insomnia disorder (ID), but their neural basis is unclear. Glymphatic dysfunction and disrupted structural-functional coupling have been implicated, yet they have rarely been examined together, particularly in clinical populations.

We conducted a multimodal MRI study in 391 ID patients. Glymphatic function was estimated using the diffusion tensor image analysis along the perivascular space (DTI-ALPS). The SFC was derived by correlating structural connectivity and functional connectivity. WM was measured by the longest span on the digit span backward task. Partial correlations and mediation analyses were performed to examine associations among sleep quality (Pittsburgh Sleep Quality Index, PSQI), DTI-ALPS, SFC, and WM performance.

DTI-ALPS was negatively correlated with PSQI (r = −0.17, p = 0.006), indicating reduced glymphatic clearance with poorer sleep quality. Global SFC was positively associated with DTI-ALPS (r = 0.32, pFDR < 0.001), but not with WM (r = 0.01, p = 0.84). At the network level, SFC within the subcortical network (Sub-SFC) correlated with both DTI-ALPS (r = 0.29, pFDR < 0.001) and WM performance (r = 0.28, pFDR < 0.001). Mediation analysis revealed that DTI-ALPS and Sub-SFC jointly mediated the association between PSQI and WM performance, with a significant indirect effect (indirect effect = −0.074).

This study provides novel evidence that impaired glymphatic clearance and reduced Sub-SFC form key neural pathways linking poor sleep quality to working-memory deficits in ID, and that DTI-ALPS and Sub-SFC may serve as useful biomarkers of cognitive vulnerability.

Schizophrenia (SCZ) and bipolar disorder (BD) are severe psychiatric conditions with overlapping clinical presentations, genetic risk factors, and brain network dysfunction. Whether alterations in large-scale intrinsic brain networks reflect shared or disorder-specific genetic influences remains poorly understood. Clarifying this distinction is essential for refining etiological models and improving diagnostic precision.

Genome-wide inferred statistics (GWIS) were applied to decompose the genetic architecture of SCZ and BD into shared and unique components. Using resting-state network (RSN) data from the UK Biobank, functional connectivity (FC) and structural connectivity (SC) were extracted as neuroimaging phenotypes. Causal inference approaches were subsequently employed to infer potential directional relationships between brain network connectivity and each disorder.

Analyses revealed both common and distinct patterns of brain network connectivity associated with SCZ and BD. Notably, SC within the default mode network (DMN) exhibited opposing effects across the two disorders, suggesting divergent structural underpinnings despite clinical overlap. Additionally, SC within the limbic network (LN) and frontotemporal control network demonstrated potential causal relationships with both conditions, implicating these circuits astransdiagnostic neural substrates.

These findings illuminate the shared and disorder-specific genetic and neural architecture underlying SCZ and BD. Integrating genome-wide genetic methods with large-scale neuroimaging data offers a powerful framework for disentangling psychiatric comorbidity and may inform more targeted diagnostic criteria and individualized treatment strategies.