Dynamic Brain Age Modeling Identifies Network-Specific Cognitive Deficits in Schizophrenia

Dynamic Brain Age Modeling Identifies Network-Specific Cognitive Deficits in Schizophrenia | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Dynamic Brain Age Modeling Identifies Network-Specific Cognitive Deficits in Schizophrenia Mohammad Sendi, Sabrina Edwards-Swart, Bradley Baker, Daniel Mathalon, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7336363/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Schizophrenia is characterized by deficits in attention and working memory. The brain age gap (BAG), the difference between brain-predicted and chronological age, has emerged as a biomarker of brain dysfunction, but its association with dynamic brain function remains unclear. We developed brain age models using static (sFNC) and dynamic (dFNC) functional network connectivity from a large resting-state fMRI dataset ( N = 22,569; UK Biobank, HCP-Young Adult, HCP-Aging) and validated them in an independent schizophrenia cohort (FBIRN; N = 153). Higher BAGs were significantly associated with lower attention and working memory performance ( FDR p < 0.01 ), with dFNC-based models showing more potent effects than sFNC. Network-specific BAGs, particularly within cognitive control, default mode, and subcortical networks, were robust predictors of cognitive impairment. These findings establish dFNC-based BAG as a sensitive biomarker of cognitive dysfunction in schizophrenia and highlight the value of dynamic connectivity measures for advancing precision diagnostics and stratification. Health sciences/Biomarkers/Diagnostic markers Biological sciences/Computational biology and bioinformatics/Machine learning Biological sciences/Computational biology and bioinformatics/Computational models Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Schizophrenia, affecting 0.25% to 0.64% of the U.S. population, is a complex psychiatric disorder marked by profound disruptions in cognition, emotion, and social interactions 1 , 2 . Key symptoms include hallucinations, delusions, and motor impairments, alongside significant reductions in emotional expression and social engagement. Critically, individuals with schizophrenia exhibit marked cognitive deficits, particularly in working memory and attention vigilance. Studies indicate they perform approximately 2.5 standard deviations below healthy controls in these cognitive domains 3 . These deficits are characterized by difficulties in retaining and updating information in working memory 4 , with increasing cognitive demands leading to more pronounced declines in attention vigilance compared to healthy individuals 5 . In recent decades, neuroimaging techniques have greatly enhanced our understanding of cognitive function and attention deficits in schizophrenia 4 , 6 . Structural magnetic resonance imaging (sMRI) provides detailed anatomical insights, associations between executive function impairments and reductions in prefrontal cortex volume and thickness, and links episodic memory deficits with hippocampal atrophy. Conversely, functional magnetic resonance imaging (fMRI) measures brain activity through oxygen metabolism and blood flow, illuminating functional dynamics across various brain regions 7 . Functional MRI has been pivotal in identifying brain regions involved in working memory, offering valuable insights into cognitive impairments in patient populations compared to healthy controls. Broadly, neuroimaging plays a key role in examining executive function and episodic memory impairments and provides a valuable tool for tracking brain function changes following cognitive remediation therapies 8 . In recent years, brain age prediction from neuroimaging has emerged as a promising approach for identifying novel biomarkers of neurological and neuropsychiatric conditions 9 – 14 . By modeling the relationship between neuroimaging variables and age in healthy individuals, this method enables the detection of meaningful brain deviations in patient populations. The difference between an individual’s predicted brain age and their chronological age, termed the brain age gap (BAG), provides critical insights into the pathology of cognitive function 15 – 18 . For example, emerging evidence demonstrates that individuals with accelerated brain age consistently show marked deficits across critical cognitive domains—including IQ, verbal comprehension, perceptual reasoning, processing speed, working memory, and memory recall—as captured by gold-standard assessments such as the Rey Auditory Verbal Learning Test 17 . In schizophrenia, studies using sMRI have revealed that patients often exhibit a brain age 6 to 8 years older than their chronological age 19 . Further, the BAG was noted to increase across at-risk, recent onset 20 , and recurrent phases of schizophrenia, initially rising approximately a year and a half annually for the first five years before stabilizing 21 . Additionally, schizophrenia and bipolar disorder present distinct BAG profiles. Individuals with schizophrenia consistently exhibit elevated BAGs, whereas those with bipolar disorder maintain levels comparable to healthy controls. This distinction underscores the potential of BAG as a valuable biomarker for enhancing early differential diagnosis 22 . This suggests a progressive pathogenic component unique to schizophrenia, highlighting the utility of BAG as a biomarker 23 . Moreover, a significant negative association between BAG and cognitive functions like working memory has been observed in schizophrenia, highlighting the potential of BAG in understanding and diagnosing declined neurocognitive performance in schizophrenia 24 . While the majority of brain age prediction research has traditionally focused on structural MRI (sMRI), recent advancements include employing functional connectivity metrics from resting-state fMRI to predict brain age 14 , 25 – 27 . A notable study using resting-state fMRI data from the Philadelphia Neurodevelopmental Cohort demonstrated the value of brain age models in youth by linking older estimated brain age to greater symptom burden, especially across DSM-5 psychiatric diagnoses. These findings underscore the potential of brain age to capture deviations in brain maturation associated with mental health conditions 28 . Furthermore, an innovative study used resting-state fMRI and brain age prediction to identify neural connections related to abnormal brain aging. By systematically excluding connections from the training model, the approach pinpointed those most critical to brain age accuracy, providing new insights into the neurobiological mechanisms of age-related psychiatric conditions 9 . Although brain age prediction using static functional connectivity (sFNC), estimated from resting-state fMRI, has proven valuable, the potential of dynamic functional connectivity (dFNC) in this area is still underexplored. Unlike sFNC, which evaluates connectivity from correlations across an entire time series, dFNC analyzes connections between brain regions or networks within specific time intervals. This allows for the capture of temporal fluctuations that reveal how brain network interactions change over time 29 , 30 . The variability and dynamism of neural signals, captured by dFNC, are crucial in understanding cognitive deficits and clinical symptoms in psychiatric disorders. These dynamic measures provide insights into brain function that static approaches cannot, highlighting changes and interactions that are critical in disease progression 31 – 37 . Because disrupted brain network dynamics potentially drive cognitive symptoms in schizophrenia, leveraging a BAG derived from dFNC offers a powerful, next-generation marker of disease-related neurobiological aging. Critically, no study has yet evaluated whether dFNC-based BAG outperforms traditional static FNC (sFNC) BAG in explaining core cognitive deficits such as working memory and attention. Establishing this added predictive power would mark a transformative advance, positioning dFNC BAG as a mechanistically grounded biomarker and unlocking new precision targets for circuit-based interventions in schizophrenia and related psychiatric conditions. Here we address this gap using the most significant sample to date: 22,569 resting-state scans from 17,522 healthy adults from UK Biobank 38 , 39 and Human Connectome Project 40 – 42 to train multiple wide-brain and sub-network brain age prediction models, and an independent cohort of 153 controls and individuals with schizophrenia from the FBIRN consortium to test them 43 . Leveraging a graph-convolutional network for sFNC and a bidirectional long short-term memory or LSTM for dFNC, we generated wide-brain and sub-network BAGs and evaluated their links to attention vigilance and working memory performance while controlling for demographic and clinical covariates. We hypothesize that (1) BAGs derived from dFNC will predict age as accurately as sFNC models, and (2) larger BAGs, particularly those based on dFNC, will associate with poorer cognitive performance, reflecting accelerated dynamic functional brain ageing in schizophrenia. By integrating dynamic connectivity with brain age modeling, this study provides the first direct evidence linking aberrant network dynamics to cognitive impairment. It establishes dFNC-based BAG as a clinically relevant biomarker. Results Study Population Our training dataset comprised 22,569 resting-state fMRI scans from 17,522 healthy individuals aged 22 to 100 years (mean age = 62.42 ± 10.96 years; 8,090 females, 9,432 males), drawn from the UK Biobank (UKBB) 38 , 39 , Human Connectome Project Young Adult (HCP-YA 40 ), and Human Connectome Project Aging (HCP-A 42 ) cohorts. Stringent exclusion criteria were applied to UKBB participants to remove individuals with primary psychiatric or neurological conditions (see Method Section ). The HCP datasets provided approximately 15 minutes of resting-state fMRI per subject, while UKBB contributed ~ 6 minutes per scan. To evaluate generalizability and clinical relevance, we validated our models in an independent clinical dataset (FBIRN; N = 311), including 151 individuals with schizophrenia and 160 healthy controls aged 18 to 62 years (mean age = 37.88 years; 81 females, 230 males). Diagnoses were confirmed using structured clinical interviews. Detailed demographic characteristics are shown in Table 1 , with site-level breakdowns of the FBIRN dataset in Supplementary Table 1 . Also, Fig. 1 A shows the count density of the training datasets, including UKBB, HCP-YA, and HCP-A, and Fig. 1 B displays the count density of the test dataset (i.e., FBIRN) used in our study. Table 1 The Training Data set is composed of UKBB, HCP, and HCP-A datasets, while the FBIRN data set comprises the Test Data. N Age (M ± SD) Sex (M/F) All Training Data 17522 62.42 ± 10.96 9432/8090 UKBB 15978 64.26 ± 7.55 8734 / 7244 Training HCP 833 28.66 ± 3.66 390 / 443 HCP-A 711 60.54 ± 15.68 308 / 403 Test FBIRN 311 37.88 ± 11.28 230 / 81 Training and validating the brain age prediction model on the healthy population We extracted reliable brain networks using the NeuroMark automated ICA pipeline 44 , which applies spatially constrained templates derived from large normative datasets. This method identified 53 independent components, organized into seven functional networks. These seven networks include subcortical (SCN), auditory (AUD), sensorimotor (SMN), visual (VSN), cognitive control (CCN), default-mode (DMN), and cerebellar networks (CBN) as shown in Supplementary Fig. 1 and Supplementary Table 2 . We computed sFNC by calculating pairwise Pearson correlations between IC time courses, yielding 1,378 connectivity features per subject (see Fig. 1 C). To capture temporal fluctuations, we estimated dFNC using a sliding window approach (see Supplementary Fig. 2 ). After developing brain age prediction models using wide-brain sFNC (as shown in Fig. 1 D and Supplementary Fig. 3A ) and dFNC (as shown in Fig. 1 E and Supplementary Fig. 3B ), we validated their performance on a validation dataset. This step was essential to assess the models’ generalizability and accuracy in predicting brain age. We calculated the Pearson correlation coefficients between the chronological brain age and the expected brain age to evaluate the models’ performance. Results demonstrated a strong correlation between the predicted brain age from the sFNC-based model and chronological age, with a correlation value of 0.8755 (Fig. 2 A). Similarly, the dFNC-based model also showed a strong correlation, with a correlation value of 0.8675, illustrating its effectiveness in age prediction (Fig. 2 B). Static and dynamic FNC-based wide-brain age gap links with attention vigilance After developing and validating the brain age prediction models, we applied them to the FBIRN dataset to calculate wide-brain BAGs (wBAGs), representing brain age gaps estimated from wide-brain sFNC and dFNC for everyone. We then constructed generalized linear model (GLMs) to examine the association between wBAGs and attention vigilance while adjusting for age, sex, study site, age 2 , the age-by-sex interaction, and diagnosis as covariates. The corresponding p-values were further adjusted to account for testing two modalities (i.e., sFNC and dFNC features). Because our age prediction model performs less accurately for individuals under 38 years old (see Fig. 2 ), we restricted the FBIRN analysis to participants over 38. This resulted in a final sample of 153 participants (121 males) with a mean age of 47.73 ± 5.94 years. Figure 3 A shows a significant negative association between sFNC-based wBAGs and attention vigilance ( r =-0.2923, β = -0.9860, SE = 0.2829, 95% CI : -1.5454 to -0.4263, FDR p = 0.0013, N = 153). Figure 3 B demonstrates a significant association between dFNC-based wBAGs and attention vigilance ( r = -0.2715, β =-0.4395, S = 0.1366, 95% CI : -0.7099 to -0.1691, FDR p = 0.0016, N = 153). In both models, we observed a negative association between wBAGs and cognitive performance, indicating that individuals with older-appearing brains tend to perform worse on attention vigilance tasks. Static and dynamic FNC-based wide-brain age gap links with working memory We constructed GLMs to examine the association between sFNC- and dFNC-based wBAGs and working memory while adjusting for age, sex, study site, age 2 , age-by-sex interaction, and diagnosis as covariates. P-values were further corrected for multiple comparisons across the two modalities. Figure 3 C shows a significant negative association between sFNC-based wBAGs and working memory ( r =-0.2237, β =-0.9603, SE = 0.3615, 95% CI =-0.2918 to -0.0856, FDR p = 0.0088, N = 153). Figure 3 D demonstrates a significant negative association between dFNC-based wBAGs and working memory ( r =-0.2508, β =-0.5195, SE = 0.1732, 95% CI = -0.8621 to -0.1769, FDR p = 0.0064, N = 153). These negative correlations indicate that individuals with higher brain age relative to chronological age tend to exhibit poorer working memory performance. Sub-network brain age gaps predict attention vigilance and working memory The next question was whether sFNC- and dFNC-based sub-network brain age gaps or subBAGs predict attention vigilance, and working memory performance. To address this, we developed brain age prediction models separately for each network using data from our healthy population. Specifically, we created seven distinct brain age prediction models corresponding to seven networks, including SCN, AUC, SMN, VSN, CCN, DMN, and CBN. The correlation between chronological age and predicted brain age for each sub-network is presented in Supplementary Fig. 4. Next, we examined the association between subBAGs and attention vigilance, and working memory, controlling for age, sex, site, age², age-by-sex interaction, and diagnosis as covariates. Given seven networks and two modalities (sFNC and dFNC), a total of 14 statistical tests were performed, and p-values were corrected accordingly. Figure 4 A illustrates the relationship between sFNC- and dFNC subBAGs and attention vigilance. Each bar represents the association between subBAGs and attention vigilance, with crosshatched bars indicating dFNC-based subBAGs and solid bars representing sFNC-based subBAGs. Bars showing a significant association after FDR correction are marked with double asterisks. As shown, the sFNC-based subBAGs estimated from the subcortical network exhibit a negative association with attention vigilance ( r = -0.2357, β = -0.1813, SE = 0.0655, 95% CI : -0.3111 to -0.0516, FDR p = 0.0228, N = 153). Similarly, the dFNC-based subBAGs from the subcortical network show a significant negative association with attention vigilance ( r =-0.2806, β =-0.2989, SE = 0.0896, 95% CI : -0.4763 to -0.1215, FDR p = 0.0140, N = 153). Additionally, dFNC-based subBAGs estimated from the sensorimotor network show a significant negative association ( r = -0.2384, β =-0.2153, SE = 0.0769, 95% CI : -0.3675 to -0.0631, FDR p = 0.0228, N = 153) and from the default mode network ( r =-0.2666, β =-0.1053, SE = 0.0333, 95% CI : -0.1713 to -0.0392, FDR p = 0.0140, N = 153). Our results also indicate a significant negative association between sFNC-based subBAGs estimated from the cognitive control network and attention vigilance ( r =-0.2081, β =-0.0353, SE = 0.0145, 95% CI : -0.0641 to -0.0065, FDR p = 0.0455, N = 153). Supplementary Table 3 presents the complete set of models examining the association between subBAGs and attention vigilance. Figure 4 B shows the link between sFNC- and dFNC-based subBAGs with working memory using GLM with age, sex, site, age², age-by-sex interaction, and diagnosis as covariates. Among all models, we identified only three showing a significant association between subBAGs and working memory (as shown with single asterisk): sFNC-based subBAGs estimated from the subcortical network ( r =-0.1724, β = -0.1050, SE = 0.0518, 95% CI : -0.2076 to -0.0020, uncorrected p = 0.0447, N = 153), dFNC-based subBAGs estimated from the subcortical network ( r =-0.1754, β =-0.1473, SE = 0.0714, 95% CI : -0.2885 to -0.0060, uncorrected p = 0.0410, N = 153), and dFNC-based subBAGs estimated from the default mode network ( r =-0.2114, β =-0.0662, SE = 0.0264, 95% CI : -0.1186 to -0.0139, uncorrected p = 0.0134, N = 153). However, none of these associations remained significant after FDR correction. Supplementary Table 4 presents the complete set of models examining the association between subBAGs and working memory. Discussion The brain-age gap (BAG), the difference between a person’s predicted brain age and chronological age, has emerged as a compact indicator of neurobiological health: a larger positive BAG signals accelerated ageing, whereas a negative BAG reflects resilience. Schizophrenia reliably shows an elevated BAG, and larger BAGs link with lower working-memory performance 10 – 12 . Crucially, almost all BAG work to date relies on structural MRI or static FNC, overlooking the millisecond-scale reconfiguration of brain networks. We present the first dynamic FNC brain age models—at both wide brain (wBAG) and network-specific (subBAG) levels—leveraging time-resolved connectivity to capture the neural volatility that likely underlies cognitive dysfunction. By testing dFNC-BAG against working memory and attention vigilance performance in the schizophrenia cohort, we aim to uncover a previously inaccessible layer of pathophysiology and position dFNC-BAG as a new precision biomarker capable of guiding circuit-based interventions and accelerating translational psychiatry. Extant research shows that sustained attention declines with normal aging 45 , and schizophrenia magnifies this decline, with patients repeatedly scoring well below normative thresholds across sites, ages, and sexes 46 . Our data reveal a robust negative association between attention vigilance measure and both wBAG and subBAG indices of brain age acceleration, whether derived from static or dynamic FNC. Put simply, the “older” the brain looks, the poorer the patient performs. This powerful link spotlights accelerated macro- and micro-scale brain aging as a driving force behind attentional breakdown in schizophrenia and elevates BAG to a precision and network-level biomarker with clear therapeutic implications. Convergent neuroimaging work has already mapped attentional control to frontoparietal and default mode circuitry; our findings now tie those circuits’ temporal dysfunction directly to real-world cognitive deficits. Neuroimaging studies have localized specific brain regions and connectivity networks that facilitate attentional processes. For instance, processing temporal stimuli has been linked to activity in the pre-supplementary motor area 47 . In contrast, joint attention engages the ventromedial frontal cortex, cingulate cortex, caudate nuclei, and left superior frontal gyrus 48 . Mirroring these maps, our strongest BAG-attention associations localize to sensorimotor and subcortical networks, notably the caudate, thalamus, and putamen, underscoring their central role in the disorder’s cognitive burden. Overall, this work bridges neuroanatomical change with clinical symptomatology and supports BAG as a precision tool for early identification and stratification in neuropsychiatric disorders. Working memory deficits, spanning visuospatial, phonological, and executive domains, are a core cognitive feature of schizophrenia, consistently observed across cohorts relative to healthy controls 49 . These deficits are compounded by normative age-related decline, with working memory performance deteriorating progressively across adulthood 50 . Building on this foundation, we examined the relationship between BAG and working memory using both sFNC and dFNC models. Both sFNC- and dFNC-based wBAG showed robust negative associations with working memory performance in the FBIRN cohort. In other words, greater brain age acceleration predicted worse working memory capacity. This finding extends prior work by linking functional BAG to cognition in schizophrenia, reinforcing the relevance of BAG as a mechanistic marker of cognitive vulnerability. Notably, our results highlight the subcortical and default mode networks as key contributors to this relationship, pointing to specific neural systems where accelerated ageing most profoundly impacts working memory function. Notably, the association between wBAGs and working memory, and subBAGs and attention vigilance, was stronger for dFNC-derived BAG, suggesting cognitive deficits in schizophrenia may be more closely linked to dynamic reconfiguration of brain networks than to static connectivity patterns. This aligns with evidence that cognition depends on flexible, moment-to-moment network adaptations to meet cognitive demands 51 . By capturing temporal fluctuations in connectivity, the dFNC approach provides a richer, more ecologically valid representation of brain function, making it more sensitive to the cognitive disruptions observed in schizophrenia. In contrast, sFNC averages connectivity over time, potentially obscuring critical dynamic alterations. The higher temporal resolution of dFNC also yields more data points per scan, enhancing the detection of subtle brain–behavior associations 32 , 33 , 35 . These findings highlight the unique potential of dFNC-based BAG as a precision biomarker to capture clinically meaningful cognitive decline and inform individualized therapeutic strategies in schizophrenia. Our study has several limitations that merit consideration. Firstly, the choice of window size in dynamic connectivity analysis inherently assumes specific characteristics about temporal dynamics. Shorter windows capture rapid fluctuations effectively, while longer windows tend to smooth these fluctuations, potentially obscuring meaningful variability. Future research should explore a broader range of window sizes to more comprehensively assess their impact on brain connectivity measures 52 . Additionally, our study excluded participants with any health conditions from the UKBB dataset. Future studies may benefit from including a broader cohort by aggregating data from multiple datasets that include healthy individuals, which could enhance the generalizability and robustness of the findings. Moreover, prior research suggests that factors like increased income or engagement in cognitively stimulating activities can mitigate age-related declines in working memory. However, our model did not account for variables such as average activity levels, diet, or other environmental factors that could also influence cognitive and brain function 53 , 54 . Adjusting both brain and cognitive measures for these variables in future studies could enhance our understanding of the complex interactions between lifestyle, health conditions, and cognitive aging, providing a more nuanced perspective on these relationships. Finally, due to deviations observed in our brain-age predictions for younger adults, likely driven by the limited representation of this age group in our healthy training dataset, we excluded FBIRN participants under 38. Expanding brain-age modeling efforts to include more younger adults from diverse datasets is an essential next step, enabling exploration of whether BAG meaningfully captures cognitive deficits earlier in life and across the whole adult lifespan. In conclusion, our study establishes BAG, derived from both static and dynamic functional connectivity at wide-brain and subnetwork levels, as a powerful and clinically relevant marker of cognitive deficits in schizophrenia. We demonstrate that higher BAG, reflecting accelerated brain aging, is significantly associated with worse working memory and attention vigilance, capturing core cognitive dysfunctions of the disorder. Notably, the dynamic approach provides unique sensitivity by incorporating temporal fluctuations in connectivity, offering more profound insight into the brain’s functional organization than traditional static methods. Our findings highlight dFNC-based BAG as especially informative, uncovering stronger links to the neural substrates of cognitive decline. These results not only advance mechanistic understanding but also open new avenues for clinical translation. Future research expanding dynamic modeling and integrating lifestyle and health factors holds tremendous promise for optimizing prediction models. Overall, this work lays the foundation for next-generation precision psychiatry tools capable of early detection and personalized intervention strategies to mitigate cognitive deterioration and improve long-term outcomes in schizophrenia. Methods Inclusion and Ethics Statement The Northwest Multicenter Research Ethics Committee granted ethical approval for the UK Biobank (reference 11/NW/0382), and the present study was conducted under UK Biobank application number 34175. The Human Connectome Project Young Adult (HCP-YA) and Aging (HCP-A) datasets were approved by the Washington University Institutional Review Board, and all participants provided written informed consent, including consent to share de-identified data. The FBIRN study received institutional review board (IRB) approval at each participating site, including the University of California (Irvine, Los Angeles, San Francisco), Duke University/University of North Carolina at Chapel Hill, the University of Iowa, the University of Minnesota, and the University of New Mexico. All participants in the FBIRN study provided written informed consent. Diagnoses for individuals with schizophrenia were confirmed using the Structured Clinical Interview for DSM-IV (SCID), and healthy controls were screened using SCID-I/NP to ensure the absence of psychiatric or neurological disorders. Study Population Our study leveraged multiple datasets, including the UK Biobank (UKBB 38,39 and the Human Connectome Project’s Young Adult (HCP-YA 40 and Aging (HCP-A 42 cohorts, to develop brain age prediction models. For the UKBB dataset, exclusion criteria encompassed a range of mental and behavioral disorders as categorized by the International Classification of Diseases version 10 (ICD-10): delirium not induced by alcohol and other psychoactive substances (F05); mental disorders due to brain damage, dysfunction, or physical disease (F06); personality and behavioral disorders due to brain disease, damage, or dysfunction (F07); unspecified organic or symptomatic mental disorders (F09); disorders due to psychoactive substance use (F10-F19); schizophrenia, schizotypal, and delusional disorders (F20-29); manic episodes (F30); and bipolar affective disorder (F31). We also excluded individuals who had sought treatment for “nerves, anxiety, tension, or depression” from either their general practitioner or a psychiatrist. After exclusions, the UKBB dataset comprised 15,978 samples aged 45 to 82 years, with a mean age of 64.26 ± 7.55 years and a median age of 65 years, including 7,244 females and 8,734 males 38,39 . All UKBB participants provide informed consent as part of the ethical oversight maintained by a dedicated Ethics and Guidance Council, which collaborates with UKBB to uphold an Ethics and Governance Framework. Additionally, the study received IRB approval from the Northwest Multicenter Research Ethics Committee. The HCP-YA dataset included 833 individuals aged 22 to 35 years, with a mean age of 28.66 ± 3.66 years and a median age of 29 years, consisting of 443 females and 390 males 41 . The HCP-A dataset included 711 individuals aged 36 to 100 years, with a mean age of 60.54 ± 15.68 years and a median age of 58.67 years, consisting of 403 females and 308 males 42 . Collectively, the datasets encompassed 17,522 individuals aged 22 to 100 years, with a mean age of 62.42 ± 10.96 years and a median age of 64 years, including 8,090 females and 9,432 males. All subject recruitment procedures and informed consent forms, including consent to share de-identified data, were approved by the Washington University IRB. Notably, the HCP datasets provided around 15 minutes of resting-state fMRI data per session, while the UKBB dataset included around 6 minutes of resting-state fMRI data. Because these datasets contain multiple scans per participant, there are a total of 22,569 distinct resting-state fMRI scans in the training set (see Supplementary Figure 5 ). Table 1 shows the demographic information of both the training and test datasets. Additionally, data from the Functional Imaging Biomedical Informatics Research Network (FBIRN) were employed to estimate the brain age gap (BAG) and examine its link with cognitive metrics such as working memory and attention 43 . For the FBIRN dataset, raw imaging data were collected across seven sites: University of California, Irvine; University of California, Los Angeles; University of California, San Francisco; Duke University/University of North Carolina at Chapel Hill; the University of Iowa; the University of Minnesota; and the University of New Mexico. Each participant provided written informed consent, with protocols approved by the institutional review boards at each respective site. All subjects with schizophrenia (SZ) were clinically stable at the time of scanning. Diagnoses were validated through the Structured Clinical Interview for DSM-IV (SCID-IV), while healthy controls (HC) were assessed using the SCID-I/NP to confirm the absence of schizophrenia. Exclusion criteria for HC included a current or history of major neurological or psychiatric disorders and having a first-degree relative with an Axis-I psychotic disorder, as determined by SCID evaluations. The FBIRN dataset contains individuals between 18 and 62 years old, with a mean age of 37.88 and a median age of 38. Out of 311 individuals in the set, 151 of them have a diagnosis of schizophrenia; these participants are between 18 and 62 years old, with a mean age of 38.77 and a median age of 39. Overall, there were 81 female participants and 230 male participants; among those diagnosed with schizophrenia, there were 36 female participants and 115 male participants. For the FBIRN dataset, written informed consent was obtained from all participants. Institutional review boards approved the consent process of each study site. Table 1 presents the demographic information for all FBIRN participants, while Supplementary Table 1 details the demographic and clinical data for each site separately. Figure 1A illustrates the count density of the training datasets (UKBB, HCP-YA, and HCP-A), while Figure 1B displays the count density of the test dataset (FBIRN) used in our study. It is worth noting that in our research, we only focused on participants older than 38 years, since our brain age prediction model was not very accurate in the younger group (see Supplementary Figure 6 ). Imaging Protocol For UKBB, the imaging data were collected using Siemens Skyra 3T scanners. The resolution was 2.4 × 2.4 × 2.4 mm³. The data was collected as 490 timeframes over 6 minutes, with a TR of 0.735 s, a TE of 39ms, and a flip angle of 52°. During the scan, participants are instructed to fixate on a cross displayed on a screen to minimize eye movement and maximize data consistency. For HCP, the imaging data were collected using a Siemens 3T scanner as 1200 frames per run over 14:33 minutes per run, with four runs total. The TR was 720 ms, the TE was 33.1, the flip angle was 52°, and the slice thickness was 2.0mm. For the HCP-YA, imaging data were captured using a 3T Siemens Prisma scanner equipped with a 32-channel phased-array head coil. The rs-fMRI acquisition parameters included a TR of 720 ms, a TE of 33.1 ms, and a field of view of 208 × 180 mm². A flip angle of 52° was used, with images obtained across 72 oblique-axial slices at a resolution of 2.0 × 2.0 × 2.0 mm³. Each scanning session was conducted for approximately 14 minutes and 40 seconds, and the HCP-A imaging data were collected using a 3T Siemens Prisma scanner equipped with a 32-channel phased-array head coil. The rs-fMRI protocol included the following parameters: TR of 800 ms, TE of 37 ms, a field of view of 810 × 936 mm², and a flip angle of 52°. Images were acquired with a resolution of 2.0 × 2.0 × 2.0 mm³ across 72 oblique-axial slices. Each scanning session lasted approximately 14 minutes and 40 seconds. Similar to the UKBB, participants in the HCP are asked to fixate on a cross presented on a screen during the scan to reduce eye movements and ensure consistent data collection. For FBIRN, imaging data were acquired using six Siemens 3T scanners and one General Electric 3T scanner. All sites followed a uniform rs-fMRI protocol. T2*-weighted functional images were captured using an echo-planar imaging sequence, aligned along the anterior and posterior commissure (AC-PC) line, with the following parameters: TE of 30 ms, TR of 2 s, flip angle of 77°, slice gap of 1 mm, voxel dimensions of 3.4 × 3.4 × 4 mm³, and a series of 162 frames spanning 5 minutes and 38 seconds. During scans, participants were instructed to keep their eyes closed. Data Processing Data from fMRI were preprocessed using Statistical Parametric Mapping (SPM12, https://www.fil.ion.ucl.ac.uk/spm/) in the MATLAB 2019 environment. We conducted a rigid body motion correction with SPM’s toolbox to address head motion. Subsequently, the imaging data were spatially normalized to an echo-planar imaging (EPI) template in standard Montreal Neurological Institute (MNI) space. Finally, a Gaussian kernel with a full width at half maximum (FWHM) of 6 mm was applied to smooth the fMRI images. Extracting independent components using NeuroMark To obtain reliable independent components (ICs), we employed the NeuroMark fully automated ICA pipeline, which integrates previously derived component maps as spatial constraints 44 . The NeuroMark framework utilizes templates developed from substantial datasets, specifically the Human Connectome Project (HCP: https://www.humanconnectome.org/study/hcp-young-adult/document/1200-subjects-data-release, 823 subjects after the subject selection) and the Genomics Superstruct Project (GSP: https://dataverse.harvard.edu/dataverse/GSP, 1005 subjects post-selection). This approach has proven effective across numerous studies, identifying a broad array of imaging markers for various brain disorders. Further information on template development is available in our prior publication on the NeuroMark method 44 . The NeuroMark template includes 53 independent components (ICs), categorized into seven functional networks: subcortical (SCN), auditory (AUD), sensorimotor (SMN), visual (VSN), cognitive control (CCN), default-mode (DMN), and cerebellar networks (CBN), as shown in Figure 1C and Supplementary Figure 1 . Supplementary Table 2 also shows all 53 ICs and their interactions. For the ICA analysis, we utilized these templates via the NeuroMark_fMRI_1.0 template, accessible through GIFT v4.0.5.14 GIFT (http://trendscenter.org/software/gift and on the TReNDS website @ http://trendscenter.org/data). Additional denoising and artifact removal steps prior to calculating dynamic functional connectivity included: 1) linear, quadratic, and cubic de-trending; 2) multiple regression of the six realignment parameters and their temporal derivatives; 3) outlier removal; and 4) low-pass filtering below a frequency of 0.15 Hz. Static functional network connectivity To estimate sFNC, we calculated the Pearson correlation between pairs of ICs in each subject as shown in equation 1 (1) where and are time course signals and and are the mean of and , respectively. This Pearson correlation takes values in the interval [− 1, 1] and measures the strength of the linear relationship between and . Each FNC is a 53×53 matrix, from which we derived a total of =1378 connectivity features. Figure 1C shows a representative FNC we used in our study. Dynamic functional network connectivity To compute dynamic functional network connectivity (dFNC), we employed a sliding window approach. We used a tapered window, formed by convolving a rectangular window with the duration of 20TR (UKBB: 14.7 s, HCP-YA: 14.4 s, HCP-A: 16 s, FBIRN: 40s) with a Gaussian kernel (σ = 3), to precisely focus on data at each time point (see Supplementary Figure 2 ) 32,33,35,36,55–57 . This approach, detailed in Equation 1, was utilized to compute the FNC at each time point. We then aggregated the dFNC estimates for each window and each subject to construct a three-dimensional array (C × C × T), where C represents the 53 independent components, and T represents the number of windows. This array captures the temporal variations in connectivity among the independent components. Training the model In our study, the resting-state fMRI time series were uniformly downsampled to the shortest duration across datasets, corresponding to approximately 5.4 minutes. We then computed dFNC matrices using a sliding window approach with a window size of 40 TRs. To evaluate model stability, K=5-fold cross-validation was performed on the training set of 22,569 scans from the UKB, HCP-YA, and HCP-A data sets. For sFNC, we deployed a brain connectivity graph convolutional network, or BCGCN, to predict the brain age 58,59 . The BCGCN model utilized ReLU activation for all layers except the output, which employed a linear activation (see Supplementary Figure 3A ). For the analysis of the dFNC data, we employed a bi-directional Long Short-Term Memory network (biLSTM) 60 configured with three recurrent layers with 128 hidden units each, a dropout rate of 0.1, and a fully connected regression layer to predict brain age (see Supplementary Figure 3B ). In both models, we utilized an Adam optimizer targeting Mean Absolute Error, with a learning rate of 1e -3 and a batch size of 64, conducting training over 100 epochs. The epoch with the optimal cross-validation performance was selected for inference on the test set. Figure 1D and Figure 1E illustrate the modeling and testing procedures for the sFNC and dFNC features, respectively. Statistical analysis We developed brain age prediction models incorporating both sFNC and dFNC features. These models were then applied to the FBIRN dataset to estimate the predicted brain age and subsequently the BAG for each participant. To investigate the associations between these BAGs and cognitive measures, including working memory and attention vigilance, we constructed a General Linear Model (GLM). This model included age, sex, site, age 2 , the interaction of age and sex, and the diagnosis (schizophrenia or control) as covariates. For each cognitive measure, we tested two hypotheses—assessing the association of BAG derived from two modalities (sFNC-based and dFNC-based BAGs). We corrected for multiple comparisons across the two modalities using the False Discovery Rate (FDR) correction method to ensure the robustness of our findings 61 . Declarations Disclosures Dr. Sendi has served as a consultant for Niji Corp for unrelated work. Dr. Mathalon has served as a consultant for Aptinyx, Boehringer-Ingelheim Pharmaceuticals, Cadent Therapeutics, and Greenwich Biosciences for unrelated work. The remaining authors declare no competing interests. Funding /Support This work was supported by the National Institutes of Health (NIH) grants R01MH123610 and T32MH125786, the National Science Foundation (NSF) grant 2112455, and the Phyllis and Jerome Lyle Rappaport Mental Health Research Scholars Award. Author Contributions Sabrina J. Edwards-Swart, Vince D. Calhoun, and Mohammad S. E. Sendi designed the study. Acknowledgements We thank the participants of this study and those involved in data collection. Data Availability The data utilized in the preparation of this manuscript are publicly available. The UK Biobank (UKBB) data are available via application at https://www.ukbiobank.ac.uk . The Human Connectome Project Young Adult (HCP-YA) and Aging (HCP-A) datasets are available through the Connectome Coordination Facility at https://www.humanconnectome.org . The FBIRN dataset is publicly accessible through the NITRC Image Repository ( https://www.nitrc.org/ ), under the Functional Biomedical Informatics Research Network (FBIRN) project. All data used in this study were obtained under appropriate data use agreements and institutional review board approvals. Code Availability The code used for preprocessing and FNC calculation is available at https://trendscenter.org/software/ . Also, statistical parametric mapping (SPM 12) is available at https://www.fil.ion.ucl.ac.uk/spm/ . The NeuroMark framework and the NeuroMark template (Neuromark_fMRI_1.0) have been made available and incorporated into the Group ICA Toolbox ( https://github.com/trendscenter/gift ). Users worldwide can now directly download and utilize these resources. We also use this https://www.nitrc.org/projects/bnv/ for the brain graph. References Desai, P. R., Lawson, K. A., Barner, J. C. & Rascati, K. L. Estimating the direct and indirect costs for community-dwelling patients with schizophrenia. Journal of Pharmaceutical Health Services Research 4, 187–194 (2013). Kessler, R. C. et al. The prevalence and correlates of nonaffective psychosis in the National Comorbidity Survey Replication (NCS-R). Biol Psychiatry 58, 668–676 (2005). Guo, J. Y., Ragland, J. D. & Carter, C. S. Memory and cognition in schizophrenia. Molecular Psychiatry vol. 24 633–642 Preprint at https://doi.org/10.1038/s41380-018-0231-1 (2019). Van Snellenberg, J. X. et al. Mechanisms of Working Memory Impairment in Schizophrenia. Biol Psychiatry 80, 617–626 (2016). Dutterer, J., Bansal, S., Robinson, B. & Gold, J. M. Sustained attention deficits in schizophrenia: Effect of memory load on the Identical Pairs Continuous Performance Test. Schizophr Res Cogn 33, (2023). Li, Y. T. et al. Neuroimaging features of cognitive impairments in schizophrenia and major depressive disorder. Therapeutic Advances in Psychopharmacology vol. 14 Preprint at https://doi.org/10.1177/20451253241243290 (2024). Gur, R. E. & Gur, R. C. Functional magnetic resonance imaging in schizophrenia. Dialogues Clin Neurosci 12, 333–343 (2010). Penadés, R., Franck, N., González-Vallespí, L. & Dekerle, M. Neuroimaging Studies of Cognitive Function in Schizophrenia. in Advances in Experimental Medicine and Biology vol. 1118 117–134 (Springer New York LLC, 2019). Chang, J. R., Yao, Z. F., Hsieh, S. & Nordling, T. E. M. Age Prediction Using Resting-State Functional MRI. Neuroinformatics 22, 119–134 (2024). Koutsouleris, N. et al. Accelerated brain aging in schizophrenia and beyond: A neuroanatomical marker of psychiatric disorders. Schizophr Bull 40, 1140–1153 (2014). Schnack, H. G. et al. Accelerated brain aging in schizophrenia: A longitudinal pattern recognition study. American Journal of Psychiatry 173, 607–616 (2016). Shahab, S. et al. Brain structure, cognition, and brain age in schizophrenia, bipolar disorder, and healthy controls. Neuropsychopharmacology 44, 898–906 (2019). Joo, Y. et al. Brain age prediction using combined deep convolutional neural network and multi-layer perceptron algorithms. Sci Rep 13, (2023). Sendi, M. S., Salat, D. H. & Calhoun, V. D. Brain age gap difference between healthy and mild dementia subjects: Functional network connectivity analysis. in 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (IEEE, Virtual Conference, 2021). Wrigglesworth, J. et al. Brain-predicted age difference is associated with cognitive processing in later-life. Neurobiol Aging 109, 195–203 (2022). Anatürk, M. et al. Prediction of brain age and cognitive age: Quantifying brain and cognitive maintenance in aging. Hum Brain Mapp 42, 1626–1640 (2021). Elliott, M. L. et al. Brain-age in midlife is associated with accelerated biological aging and cognitive decline in a longitudinal birth cohort. Mol Psychiatry 26, 3829–3838 (2021). Stefaniak, J. D. et al. Brain age gap, dementia risk factors and cognition in middle age. Brain Commun 6, (2024). Shahab, S. et al. Brain structure, cognition, and brain age in schizophrenia, bipolar disorder, and healthy controls. Neuropsychopharmacology 44, 898–906 (2019). Koutsouleris, N. et al. Accelerated brain aging in schizophrenia and beyond: A neuroanatomical marker of psychiatric disorders. Schizophr Bull 40, 1140–1153 (2014). Schnack, H. G. et al. Accelerated brain aging in schizophrenia: A longitudinal pattern recognition study. American Journal of Psychiatry 173, 607–616 (2016). Hajek, T. et al. Brain Age in Early Stages of Bipolar Disorders or Schizophrenia. Schizophr Bull 45, 191–198 (2019). Nenadić, I., Dietzek, M., Langbein, K., Sauer, H. & Gaser, C. BrainAGE score indicates accelerated brain aging in schizophrenia, but not bipolar disorder. Psychiatry Res Neuroimaging 266, 86–89 (2017). Wang, J. et al. White matter brain aging in relationship to schizophrenia and its cognitive deficit. Schizophr Res 230, 9–16 (2021). Zhu, J. D., Wu, Y. F., Tsai, S. J., Lin, C. P. & Yang, A. C. Investigating brain aging trajectory deviations in different brain regions of individuals with schizophrenia using multimodal magnetic resonance imaging and brain-age prediction: a multicenter study. Transl Psychiatry 13, (2023). Millar, P. R. et al. Predicting brain age from functional connectivity in symptomatic and preclinical Alzheimer disease. Neuroimage 256, (2022). Monti, R. P. et al. Interpretable brain age prediction using linear latent variable models of functional connectivity. PLoS One 15, (2020). Lund, M. J. et al. Brain age prediction using fMRI network coupling in youths and associations with psychiatric symptoms. Neuroimage Clin 33, (2022). Sakoğlu, Ü. et al. A method for evaluating dynamic functional network connectivity and task-modulation: Application to schizophrenia. Magnetic Resonance Materials in Physics, Biology and Medicine 23, 351–366 (2010). Calhoun, V. D., Miller, R., Pearlson, G. & Adali, T. The Chronnectome: Time-Varying Connectivity Networks as the Next Frontier in fMRI Data Discovery. Neuron 84, 262–274 (2014). Lurie, D. J. et al. Questions and controversies in the study of time-varying functional connectivity in resting fMRI. Network Neuroscience vol. 4 30–69 Preprint at https://doi.org/10.1162/netn_a_00116 (2020). Sendi, M. S. E. et al. Multiple overlapping dynamic patterns of the visual sensory network in schizophrenia. Schizophr Res 228, 103–111 (2020). Sendi, M. S. E. et al. The link between static and dynamic brain functional network connectivity and genetic risk of Alzheimer’s disease. Neuroimage Clin 37, 103363 (2023). Sendi, M. S. et al. Brain dynamics reflecting an intra-network brain state is associated with increased posttraumatic stress symptoms in the early aftermath of trauma. Nature Mental Health (2024). Sendi, M. S. E. et al. Alzheimer’s Disease Projection From Normal to Mild Dementia Reflected in Functional Network Connectivity: A Longitudinal Study. Front Neural Circuits 14, (2021). Dini, H. et al. Dynamic functional connectivity predicts treatment response to electroconvulsive therapy in major depressive disorder. Front Hum Neurosci 15, 1–11 (2021). Allen, E. A. et al. Tracking whole-brain connectivity dynamics in the resting state. Cerebral Cortex 24, 663–676 (2014). Miller, K. L. et al. Multimodal population brain imaging in the UK Biobank prospective epidemiological study. Nat Neurosci 19, 1523–1536 (2016). Littlejohns, T. J. et al. The UK Biobank imaging enhancement of 100,000 participants: rationale, data collection, management and future directions. Nature Communications vol. 11 Preprint at (2020). Glasser, M. F. et al. The Human Connectome Project’s neuroimaging approach. Nat Neurosci 19, 1175–1187 (2016). Van Essen, D. C. et al. The WU-Minn Human Connectome Project: An overview. Neuroimage 80, 62–79 (2013). Bookheimer, S. Y. et al. The Lifespan Human Connectome Project in Aging: An overview. Neuroimage 185, 335–348 (2019). Keator, D. B. et al. The Function Biomedical Informatics Research Network Data Repository. Neuroimage 124, 1074–1079 (2016). Du, Y. et al. NeuroMark: An automated and adaptive ICA based pipeline to identify reproducible fMRI markers of brain disorders. Neuroimage Clin 28, 102375 (2020). Filley, C. M. & Cullum, C. M. Attention and vigilance functions in normal aging. Appl Neuropsychol 1, 29–32 (1994). Nuechterlein, K. H. et al. Attention/vigilance in schizophrenia: Performance results from a large multi-site study of the Consortium on the Genetics of Schizophrenia (COGS). Schizophr Res 163, 38–46 (2015). Coull, J. T. fMRI studies of temporal attention: Allocating attention within, or towards, time. Cognitive Brain Research 21, 216–226 (2004). Williams, J. H. G., Waiter, G. D., Perra, O., Perrett, D. I. & Whiten, A. An fMRI study of joint attention experience. Neuroimage 25, 133–140 (2005). Forbes, N. F., Carrick, L. A., McIntosh, A. M. & Lawrie, S. M. Working memory in schizophrenia: A meta-analysis. Psychological Medicine vol. 39 889–905 Preprint at https://doi.org/10.1017/S0033291708004558 (2009). Wingfield, A., Stine, E. A. L., Lahar, C. J. & Aberdeen, J. S. Does the capacity of working memory change with age? Exp Aging Res 14, 103–107 (1988). Zhu, J. et al. Dynamic functional connectome predicts individual working memory performance across diagnostic categories. Neuroimage Clin 30, (2021). Faghiri, A., Thomas, D., Oktay, A. & Calhoun, V. D. A unified approach to study the brain connectivity frequency profile. in International Symposium on Biomedical Imaging (ISBI) (2020). Hsu, H. C. & Bai, C. H. Individual and environmental factors associated with cognitive function in older people: a longitudinal multilevel analysis. BMC Geriatr 22, (2022). Jeong, H. J. et al. General and Specific Factors of Environmental Stress and Their Associations With Brain Structure and Dimensions of Psychopathology. Biological Psychiatry Global Open Science 3, 480–489 (2023). Fu, Z. et al. Whole-Brain Functional Connectivity Dynamics Associated With Electroconvulsive Therapy Treatment Response. Biol Psychiatry Cogn Neurosci Neuroimaging 7, 312–322 (2022). Sendi, M. S. E., Salat, D. H., Miller, R. L. & Calhoun, V. D. Two-step clustering-based pipeline for big dynamic functional network connectivity data. Front Neurosci 16, (2022). Sendi, M. S. E. et al. Aberrant dynamic functional connectivity of default mode network predicts symptom severity in major depressive disorder. Brain Connect 11, 838–849 (2021). Yao, D. et al. A Mutual Multi-Scale Triplet Graph Convolutional Network for Classification of Brain Disorders Using Functional or Structural Connectivity. IEEE Trans Med Imaging 40, 1279–1289 (2021). Li, Y. et al. Brain Connectivity Based Graph Convolutional Networks and Its Application to Infant Age Prediction. IEEE Trans Med Imaging 41, 2764–2776 (2022). Hochreiter, S. & Schmidhuber, J. Long short-term memory. Neural Comput 9, (1997). Hochberg, Y. & Tamhane, A. C. Multiple Comparison Procedures. (John Wiley and Sons Inc., 1987). Additional Declarations Yes there is potential Competing Interest. Dr. Sendi has served as a consultant for Niji Corp for unrelated work. Dr. Mathalon has served as a consultant for Aptinyx, Boehringer-Ingelheim Pharmaceuticals, Cadent Therapeutics, and Greenwich Biosciences for unrelated work. The remaining authors declare no competing interests. Supplementary Files 2025EdwardsSwartSupplementaryInfoNMH.docx Supplementary Materials Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7336363","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":506749278,"identity":"a3d8cfb9-9a0b-4a0e-b89b-3931df9d6518","order_by":0,"name":"Mohammad Sendi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA90lEQVRIiWNgGAWjYLACHgMbBgZ2MPMAVISBgbEBv5Y0BgZm0rQwHCZBi27/4WMP3hScj+ZvZj74uYDhjrz5jORnH94w2MhuOIBdi9mNtHTDOQa3c2ccZkuWnsHwzHDOjTTjmXMY0oxxa+Exk+YBatnAzGMgDXQh4wyeA8bMQEYiTi3nz38DajkH1ML/+TdQpf0MnuOfgVr+49ZyIIcNqOUAyBY2kC2JM9h7QLYcwK3lRpqZ5ByDZJBfzKx5DA4nA7UUMwJFjGfidNjhZxJv/tjl9rc3P77NU3HYdgYz+2aGNxV2sn04tKABAwzGKBgFo2AUjAJyAAADcVsq0eScjAAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0003-2946-1786","institution":"Harvard Medical School/McLean Hospital","correspondingAuthor":true,"prefix":"","firstName":"Mohammad","middleName":"","lastName":"Sendi","suffix":""},{"id":506749279,"identity":"e6301f2a-8d73-47b1-95dd-000afd7bb4d3","order_by":1,"name":"Sabrina Edwards-Swart","email":"","orcid":"","institution":"Georgia Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Sabrina","middleName":"","lastName":"Edwards-Swart","suffix":""},{"id":506749280,"identity":"1ff982e9-9e11-46c2-8fcb-d962d9b0fe94","order_by":2,"name":"Bradley Baker","email":"","orcid":"","institution":"Georgia State University","correspondingAuthor":false,"prefix":"","firstName":"Bradley","middleName":"","lastName":"Baker","suffix":""},{"id":506749281,"identity":"5b5d36d4-f331-43cc-86ff-509bef801aaf","order_by":3,"name":"Daniel Mathalon","email":"","orcid":"","institution":"University of California, San Francisco","correspondingAuthor":false,"prefix":"","firstName":"Daniel","middleName":"","lastName":"Mathalon","suffix":""},{"id":506749282,"identity":"d2fce6fb-8e92-4918-8851-6775ce0a388d","order_by":4,"name":"Judith Ford","email":"","orcid":"","institution":"Department of Psychiatry and Behavioral Sciences, University of California, San Francisco","correspondingAuthor":false,"prefix":"","firstName":"Judith","middleName":"","lastName":"Ford","suffix":""},{"id":506749283,"identity":"05f5f424-405e-4dc3-b4ae-4948f8e288d8","order_by":5,"name":"Adrian Preda","email":"","orcid":"","institution":"University of California, Irvine","correspondingAuthor":false,"prefix":"","firstName":"Adrian","middleName":"","lastName":"Preda","suffix":""},{"id":506749284,"identity":"3a7935c7-1b69-4c91-8f17-2c11697f1a23","order_by":6,"name":"Theo van Erp","email":"","orcid":"","institution":"University of California Irvine","correspondingAuthor":false,"prefix":"","firstName":"Theo","middleName":"van","lastName":"Erp","suffix":""},{"id":506749285,"identity":"9a6f4d4b-9be7-4862-83c6-f74401986eeb","order_by":7,"name":"Godfrey Pearlson","email":"","orcid":"","institution":"Yale University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Godfrey","middleName":"","lastName":"Pearlson","suffix":""},{"id":506749286,"identity":"2a963ec1-3860-4573-9fa5-a098d41c9d17","order_by":8,"name":"Jessica Turner","email":"","orcid":"","institution":"The Ohio State University","correspondingAuthor":false,"prefix":"","firstName":"Jessica","middleName":"","lastName":"Turner","suffix":""},{"id":506749287,"identity":"964abf03-ce0d-4af0-a11f-0bfd0e3572a8","order_by":9,"name":"Vince Calhoun","email":"","orcid":"https://orcid.org/0000-0001-9058-0747","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Vince","middleName":"","lastName":"Calhoun","suffix":""}],"badges":[],"createdAt":"2025-08-10 02:20:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7336363/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7336363/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93268504,"identity":"fa41ff80-b24d-4957-a44d-b371f0b182f2","added_by":"auto","created_at":"2025-10-10 21:13:19","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1636884,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eA)\u003c/strong\u003eCount density of the training dataset used in the brain age prediction model employing both sFNC and dFNC variables. The training set comprises 17,522 individuals aged 22 to 100 years, including 8,090 females and 9,432 males. \u003cstrong\u003eB)\u003c/strong\u003eCount density of the FBIRN dataset used as test data to examine the link between BAG and cognitive/attention measures. The dataset includes 311 individuals aged 18 to 62 years (81 females and 230 males); our analysis focused on those with schizophrenia. \u003cstrong\u003eC)\u003c/strong\u003e NeuroMark pipeline extraction of 53 regions across seven networks: subcortical (SCN), auditory (AUD), sensorimotor (SMN), visual (VSN), cognitive control (CCN), default mode (DMN), and cerebellar (CBN). Both static and dynamic functional network connectivity (FNC) were computed between each pair of regions using the NeuroMark template, yielding a 53×53 connectivity matrix. A representative mean FNC was estimated from UK Biobank participants. \u003cstrong\u003eD)\u003c/strong\u003e Illustration of sFNC data processing: Training data (UKBB, HCP, and HCP-Aging) are input into an untrained model with chronological age as the label. The trained model then predicts brain age from test data (FBIRN). Comparison of predicted versus chronological age determines the brain age gap. \u003cstrong\u003eE)\u003c/strong\u003e Illustration of the dFNC data processing, analogous to panel D, demonstrating the training and application of the model using dynamic connectivity data.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7336363/v1/b428f4c7273982d81512dc54.png"},{"id":93268506,"identity":"b607fa86-f4fa-4bc8-9c56-d348a2c44af4","added_by":"auto","created_at":"2025-10-10 21:13:19","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":144060,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCorrelation between brain age predictions and chronological age in sFNC- and dFNC-based models. \u003c/strong\u003eThese graphs illustrate the relationship between chronological age and predicted brain age in the training dataset (UK Biobank, HCP, and HCP Aging).\u0026nbsp;\u003cstrong\u003eA)\u003c/strong\u003e\u0026nbsp;sFNC-based predictions (green) show a strong positive correlation with chronological age (r = 0.8324, p \u0026lt; 1.1755e\u003csup\u003e-38\u003c/sup\u003e).\u0026nbsp;\u003cstrong\u003eB)\u003c/strong\u003e\u0026nbsp;dFNC-based predictions (orange) also show a significant positive correlation (r = 0.801, p \u0026lt; 1.1755e\u003csup\u003e-38\u003c/sup\u003e).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7336363/v1/8f8cd47124a5d86529dbbd87.png"},{"id":93268588,"identity":"f2dec557-d523-4268-82ad-0bbdeeadc755","added_by":"auto","created_at":"2025-10-10 21:21:19","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":119868,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAssociations between sFNC- and dFNC-based wide-brain age gap (wBAG) and cognitive performance in individuals with schizophrenia.\u003c/strong\u003e After developing and validating the brain age prediction models, we applied them to the FBIRN dataset to calculate wBAGs based on static (sFNC) and dynamic (dFNC) functional network connectivity features. Analyses were restricted to individuals older than 38 years due to reduced model accuracy in younger participants (see \u003cstrong\u003eFigure 2\u003c/strong\u003e), resulting in a final sample of 153 participants (mean age = 47.73 ± 5.94 years; 121 males). General Linear Models (GLMs) were used to examine associations between wBAGs and two cognitive domains—attention vigilance and working memory—controlling for age, sex, study site, age squared, age-by-sex interaction, and diagnosis. P-values were adjusted for multiple comparisons across modalities using the Benjamini-Hochberg FDR correction. \u003cstrong\u003eA)\u003c/strong\u003esFNC-based wBAG was significantly negatively associated with attention vigilance (\u003cem\u003er\u003c/em\u003e=–0.2923, \u003cem\u003eβ\u003c/em\u003e=–0.9860, \u003cem\u003eSE\u003c/em\u003e=0.2829, \u003cem\u003e95% CI\u003c/em\u003e: –1.5454 to –0.4263, \u003cem\u003eFDR p\u003c/em\u003e=0.0013, \u003cem\u003eN\u003c/em\u003e=153). \u003cstrong\u003eB)\u003c/strong\u003e dFNC-based wBAG was significantly negatively associated with attention vigilance (\u003cem\u003er\u003c/em\u003e=–0.2715, \u003cem\u003eβ\u003c/em\u003e=–0.4395, \u003cem\u003eSE\u003c/em\u003e=0.1366, \u003cem\u003e95% CI\u003c/em\u003e: –0.7099 to –0.1691, \u003cem\u003eFDR p\u003c/em\u003e=0.0016, N=153). \u003cstrong\u003eC)\u003c/strong\u003e sFNC-based wBAG was significantly negatively associated with working memory (\u003cem\u003er\u003c/em\u003e=–0.2237, \u003cem\u003eβ\u003c/em\u003e=–0.9603, \u003cem\u003eSE\u003c/em\u003e=0.3615, \u003cem\u003e95% CI\u003c/em\u003e: –1.6761 to –0.2445, \u003cem\u003eFDR p\u003c/em\u003e=0.0088, \u003cem\u003eN\u003c/em\u003e=153). D) dFNC-based wBAG was significantly negatively associated with working memory (\u003cem\u003er\u003c/em\u003e=–0.2508, \u003cem\u003eβ\u003c/em\u003e=–0.5195, \u003cem\u003eSE\u003c/em\u003e=0.1732, \u003cem\u003e95% CI\u003c/em\u003e: –0.8621 to –0.1769, \u003cem\u003eFDR p\u003c/em\u003e=0.0064, \u003cem\u003eN\u003c/em\u003e=153). Across both cognitive domains, higher wBAGs—reflecting accelerated brain aging—were associated with poorer cognitive performance.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7336363/v1/c93259fd1b085c2ecd1fb60f.png"},{"id":93268508,"identity":"1e8df56b-6377-43d8-9298-4ed1ab720d21","added_by":"auto","created_at":"2025-10-10 21:13:19","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1114174,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAssociations between subBAGs and cognitive performance. A)\u003c/strong\u003e Relationship between subBAGs and attention vigilance. Crosshatched bars represent \u003cem\u003edFNC\u003c/em\u003e-based subBAGs, and solid bars represent \u003cem\u003esFNC-\u003c/em\u003ebased subBAGs. Bars with double asterisks indicate significant associations after FDR correction. Significant negative associations were observed for \u003cem\u003esFNC\u003c/em\u003e-based subBAGs from the subcortical network (\u003cem\u003er\u003c/em\u003e=-0.2357, \u003cem\u003eβ\u003c/em\u003e=-0.1813, \u003cem\u003eSE\u003c/em\u003e=0.0655, \u003cem\u003e95% CI\u003c/em\u003e=-0.3111 to -0.0516, \u003cem\u003eFDR p\u003c/em\u003e=0.0228, \u003cem\u003eN\u003c/em\u003e=153) and \u003cem\u003edFNC-\u003c/em\u003ebased subBAGs from the subcortical network (\u003cem\u003er\u003c/em\u003e=-0.2806, \u003cem\u003eβ\u003c/em\u003e=-0.2989, \u003cem\u003eSE\u003c/em\u003e=0.0896, 95% \u003cem\u003eCI\u003c/em\u003e=-0.4763 to -0.1215, \u003cem\u003eFDR p\u003c/em\u003e=0.0140, \u003cem\u003eN\u003c/em\u003e=153). Additional significant negative associations were observed for \u003cem\u003edFNC\u003c/em\u003e-based subBAGs from the sensorimotor network (\u003cem\u003er\u003c/em\u003e=-0.2384, \u003cem\u003eβ\u003c/em\u003e=-0.2153, \u003cem\u003eSE\u003c/em\u003e=0.0769, 95% \u003cem\u003eCI\u003c/em\u003e=-0.3675 to -0.0631, \u003cem\u003eFDR p\u003c/em\u003e=0.0228, \u003cem\u003eN\u003c/em\u003e=153) and the default mode network (\u003cem\u003er\u003c/em\u003e=-0.2666, \u003cem\u003eβ\u003c/em\u003e=-0.1053, \u003cem\u003eSE\u003c/em\u003e=0.0333, 95% \u003cem\u003eCI\u003c/em\u003e=-0.1713 to -0.0392, \u003cem\u003eFDR p\u003c/em\u003e=0.0140, \u003cem\u003eN\u003c/em\u003e=153). The \u003cem\u003esFNC\u003c/em\u003e-based subBAGs from the cognitive control network also showed a significant negative association (\u003cem\u003er\u003c/em\u003e=-0.2081, \u003cem\u003eβ\u003c/em\u003e=-0.0353, \u003cem\u003eSE\u003c/em\u003e=0.0145, 95% \u003cem\u003eCI\u003c/em\u003e=-0.0641 to -0.0065, \u003cem\u003eFDR p\u003c/em\u003e=0.0455, \u003cem\u003eN\u003c/em\u003e=153). \u003cstrong\u003eB)\u003c/strong\u003e Relationship between subBAGs and working memory. Significant negative associations (\u003cem\u003euncorrected p\u003c/em\u003e\u0026lt;0.05) were identified for \u003cem\u003esFNC-\u003c/em\u003ebased subBAGs from the subcortical network (\u003cem\u003er\u003c/em\u003e=-0.1724, \u003cem\u003eβ\u003c/em\u003e=-0.1050, \u003cem\u003eSE\u003c/em\u003e=0.0518, 95% \u003cem\u003eCI\u003c/em\u003e=-0.2076 to -0.0020, \u003cem\u003euncorrected p\u003c/em\u003e=0.0447, \u003cem\u003eN\u003c/em\u003e=153), \u003cem\u003edFNC\u003c/em\u003e-based subBAGs from the subcortical network (\u003cem\u003er\u003c/em\u003e=-0.1754, \u003cem\u003eβ\u003c/em\u003e=-0.1473, \u003cem\u003eSE\u003c/em\u003e=0.0714, 95% \u003cem\u003eCI\u003c/em\u003e=-0.2885 to -0.0060, \u003cem\u003euncorrected p\u003c/em\u003e=0.0410, \u003cem\u003eN\u003c/em\u003e=153), and \u003cem\u003edFNC\u003c/em\u003e-based subBAGs from the default mode network (\u003cem\u003er\u003c/em\u003e=-0.2114, \u003cem\u003eβ\u003c/em\u003e=-0.0662, \u003cem\u003eSE\u003c/em\u003e=0.0264, 95% \u003cem\u003eCI\u003c/em\u003e=-0.1186 to -0.0139, \u003cem\u003euncorrected p\u003c/em\u003e=0.0134, \u003cem\u003eN\u003c/em\u003e=153). However, none of these associations survived FDR correction. All models were adjusted for age, sex, site, age², age-by-sex interaction, and diagnosis.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7336363/v1/e64fc0bdd8fbaaa5d0ed7e32.png"},{"id":93268811,"identity":"5595a288-afaf-423c-af0f-a0f5be679611","added_by":"auto","created_at":"2025-10-10 21:29:25","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4237068,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7336363/v1/57f9943e-a2ea-479a-80dd-de853ccec75e.pdf"},{"id":93268507,"identity":"717431f5-f0e9-47d5-81a5-7450cd7bb140","added_by":"auto","created_at":"2025-10-10 21:13:19","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":4485719,"visible":true,"origin":"","legend":"Supplementary Materials","description":"","filename":"2025EdwardsSwartSupplementaryInfoNMH.docx","url":"https://assets-eu.researchsquare.com/files/rs-7336363/v1/43da5658f73062cd9a01a88e.docx"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nDr. Sendi has served as a consultant for Niji Corp for unrelated work. Dr. Mathalon has served as a consultant for Aptinyx, Boehringer-Ingelheim Pharmaceuticals, Cadent Therapeutics, and Greenwich Biosciences for unrelated work. The remaining authors declare no competing interests.","formattedTitle":"Dynamic Brain Age Modeling Identifies Network-Specific Cognitive Deficits in Schizophrenia","fulltext":[{"header":"Introduction","content":"\u003cp\u003eSchizophrenia, affecting 0.25% to 0.64% of the U.S. population, is a complex psychiatric disorder marked by profound disruptions in cognition, emotion, and social interactions\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Key symptoms include hallucinations, delusions, and motor impairments, alongside significant reductions in emotional expression and social engagement. Critically, individuals with schizophrenia exhibit marked cognitive deficits, particularly in working memory and attention vigilance. Studies indicate they perform approximately 2.5 standard deviations below healthy controls in these cognitive domains\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. These deficits are characterized by difficulties in retaining and updating information in working memory\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e, with increasing cognitive demands leading to more pronounced declines in attention vigilance compared to healthy individuals\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eIn recent decades, neuroimaging techniques have greatly enhanced our understanding of cognitive function and attention deficits in schizophrenia\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Structural magnetic resonance imaging (sMRI) provides detailed anatomical insights, associations between executive function impairments and reductions in prefrontal cortex volume and thickness, and links episodic memory deficits with hippocampal atrophy. Conversely, functional magnetic resonance imaging (fMRI) measures brain activity through oxygen metabolism and blood flow, illuminating functional dynamics across various brain regions\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. Functional MRI has been pivotal in identifying brain regions involved in working memory, offering valuable insights into cognitive impairments in patient populations compared to healthy controls. Broadly, neuroimaging plays a key role in examining executive function and episodic memory impairments and provides a valuable tool for tracking brain function changes following cognitive remediation therapies\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eIn recent years, brain age prediction from neuroimaging has emerged as a promising approach for identifying novel biomarkers of neurological and neuropsychiatric conditions\u003csup\u003e\u003cspan additionalcitationids=\"CR10 CR11 CR12 CR13\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. By modeling the relationship between neuroimaging variables and age in healthy individuals, this method enables the detection of meaningful brain deviations in patient populations. The difference between an individual\u0026rsquo;s predicted brain age and their chronological age, termed the brain age gap (BAG), provides critical insights into the pathology of cognitive function\u003csup\u003e\u003cspan additionalcitationids=\"CR16 CR17\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. For example, emerging evidence demonstrates that individuals with accelerated brain age consistently show marked deficits across critical cognitive domains\u0026mdash;including IQ, verbal comprehension, perceptual reasoning, processing speed, working memory, and memory recall\u0026mdash;as captured by gold-standard assessments such as the Rey Auditory Verbal Learning Test\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. In schizophrenia, studies using sMRI have revealed that patients often exhibit a brain age 6 to 8 years older than their chronological age\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. Further, the BAG was noted to increase across at-risk, recent onset\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e, and recurrent phases of schizophrenia, initially rising approximately a year and a half annually for the first five years before stabilizing\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eAdditionally, schizophrenia and bipolar disorder present distinct BAG profiles. Individuals with schizophrenia consistently exhibit elevated BAGs, whereas those with bipolar disorder maintain levels comparable to healthy controls. This distinction underscores the potential of BAG as a valuable biomarker for enhancing early differential diagnosis\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. This suggests a progressive pathogenic component unique to schizophrenia, highlighting the utility of BAG as a biomarker\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. Moreover, a significant negative association between BAG and cognitive functions like working memory has been observed in schizophrenia, highlighting the potential of BAG in understanding and diagnosing declined neurocognitive performance in schizophrenia\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eWhile the majority of brain age prediction research has traditionally focused on structural MRI (sMRI), recent advancements include employing functional connectivity metrics from resting-state fMRI to predict brain age\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan additionalcitationids=\"CR26\" citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. A notable study using resting-state fMRI data from the Philadelphia Neurodevelopmental Cohort demonstrated the value of brain age models in youth by linking older estimated brain age to greater symptom burden, especially across DSM-5 psychiatric diagnoses. These findings underscore the potential of brain age to capture deviations in brain maturation associated with mental health conditions\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. Furthermore, an innovative study used resting-state fMRI and brain age prediction to identify neural connections related to abnormal brain aging. By systematically excluding connections from the training model, the approach pinpointed those most critical to brain age accuracy, providing new insights into the neurobiological mechanisms of age-related psychiatric conditions\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eAlthough brain age prediction using static functional connectivity (sFNC), estimated from resting-state fMRI, has proven valuable, the potential of dynamic functional connectivity (dFNC) in this area is still underexplored. Unlike sFNC, which evaluates connectivity from correlations across an entire time series, dFNC analyzes connections between brain regions or networks within specific time intervals. This allows for the capture of temporal fluctuations that reveal how brain network interactions change over time\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e,\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. The variability and dynamism of neural signals, captured by dFNC, are crucial in understanding cognitive deficits and clinical symptoms in psychiatric disorders. These dynamic measures provide insights into brain function that static approaches cannot, highlighting changes and interactions that are critical in disease progression \u003csup\u003e\u003cspan additionalcitationids=\"CR32 CR33 CR34 CR35 CR36\" citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. Because disrupted brain network dynamics potentially drive cognitive symptoms in schizophrenia, leveraging a BAG derived from dFNC offers a powerful, next-generation marker of disease-related neurobiological aging. Critically, no study has yet evaluated whether dFNC-based BAG outperforms traditional static FNC (sFNC) BAG in explaining core cognitive deficits such as working memory and attention. Establishing this added predictive power would mark a transformative advance, positioning dFNC BAG as a mechanistically grounded biomarker and unlocking new precision targets for circuit-based interventions in schizophrenia and related psychiatric conditions.\u003c/p\u003e\u003cp\u003eHere we address this gap using the most significant sample to date: 22,569 resting-state scans from 17,522 healthy adults from UK Biobank\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e,\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e and Human Connectome Project\u003csup\u003e\u003cspan additionalcitationids=\"CR41\" citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e to train multiple wide-brain and sub-network brain age prediction models, and an independent cohort of 153 controls and individuals with schizophrenia from the FBIRN consortium to test them \u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. Leveraging a graph-convolutional network for sFNC and a bidirectional long short-term memory or LSTM for dFNC, we generated wide-brain and sub-network BAGs and evaluated their links to attention vigilance and working memory performance while controlling for demographic and clinical covariates. We hypothesize that (1) BAGs derived from dFNC will predict age as accurately as sFNC models, and (2) larger BAGs, particularly those based on dFNC, will associate with poorer cognitive performance, reflecting accelerated dynamic functional brain ageing in schizophrenia. By integrating dynamic connectivity with brain age modeling, this study provides the first direct evidence linking aberrant network dynamics to cognitive impairment. It establishes dFNC-based BAG as a clinically relevant biomarker.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eStudy Population\u003c/h2\u003e\u003cp\u003eOur training dataset comprised 22,569 resting-state fMRI scans from 17,522 healthy individuals aged 22 to 100 years (mean age\u0026thinsp;=\u0026thinsp;62.42\u0026thinsp;\u0026plusmn;\u0026thinsp;10.96 years; 8,090 females, 9,432 males), drawn from the UK Biobank (UKBB)\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e,\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e, Human Connectome Project Young Adult (HCP-YA\u003csup\u003e40\u003c/sup\u003e ), and Human Connectome Project Aging (HCP-A\u003csup\u003e42\u003c/sup\u003e) cohorts. Stringent exclusion criteria were applied to UKBB participants to remove individuals with primary psychiatric or neurological conditions (see \u003cem\u003eMethod Section\u003c/em\u003e). The HCP datasets provided approximately 15 minutes of resting-state fMRI per subject, while UKBB contributed\u0026thinsp;~\u0026thinsp;6 minutes per scan. To evaluate generalizability and clinical relevance, we validated our models in an independent clinical dataset (FBIRN; N\u0026thinsp;=\u0026thinsp;311), including 151 individuals with schizophrenia and 160 healthy controls aged 18 to 62 years (mean age\u0026thinsp;=\u0026thinsp;37.88 years; 81 females, 230 males). Diagnoses were confirmed using structured clinical interviews. Detailed demographic characteristics are shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, with site-level breakdowns of the FBIRN dataset in \u003cb\u003eSupplementary Table\u0026nbsp;1\u003c/b\u003e. Also, Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA shows the count density of the training datasets, including UKBB, HCP-YA, and HCP-A, and Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB displays the count density of the test dataset (i.e., FBIRN) used in our study.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eThe Training Data set is composed of UKBB, HCP, and HCP-A datasets, while the FBIRN data set comprises the Test Data.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eAge (M\u0026thinsp;\u0026plusmn;\u0026thinsp;SD)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eSex (M/F)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAll Training Data\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e17522\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e62.42\u0026thinsp;\u0026plusmn;\u0026thinsp;10.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e9432/8090\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUKBB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e15978\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e64.26\u0026thinsp;\u0026plusmn;\u0026thinsp;7.55\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e8734 / 7244\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eTraining\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHCP\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e833\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e28.66\u0026thinsp;\u0026plusmn;\u0026thinsp;3.66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e390 / 443\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHCP-A\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e711\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e60.54\u0026thinsp;\u0026plusmn;\u0026thinsp;15.68\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e308 / 403\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eTest\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFBIRN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e311\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e\u003cp\u003e37.88\u0026thinsp;\u0026plusmn;\u0026thinsp;11.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e230 / 81\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eTraining and validating the brain age prediction model on the healthy population\u003c/h3\u003e\n\u003cp\u003eWe extracted reliable brain networks using the NeuroMark automated ICA pipeline\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e, which applies spatially constrained templates derived from large normative datasets. This method identified 53 independent components, organized into seven functional networks. These seven networks include subcortical (SCN), auditory (AUD), sensorimotor (SMN), visual (VSN), cognitive control (CCN), default-mode (DMN), and cerebellar networks (CBN) as shown in \u003cb\u003eSupplementary Fig.\u0026nbsp;1\u003c/b\u003e and \u003cb\u003eSupplementary Table\u0026nbsp;2\u003c/b\u003e. We computed sFNC by calculating pairwise Pearson correlations between IC time courses, yielding 1,378 connectivity features per subject (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC). To capture temporal fluctuations, we estimated dFNC using a sliding window approach (see \u003cb\u003eSupplementary Fig.\u0026nbsp;2\u003c/b\u003e). After developing brain age prediction models using wide-brain sFNC (as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD and \u003cb\u003eSupplementary Fig.\u0026nbsp;3A\u003c/b\u003e) and dFNC (as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE and \u003cb\u003eSupplementary Fig.\u0026nbsp;3B\u003c/b\u003e), we validated their performance on a validation dataset. This step was essential to assess the models\u0026rsquo; generalizability and accuracy in predicting brain age. We calculated the Pearson correlation coefficients between the chronological brain age and the expected brain age to evaluate the models\u0026rsquo; performance. Results demonstrated a strong correlation between the predicted brain age from the sFNC-based model and chronological age, with a correlation value of 0.8755 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). Similarly, the dFNC-based model also showed a strong correlation, with a correlation value of 0.8675, illustrating its effectiveness in age prediction (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\n\u003ch3\u003eStatic and dynamic FNC-based wide-brain age gap links with attention vigilance\u003c/h3\u003e\n\u003cp\u003eAfter developing and validating the brain age prediction models, we applied them to the FBIRN dataset to calculate wide-brain BAGs (wBAGs), representing brain age gaps estimated from wide-brain sFNC and dFNC for everyone. We then constructed generalized linear model (GLMs) to examine the association between wBAGs and attention vigilance while adjusting for age, sex, study site, age\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e, the age-by-sex interaction, and diagnosis as covariates. The corresponding p-values were further adjusted to account for testing two modalities (i.e., sFNC and dFNC features). Because our age prediction model performs less accurately for individuals under 38 years old (see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), we restricted the FBIRN analysis to participants over 38. This resulted in a final sample of 153 participants (121 males) with a mean age of 47.73\u0026thinsp;\u0026plusmn;\u0026thinsp;5.94 years. Figure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA shows a significant negative association between sFNC-based wBAGs and attention vigilance (\u003cem\u003er\u003c/em\u003e =-0.2923, \u003cem\u003eβ\u003c/em\u003e = -0.9860, \u003cem\u003eSE\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.2829, \u003cem\u003e95% CI\u003c/em\u003e: -1.5454 to -0.4263, \u003cem\u003eFDR p\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0013, \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;153). Figure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB demonstrates a significant association between dFNC-based wBAGs and attention vigilance (\u003cem\u003er\u003c/em\u003e = -0.2715, \u003cem\u003eβ\u003c/em\u003e=-0.4395, \u003cem\u003eS\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.1366, \u003cem\u003e95% CI\u003c/em\u003e: -0.7099 to -0.1691, \u003cem\u003eFDR p\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0016, \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;153). In both models, we observed a negative association between wBAGs and cognitive performance, indicating that individuals with older-appearing brains tend to perform worse on attention vigilance tasks.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\n\u003ch3\u003eStatic and dynamic FNC-based wide-brain age gap links with working memory\u003c/h3\u003e\n\u003cp\u003eWe constructed GLMs to examine the association between sFNC- and dFNC-based wBAGs and working memory while adjusting for age, sex, study site, age\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e, age-by-sex interaction, and diagnosis as covariates. P-values were further corrected for multiple comparisons across the two modalities. Figure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC shows a significant negative association between sFNC-based wBAGs and working memory (\u003cem\u003er\u003c/em\u003e=-0.2237, \u003cem\u003eβ\u003c/em\u003e=-0.9603, \u003cem\u003eSE\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.3615, \u003cem\u003e95% CI\u003c/em\u003e=-0.2918 to -0.0856, \u003cem\u003eFDR p\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0088, \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;153). Figure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD demonstrates a significant negative association between dFNC-based wBAGs and working memory (\u003cem\u003er\u003c/em\u003e=-0.2508, \u003cem\u003eβ\u003c/em\u003e=-0.5195, \u003cem\u003eSE\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.1732, \u003cem\u003e95% CI\u003c/em\u003e= -0.8621 to -0.1769, \u003cem\u003eFDR p\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0064, \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;153). These negative correlations indicate that individuals with higher brain age relative to chronological age tend to exhibit poorer working memory performance.\u003c/p\u003e\n\u003ch3\u003eSub-network brain age gaps predict attention vigilance and working memory\u003c/h3\u003e\n\u003cp\u003eThe next question was whether sFNC- and dFNC-based sub-network brain age gaps or subBAGs predict attention vigilance, and working memory performance. To address this, we developed brain age prediction models separately for each network using data from our healthy population. Specifically, we created seven distinct brain age prediction models corresponding to seven networks, including SCN, AUC, SMN, VSN, CCN, DMN, and CBN. The correlation between chronological age and predicted brain age for each sub-network is presented in \u003cb\u003eSupplementary Fig.\u0026nbsp;4.\u003c/b\u003e Next, we examined the association between subBAGs and attention vigilance, and working memory, controlling for age, sex, site, age\u0026sup2;, age-by-sex interaction, and diagnosis as covariates. Given seven networks and two modalities (sFNC and dFNC), a total of 14 statistical tests were performed, and p-values were corrected accordingly.\u003c/p\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA illustrates the relationship between sFNC- and dFNC subBAGs and attention vigilance. Each bar represents the association between subBAGs and attention vigilance, with crosshatched bars indicating dFNC-based subBAGs and solid bars representing sFNC-based subBAGs. Bars showing a significant association after FDR correction are marked with double asterisks. As shown, the sFNC-based subBAGs estimated from the subcortical network exhibit a negative association with attention vigilance (\u003cem\u003er\u003c/em\u003e = -0.2357, \u003cem\u003eβ\u003c/em\u003e = -0.1813, \u003cem\u003eSE\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0655, \u003cem\u003e95% CI\u003c/em\u003e: -0.3111 to -0.0516, \u003cem\u003eFDR p\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0228, \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;153). Similarly, the dFNC-based subBAGs from the subcortical network show a significant negative association with attention vigilance (\u003cem\u003er\u003c/em\u003e=-0.2806, \u003cem\u003eβ\u003c/em\u003e=-0.2989, \u003cem\u003eSE\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0896, \u003cem\u003e95% CI\u003c/em\u003e: -0.4763 to -0.1215, \u003cem\u003eFDR p\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0140, \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;153). Additionally, dFNC-based subBAGs estimated from the sensorimotor network show a significant negative association (\u003cem\u003er\u003c/em\u003e = -0.2384, \u003cem\u003eβ\u003c/em\u003e=-0.2153, \u003cem\u003eSE\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0769, \u003cem\u003e95% CI\u003c/em\u003e: -0.3675 to -0.0631, \u003cem\u003eFDR p\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0228, \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;153) and from the default mode network (\u003cem\u003er\u003c/em\u003e =-0.2666, \u003cem\u003eβ\u003c/em\u003e =-0.1053, \u003cem\u003eSE\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0333, \u003cem\u003e95% CI\u003c/em\u003e: -0.1713 to -0.0392, \u003cem\u003eFDR p\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0140, \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;153). Our results also indicate a significant negative association between sFNC-based subBAGs estimated from the cognitive control network and attention vigilance (\u003cem\u003er\u003c/em\u003e=-0.2081, \u003cem\u003eβ\u003c/em\u003e=-0.0353, \u003cem\u003eSE\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0145, \u003cem\u003e95% CI\u003c/em\u003e: -0.0641 to -0.0065, \u003cem\u003eFDR p\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0455, \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;153). \u003cb\u003eSupplementary Table\u0026nbsp;3\u003c/b\u003e presents the complete set of models examining the association between subBAGs and attention vigilance.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB shows the link between sFNC- and dFNC-based subBAGs with working memory using GLM with age, sex, site, age\u0026sup2;, age-by-sex interaction, and diagnosis as covariates. Among all models, we identified only three showing a significant association between subBAGs and working memory (as shown with single asterisk): sFNC-based subBAGs estimated from the subcortical network (\u003cem\u003er\u003c/em\u003e=-0.1724, \u003cem\u003eβ\u003c/em\u003e = -0.1050, \u003cem\u003eSE\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0518, \u003cem\u003e95% CI\u003c/em\u003e: -0.2076 to -0.0020, \u003cem\u003euncorrected p\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0447, \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;153), dFNC-based subBAGs estimated from the subcortical network (\u003cem\u003er\u003c/em\u003e =-0.1754, \u003cem\u003eβ\u003c/em\u003e=-0.1473, \u003cem\u003eSE\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0714, \u003cem\u003e95% CI\u003c/em\u003e: -0.2885 to -0.0060, \u003cem\u003euncorrected p\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0410, \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;153), and dFNC-based subBAGs estimated from the default mode network (\u003cem\u003er\u003c/em\u003e =-0.2114, \u003cem\u003eβ\u003c/em\u003e =-0.0662, \u003cem\u003eSE\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0264, \u003cem\u003e95% CI\u003c/em\u003e: -0.1186 to -0.0139, \u003cem\u003euncorrected p\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0134, \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;153). However, none of these associations remained significant after FDR correction. \u003cb\u003eSupplementary Table\u0026nbsp;4\u003c/b\u003e presents the complete set of models examining the association between subBAGs and working memory.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe brain-age gap (BAG), the difference between a person\u0026rsquo;s predicted brain age and chronological age, has emerged as a compact indicator of neurobiological health: a larger positive BAG signals accelerated ageing, whereas a negative BAG reflects resilience. Schizophrenia reliably shows an elevated BAG, and larger BAGs link with lower working-memory performance \u003csup\u003e\u003cspan additionalcitationids=\"CR11\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Crucially, almost all BAG work to date relies on structural MRI or static FNC, overlooking the millisecond-scale reconfiguration of brain networks. We present the first dynamic FNC brain age models\u0026mdash;at both wide brain (wBAG) and network-specific (subBAG) levels\u0026mdash;leveraging time-resolved connectivity to capture the neural volatility that likely underlies cognitive dysfunction. By testing dFNC-BAG against working memory and attention vigilance performance in the schizophrenia cohort, we aim to uncover a previously inaccessible layer of pathophysiology and position dFNC-BAG as a new precision biomarker capable of guiding circuit-based interventions and accelerating translational psychiatry.\u003c/p\u003e\u003cp\u003eExtant research shows that sustained attention declines with normal aging\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e, and schizophrenia magnifies this decline, with patients repeatedly scoring well below normative thresholds across sites, ages, and sexes\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e. Our data reveal a robust negative association between attention vigilance measure and both wBAG and subBAG indices of brain age acceleration, whether derived from static or dynamic FNC. Put simply, the \u0026ldquo;older\u0026rdquo; the brain looks, the poorer the patient performs. This powerful link spotlights accelerated macro- and micro-scale brain aging as a driving force behind attentional breakdown in schizophrenia and elevates BAG to a precision and network-level biomarker with clear therapeutic implications. Convergent neuroimaging work has already mapped attentional control to frontoparietal and default mode circuitry; our findings now tie those circuits\u0026rsquo; temporal dysfunction directly to real-world cognitive deficits. Neuroimaging studies have localized specific brain regions and connectivity networks that facilitate attentional processes. For instance, processing temporal stimuli has been linked to activity in the pre-supplementary motor area\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eIn contrast, joint attention engages the ventromedial frontal cortex, cingulate cortex, caudate nuclei, and left superior frontal gyrus\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. Mirroring these maps, our strongest BAG-attention associations localize to sensorimotor and subcortical networks, notably the caudate, thalamus, and putamen, underscoring their central role in the disorder\u0026rsquo;s cognitive burden. Overall, this work bridges neuroanatomical change with clinical symptomatology and supports BAG as a precision tool for early identification and stratification in neuropsychiatric disorders.\u003c/p\u003e\u003cp\u003eWorking memory deficits, spanning visuospatial, phonological, and executive domains, are a core cognitive feature of schizophrenia, consistently observed across cohorts relative to healthy controls\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e. These deficits are compounded by normative age-related decline, with working memory performance deteriorating progressively across adulthood\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e. Building on this foundation, we examined the relationship between BAG and working memory using both sFNC and dFNC models. Both sFNC- and dFNC-based wBAG showed robust negative associations with working memory performance in the FBIRN cohort. In other words, greater brain age acceleration predicted worse working memory capacity. This finding extends prior work by linking functional BAG to cognition in schizophrenia, reinforcing the relevance of BAG as a mechanistic marker of cognitive vulnerability. Notably, our results highlight the subcortical and default mode networks as key contributors to this relationship, pointing to specific neural systems where accelerated ageing most profoundly impacts working memory function.\u003c/p\u003e\u003cp\u003eNotably, the association between wBAGs and working memory, and subBAGs and attention vigilance, was stronger for dFNC-derived BAG, suggesting cognitive deficits in schizophrenia may be more closely linked to dynamic reconfiguration of brain networks than to static connectivity patterns. This aligns with evidence that cognition depends on flexible, moment-to-moment network adaptations to meet cognitive demands\u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e. By capturing temporal fluctuations in connectivity, the dFNC approach provides a richer, more ecologically valid representation of brain function, making it more sensitive to the cognitive disruptions observed in schizophrenia. In contrast, sFNC averages connectivity over time, potentially obscuring critical dynamic alterations. The higher temporal resolution of dFNC also yields more data points per scan, enhancing the detection of subtle brain\u0026ndash;behavior associations\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. These findings highlight the unique potential of dFNC-based BAG as a precision biomarker to capture clinically meaningful cognitive decline and inform individualized therapeutic strategies in schizophrenia.\u003c/p\u003e\u003cp\u003eOur study has several limitations that merit consideration. Firstly, the choice of window size in dynamic connectivity analysis inherently assumes specific characteristics about temporal dynamics. Shorter windows capture rapid fluctuations effectively, while longer windows tend to smooth these fluctuations, potentially obscuring meaningful variability. Future research should explore a broader range of window sizes to more comprehensively assess their impact on brain connectivity measures\u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e. Additionally, our study excluded participants with any health conditions from the UKBB dataset. Future studies may benefit from including a broader cohort by aggregating data from multiple datasets that include healthy individuals, which could enhance the generalizability and robustness of the findings. Moreover, prior research suggests that factors like increased income or engagement in cognitively stimulating activities can mitigate age-related declines in working memory. However, our model did not account for variables such as average activity levels, diet, or other environmental factors that could also influence cognitive and brain function\u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e,\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e. Adjusting both brain and cognitive measures for these variables in future studies could enhance our understanding of the complex interactions between lifestyle, health conditions, and cognitive aging, providing a more nuanced perspective on these relationships. Finally, due to deviations observed in our brain-age predictions for younger adults, likely driven by the limited representation of this age group in our healthy training dataset, we excluded FBIRN participants under 38. Expanding brain-age modeling efforts to include more younger adults from diverse datasets is an essential next step, enabling exploration of whether BAG meaningfully captures cognitive deficits earlier in life and across the whole adult lifespan.\u003c/p\u003e\u003cp\u003eIn conclusion, our study establishes BAG, derived from both static and dynamic functional connectivity at wide-brain and subnetwork levels, as a powerful and clinically relevant marker of cognitive deficits in schizophrenia. We demonstrate that higher BAG, reflecting accelerated brain aging, is significantly associated with worse working memory and attention vigilance, capturing core cognitive dysfunctions of the disorder. Notably, the dynamic approach provides unique sensitivity by incorporating temporal fluctuations in connectivity, offering more profound insight into the brain\u0026rsquo;s functional organization than traditional static methods. Our findings highlight dFNC-based BAG as especially informative, uncovering stronger links to the neural substrates of cognitive decline. These results not only advance mechanistic understanding but also open new avenues for clinical translation. Future research expanding dynamic modeling and integrating lifestyle and health factors holds tremendous promise for optimizing prediction models. Overall, this work lays the foundation for next-generation precision psychiatry tools capable of early detection and personalized intervention strategies to mitigate cognitive deterioration and improve long-term outcomes in schizophrenia.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eInclusion and Ethics Statement\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u003cbr\u003e\u003c/strong\u003eThe Northwest Multicenter Research Ethics Committee granted ethical approval for the UK Biobank (reference 11/NW/0382), and the present study was conducted under UK Biobank application number 34175. The Human Connectome Project Young Adult (HCP-YA) and Aging (HCP-A) datasets were approved by the Washington University Institutional Review Board, and all participants provided written informed consent, including consent to share de-identified data. The FBIRN study received institutional review board (IRB) approval at each participating site, including the University of California (Irvine, Los Angeles, San Francisco), Duke University/University of North Carolina at Chapel Hill, the University of Iowa, the University of Minnesota, and the University of New Mexico. All participants in the FBIRN study provided written informed consent. Diagnoses for individuals with schizophrenia were confirmed using the Structured Clinical Interview for DSM-IV (SCID), and healthy controls were screened using SCID-I/NP to ensure the absence of psychiatric or neurological disorders.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eStudy Population\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u003cbr\u003e\u003c/strong\u003eOur study leveraged multiple datasets, including the UK Biobank (UKBB\u003csup\u003e38,39\u003c/sup\u003e and the Human Connectome Project\u0026rsquo;s Young Adult (HCP-YA\u003csup\u003e40\u003c/sup\u003e\u0026nbsp; and Aging (HCP-A\u003csup\u003e42\u003c/sup\u003e cohorts, to develop brain age prediction models. For the UKBB dataset, exclusion criteria encompassed a range of mental and behavioral disorders as categorized by the International Classification of Diseases version 10 (ICD-10): delirium not induced by alcohol and other psychoactive substances (F05); mental disorders due to brain damage, dysfunction, or physical disease (F06); personality and behavioral disorders due to brain disease, damage, or dysfunction (F07); unspecified organic or symptomatic mental disorders (F09); disorders due to psychoactive substance use (F10-F19); schizophrenia, schizotypal, and delusional disorders (F20-29); manic episodes (F30); and bipolar affective disorder (F31). We also excluded individuals who had sought treatment for \u0026ldquo;nerves, anxiety, tension, or depression\u0026rdquo; from either their general practitioner or a psychiatrist. After exclusions, the UKBB dataset comprised 15,978 samples aged 45 to 82 years, with a mean age of 64.26 \u0026plusmn; 7.55 years and a median age of 65 years, including 7,244 females and 8,734 males\u003csup\u003e38,39\u003c/sup\u003e. All UKBB participants provide informed consent as part of the ethical oversight maintained by a dedicated Ethics and Guidance Council, which collaborates with UKBB to uphold an Ethics and Governance Framework. Additionally, the study received IRB approval from the Northwest Multicenter Research Ethics Committee.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe HCP-YA dataset included 833 individuals aged 22 to 35 years, with a mean age of 28.66 \u0026plusmn; 3.66 years and a median age of 29 years, consisting of 443 females and 390 males\u003csup\u003e41\u003c/sup\u003e. The HCP-A dataset included 711 individuals aged 36 to 100 years, with a mean age of 60.54 \u0026plusmn; 15.68 years and a median age of 58.67 years, consisting of 403 females and 308 males\u003csup\u003e42\u003c/sup\u003e. Collectively, the datasets encompassed 17,522 individuals aged 22 to 100 years, with a mean age of 62.42 \u0026plusmn; 10.96 years and a median age of 64 years, including 8,090 females and 9,432 males. All subject recruitment procedures and informed consent forms, including consent to share de-identified data, were approved by the Washington University IRB. Notably, the HCP datasets provided around 15 minutes of resting-state fMRI data per session, while the UKBB dataset included around 6 minutes of resting-state fMRI data. Because these datasets contain multiple scans per participant, there are a total of 22,569 distinct resting-state fMRI scans in the training set (see\u0026nbsp;\u003cstrong\u003eSupplementary Figure 5\u003c/strong\u003e). \u003cstrong\u003eTable 1\u003c/strong\u003e shows the demographic information of both the training and test datasets.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAdditionally, data from the Functional Imaging Biomedical Informatics Research Network (FBIRN) were employed to estimate the brain age gap (BAG) and examine its link with cognitive metrics such as working memory and attention\u003csup\u003e43\u003c/sup\u003e. For the FBIRN dataset, raw imaging data were collected across seven sites: University of California, Irvine; University of California, Los Angeles; University of California, San Francisco; Duke University/University of North Carolina at Chapel Hill; the University of Iowa; the University of Minnesota; and the University of New Mexico. Each participant provided written informed consent, with protocols approved by the institutional review boards at each respective site. All subjects with schizophrenia (SZ) were clinically stable at the time of scanning. Diagnoses were validated through the Structured Clinical Interview for DSM-IV (SCID-IV), while healthy controls (HC) were assessed using the SCID-I/NP to confirm the absence of schizophrenia. Exclusion criteria for HC included a current or history of major neurological or psychiatric disorders and having a first-degree relative with an Axis-I psychotic disorder, as determined by SCID evaluations. The FBIRN dataset contains individuals between 18 and 62 years old, with a mean age of 37.88 and a median age of 38. Out of 311 individuals in the set, 151 of them have a diagnosis of schizophrenia; these participants are between 18 and 62 years old, with a mean age of 38.77 and a median age of 39. Overall, there were 81 female participants and 230 male participants; among those diagnosed with schizophrenia, there were 36 female participants and 115 male participants. For the FBIRN dataset, written informed consent was obtained from all participants. Institutional review boards approved the consent process of each study site. \u003cstrong\u003eTable 1\u003c/strong\u003e presents the demographic information for all FBIRN participants, while \u003cstrong\u003eSupplementary Table 1\u003c/strong\u003e details the demographic and clinical data for each site separately. \u003cstrong\u003eFigure 1A\u003c/strong\u003e illustrates the count density of the training datasets (UKBB, HCP-YA, and HCP-A), while \u003cstrong\u003eFigure 1B\u003c/strong\u003e displays the count density of the test dataset (FBIRN) used in our study. It is worth noting that in our research, we only focused on participants older than 38 years, since our brain age prediction model was not very accurate in the younger group (see\u0026nbsp;\u003cstrong\u003eSupplementary Figure 6\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eImaging Protocol\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u003cbr\u003e\u003c/strong\u003eFor UKBB, the imaging data were collected using Siemens Skyra 3T scanners. The resolution was 2.4 \u0026times; 2.4 \u0026times; 2.4 mm\u0026sup3;. The data was collected as 490 timeframes over 6 minutes, with a TR of 0.735 s, a TE of 39ms, and a flip angle of 52\u0026deg;. During the scan, participants are instructed to fixate on a cross displayed on a screen to minimize eye movement and maximize data consistency. For HCP, the imaging data were collected using a Siemens 3T scanner as 1200 frames per run over 14:33 minutes per run, with four runs total. The TR was 720 ms, the TE was 33.1, the flip angle was 52\u0026deg;, and the slice thickness was 2.0mm.\u0026nbsp;\u003cstrong\u003e\u003cbr\u003e\u0026nbsp;\u003c/strong\u003e\u0026nbsp;\u003cbr\u003e\u0026nbsp;For the HCP-YA, imaging data were captured using a 3T Siemens Prisma scanner equipped with a 32-channel phased-array head coil. The rs-fMRI acquisition parameters included a TR of 720 ms, a TE of 33.1 ms, and a field of view of 208 \u0026times; 180 mm\u0026sup2;. A flip angle of 52\u0026deg; was used, with images obtained across 72 oblique-axial slices at a resolution of 2.0 \u0026times; 2.0 \u0026times; 2.0 mm\u0026sup3;. Each scanning session was conducted for approximately 14 minutes and 40 seconds, and the HCP-A imaging data were collected using a 3T Siemens Prisma scanner equipped with a 32-channel phased-array head coil. The rs-fMRI protocol included the following parameters: TR of 800 ms, TE of 37 ms, a field of view of 810 \u0026times; 936 mm\u0026sup2;, and a flip angle of 52\u0026deg;. Images were acquired with a resolution of 2.0 \u0026times; 2.0 \u0026times; 2.0 mm\u0026sup3; across 72 oblique-axial slices. Each scanning session lasted approximately 14 minutes and 40 seconds. Similar to the UKBB, participants in the HCP are asked to fixate on a cross presented on a screen during the scan to reduce eye movements and ensure consistent data collection.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;For FBIRN, imaging data were acquired using six Siemens 3T scanners and one General Electric 3T scanner. All sites followed a uniform rs-fMRI protocol. T2*-weighted functional images were captured using an echo-planar imaging sequence, aligned along the anterior and posterior commissure (AC-PC) line, with the following parameters: TE of 30 ms, TR of 2 s, flip angle of 77\u0026deg;, slice gap of 1 mm, voxel dimensions of 3.4 \u0026times; 3.4 \u0026times; 4 mm\u0026sup3;, and a series of 162 frames spanning 5 minutes and 38 seconds. During scans, participants were instructed to keep their eyes closed.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eData Processing\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u003cbr\u003e\u003c/strong\u003eData from fMRI were preprocessed using Statistical Parametric Mapping (SPM12, https://www.fil.ion.ucl.ac.uk/spm/) in the MATLAB 2019 environment. We conducted a rigid body motion correction with SPM\u0026rsquo;s toolbox to address head motion. Subsequently, the imaging data were spatially normalized to an echo-planar imaging (EPI) template in standard Montreal Neurological Institute (MNI) space. Finally, a Gaussian kernel with a full width at half maximum (FWHM) of 6 mm was applied to smooth the fMRI images.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eExtracting independent components using NeuroMark\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u003cbr\u003e\u003c/strong\u003eTo obtain reliable independent components (ICs), we employed the NeuroMark fully automated ICA pipeline, which integrates previously derived component maps as spatial constraints\u003csup\u003e44\u003c/sup\u003e. The NeuroMark framework utilizes templates developed from substantial datasets, specifically the Human Connectome Project (HCP: https://www.humanconnectome.org/study/hcp-young-adult/document/1200-subjects-data-release, 823 subjects after the subject selection) and the Genomics Superstruct Project (GSP: https://dataverse.harvard.edu/dataverse/GSP, 1005 subjects post-selection). This approach has proven effective across numerous studies, identifying a broad array of imaging markers for various brain disorders. Further information on template development is available in our prior publication on the NeuroMark method\u003csup\u003e44\u003c/sup\u003e. The NeuroMark template includes 53 independent components (ICs), categorized into seven functional networks: subcortical (SCN), auditory (AUD), sensorimotor (SMN), visual (VSN), cognitive control (CCN), default-mode (DMN), and cerebellar networks (CBN), as shown in \u003cstrong\u003eFigure 1C\u003c/strong\u003e and \u003cstrong\u003eSupplementary Figure 1\u003c/strong\u003e. \u003cstrong\u003eSupplementary Table 2\u003c/strong\u003e also shows all 53 ICs and their interactions. For the ICA analysis, we utilized these templates via the NeuroMark_fMRI_1.0 template, accessible through GIFT v4.0.5.14 GIFT (http://trendscenter.org/software/gift and on the TReNDS website @ http://trendscenter.org/data). Additional denoising and artifact removal steps prior to calculating dynamic functional connectivity included: 1) linear, quadratic, and cubic de-trending; 2) multiple regression of the six realignment parameters and their temporal derivatives; 3) outlier removal; and 4) low-pass filtering below a frequency of 0.15 Hz.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eStatic functional network connectivity\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u003cbr\u003e\u003c/strong\u003eTo estimate sFNC, we calculated the Pearson correlation between pairs of ICs in each subject as shown in equation 1\u003c/p\u003e\n\u003cp\u003e\u003cimg width=\"217\" height=\"49\" src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; (1)\u003c/p\u003e\n\u003cp\u003ewhere\u0026nbsp;\u003cimg width=\"16\" height=\"20\" src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u0026nbsp;and\u0026nbsp;\u003cimg width=\"16\" height=\"20\" src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u0026nbsp;are time course signals and\u0026nbsp;\u003cimg width=\"14\" height=\"20\" src=\"data:image/png;base64,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\" alt=\"image\"\u003eand\u0026nbsp;\u003cimg width=\"15\" height=\"20\" src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u0026nbsp;are the mean of\u0026nbsp;\u003cimg width=\"19\" height=\"20\" src=\"data:image/png;base64,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\" alt=\"image\"\u003eand\u0026nbsp;\u003cimg width=\"16\" height=\"20\" src=\"data:image/png;base64,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\" alt=\"image\"\u003e, respectively. This Pearson correlation takes values in the interval [\u0026minus; 1, 1] and measures the strength of the linear relationship between\u0026nbsp;\u003cimg width=\"19\" height=\"20\" src=\"data:image/png;base64,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\" alt=\"image\"\u003eand\u0026nbsp;\u003cimg width=\"16\" height=\"20\" src=\"data:image/png;base64,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\" alt=\"image\"\u003e. Each FNC is a 53\u0026times;53 matrix, from which we derived a total of\u0026nbsp;\u003cimg width=\"25\" height=\"24\" src=\"data:image/png;base64,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\" alt=\"image\"\u003e=1378 connectivity features. \u003cstrong\u003eFigure 1C\u003c/strong\u003e shows a representative FNC we used in our study.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eDynamic functional network connectivity\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo compute dynamic functional network connectivity (dFNC), we employed a sliding window approach. We used a tapered window, formed by convolving a rectangular window with the duration of 20TR (UKBB: 14.7 s, HCP-YA: 14.4 s, HCP-A: 16 s, FBIRN: 40s) with a Gaussian kernel (\u0026sigma; = 3), to precisely focus on data at each time point (see \u003cstrong\u003eSupplementary Figure 2\u003c/strong\u003e) \u003csup\u003e32,33,35,36,55\u0026ndash;57\u003c/sup\u003e. This approach, detailed in Equation 1, was utilized to compute the FNC at each time point. We then aggregated the dFNC estimates for each window and each subject to construct a three-dimensional array (C \u0026times; C \u0026times; T), where C represents the 53 independent components, and T represents the number of windows. This array captures the temporal variations in connectivity among the independent components.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eTraining the model\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn our study, the resting-state fMRI time series were uniformly downsampled to the shortest duration across datasets, corresponding to approximately 5.4 minutes. We then computed dFNC matrices using a sliding window approach with a window size of 40 TRs. To evaluate model stability, K=5-fold cross-validation was performed on the training set of 22,569 scans from the UKB, HCP-YA, and HCP-A data sets. For sFNC, we deployed a brain connectivity graph convolutional network, or BCGCN, to predict the brain age\u003csup\u003e58,59\u003c/sup\u003e . The BCGCN model utilized ReLU activation for all layers except the output, which employed a linear activation (see \u003cstrong\u003eSupplementary Figure 3A\u003c/strong\u003e). For the analysis of the dFNC data, we employed a bi-directional Long Short-Term Memory network (biLSTM)\u003csup\u003e60\u003c/sup\u003e configured with three recurrent layers with 128 hidden units each, a dropout rate of 0.1, and a fully connected regression layer to predict brain age (see \u003cstrong\u003eSupplementary Figure 3B\u003c/strong\u003e). In both models, we utilized an Adam optimizer targeting Mean Absolute Error, with a learning rate of 1e\u003csup\u003e-3\u003c/sup\u003e and a batch size of 64, conducting training over 100 epochs. The epoch with the optimal cross-validation performance was selected for inference on the test set. \u003cstrong\u003eFigure 1D\u003c/strong\u003e and \u003cstrong\u003eFigure 1E\u003c/strong\u003e illustrate the modeling and testing procedures for the sFNC and dFNC features, respectively.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003cem\u003eStatistical analysis\u003c/em\u003e\u003c/strong\u003e\u003cbr\u003eWe developed brain age prediction models incorporating both sFNC and dFNC features. These models were then applied to the FBIRN dataset to estimate the predicted brain age and subsequently the BAG for each participant. To investigate the associations between these BAGs and cognitive measures, including working memory and attention vigilance, we constructed a General Linear Model (GLM). This model included age, sex, site, age\u003csup\u003e2\u003c/sup\u003e, the interaction of age and sex, and the diagnosis (schizophrenia or control) as covariates.\u0026nbsp;For each cognitive measure, we tested two hypotheses\u0026mdash;assessing the association of BAG derived from two modalities (sFNC-based and dFNC-based BAGs). We corrected for multiple comparisons across the two modalities using the False Discovery Rate (FDR) correction method to ensure the robustness of our findings\u003csup\u003e61\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003ch2\u003eDisclosures\u003c/h2\u003e\u003cp\u003eDr. Sendi has served as a consultant for Niji Corp for unrelated work. Dr. Mathalon has served as a consultant for Aptinyx, Boehringer-Ingelheim Pharmaceuticals, Cadent Therapeutics, and Greenwich Biosciences for unrelated work. The remaining authors declare no competing interests.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding /Support\u003c/p\u003e\u003cp\u003eThis work was supported by the National Institutes of Health (NIH) grants R01MH123610 and T32MH125786, the National Science Foundation (NSF) grant 2112455, and the Phyllis and Jerome Lyle Rappaport Mental Health Research Scholars Award.\u003c/p\u003e\u003ch2\u003eAuthor Contributions\u003c/h2\u003e\u003cp\u003eSabrina J. Edwards-Swart, Vince D. Calhoun, and Mohammad S. E. Sendi designed the study.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e\u003cp\u003eWe thank the participants of this study and those involved in data collection.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe data utilized in the preparation of this manuscript are publicly available. The UK Biobank (UKBB) data are available via application at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ukbiobank.ac.uk\u003c/span\u003e\u003cspan address=\"https://www.ukbiobank.ac.uk\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. The Human Connectome Project Young Adult (HCP-YA) and Aging (HCP-A) datasets are available through the Connectome Coordination Facility at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.humanconnectome.org\u003c/span\u003e\u003cspan address=\"https://www.humanconnectome.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. The FBIRN dataset is publicly accessible through the NITRC Image Repository (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.nitrc.org/\u003c/span\u003e\u003cspan address=\"https://www.nitrc.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), under the Functional Biomedical Informatics Research Network (FBIRN) project. All data used in this study were obtained under appropriate data use agreements and institutional review board approvals.\u003c/p\u003e\u003ch2\u003eCode Availability\u003c/h2\u003e\u003cp\u003eThe code used for preprocessing and FNC calculation is available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://trendscenter.org/software/\u003c/span\u003e\u003cspan address=\"https://trendscenter.org/software/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Also, statistical parametric mapping (SPM 12) is available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.fil.ion.ucl.ac.uk/spm/\u003c/span\u003e\u003cspan address=\"https://www.fil.ion.ucl.ac.uk/spm/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. The NeuroMark framework and the NeuroMark template (Neuromark_fMRI_1.0) have been made available and incorporated into the Group ICA Toolbox (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/trendscenter/gift\u003c/span\u003e\u003cspan address=\"https://github.com/trendscenter/gift\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Users worldwide can now directly download and utilize these resources. We also use this \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.nitrc.org/projects/bnv/\u003c/span\u003e\u003cspan address=\"https://www.nitrc.org/projects/bnv/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e for the brain graph.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eDesai, P. R., Lawson, K. A., Barner, J. C. \u0026amp; Rascati, K. L. Estimating the direct and indirect costs for community-dwelling patients with schizophrenia. \u003cem\u003eJournal of Pharmaceutical Health Services Research\u003c/em\u003e 4, 187\u0026ndash;194 (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKessler, R. C. \u003cem\u003eet al.\u003c/em\u003e The prevalence and correlates of nonaffective psychosis in the National Comorbidity Survey Replication (NCS-R). \u003cem\u003eBiol Psychiatry\u003c/em\u003e 58, 668\u0026ndash;676 (2005).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGuo, J. Y., Ragland, J. D. \u0026amp; Carter, C. S. Memory and cognition in schizophrenia. Molecular Psychiatry vol. 24 633\u0026ndash;642 Preprint at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41380-018-0231-1\u003c/span\u003e\u003cspan address=\"10.1038/s41380-018-0231-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVan Snellenberg, J. X. \u003cem\u003eet al.\u003c/em\u003e Mechanisms of Working Memory Impairment in Schizophrenia. \u003cem\u003eBiol Psychiatry\u003c/em\u003e 80, 617\u0026ndash;626 (2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDutterer, J., Bansal, S., Robinson, B. \u0026amp; Gold, J. M. Sustained attention deficits in schizophrenia: Effect of memory load on the Identical Pairs Continuous Performance Test. \u003cem\u003eSchizophr Res Cogn\u003c/em\u003e 33, (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi, Y. T. et al. Neuroimaging features of cognitive impairments in schizophrenia and major depressive disorder. Therapeutic Advances in Psychopharmacology vol. 14 Preprint at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/20451253241243290\u003c/span\u003e\u003cspan address=\"10.1177/20451253241243290\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGur, R. E. \u0026amp; Gur, R. C. Functional magnetic resonance imaging in schizophrenia. \u003cem\u003eDialogues Clin Neurosci\u003c/em\u003e 12, 333\u0026ndash;343 (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePenad\u0026eacute;s, R., Franck, N., Gonz\u0026aacute;lez-Vallesp\u0026iacute;, L. \u0026amp; Dekerle, M. Neuroimaging Studies of Cognitive Function in Schizophrenia. in Advances in Experimental Medicine and Biology vol. 1118 117\u0026ndash;134 (Springer New York LLC, 2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChang, J. R., Yao, Z. F., Hsieh, S. \u0026amp; Nordling, T. E. M. Age Prediction Using Resting-State Functional MRI. \u003cem\u003eNeuroinformatics\u003c/em\u003e 22, 119\u0026ndash;134 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKoutsouleris, N. \u003cem\u003eet al.\u003c/em\u003e Accelerated brain aging in schizophrenia and beyond: A neuroanatomical marker of psychiatric disorders. \u003cem\u003eSchizophr Bull\u003c/em\u003e 40, 1140\u0026ndash;1153 (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchnack, H. G. \u003cem\u003eet al.\u003c/em\u003e Accelerated brain aging in schizophrenia: A longitudinal pattern recognition study. \u003cem\u003eAmerican Journal of Psychiatry\u003c/em\u003e 173, 607\u0026ndash;616 (2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShahab, S. \u003cem\u003eet al.\u003c/em\u003e Brain structure, cognition, and brain age in schizophrenia, bipolar disorder, and healthy controls. \u003cem\u003eNeuropsychopharmacology\u003c/em\u003e 44, 898\u0026ndash;906 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJoo, Y. \u003cem\u003eet al.\u003c/em\u003e Brain age prediction using combined deep convolutional neural network and multi-layer perceptron algorithms. \u003cem\u003eSci Rep\u003c/em\u003e 13, (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSendi, M. S., Salat, D. H. \u0026amp; Calhoun, V. D. Brain age gap difference between healthy and mild dementia subjects: Functional network connectivity analysis. in 43rd Annual International Conference of the IEEE Engineering in Medicine \u0026amp; Biology Society (IEEE, Virtual Conference, 2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWrigglesworth, J. \u003cem\u003eet al.\u003c/em\u003e Brain-predicted age difference is associated with cognitive processing in later-life. \u003cem\u003eNeurobiol Aging\u003c/em\u003e 109, 195\u0026ndash;203 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAnat\u0026uuml;rk, M. \u003cem\u003eet al.\u003c/em\u003e Prediction of brain age and cognitive age: Quantifying brain and cognitive maintenance in aging. \u003cem\u003eHum Brain Mapp\u003c/em\u003e 42, 1626\u0026ndash;1640 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eElliott, M. L. \u003cem\u003eet al.\u003c/em\u003e Brain-age in midlife is associated with accelerated biological aging and cognitive decline in a longitudinal birth cohort. \u003cem\u003eMol Psychiatry\u003c/em\u003e 26, 3829\u0026ndash;3838 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStefaniak, J. D. \u003cem\u003eet al.\u003c/em\u003e Brain age gap, dementia risk factors and cognition in middle age. \u003cem\u003eBrain Commun\u003c/em\u003e 6, (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShahab, S. \u003cem\u003eet al.\u003c/em\u003e Brain structure, cognition, and brain age in schizophrenia, bipolar disorder, and healthy controls. \u003cem\u003eNeuropsychopharmacology\u003c/em\u003e 44, 898\u0026ndash;906 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKoutsouleris, N. \u003cem\u003eet al.\u003c/em\u003e Accelerated brain aging in schizophrenia and beyond: A neuroanatomical marker of psychiatric disorders. \u003cem\u003eSchizophr Bull\u003c/em\u003e 40, 1140\u0026ndash;1153 (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchnack, H. G. \u003cem\u003eet al.\u003c/em\u003e Accelerated brain aging in schizophrenia: A longitudinal pattern recognition study. \u003cem\u003eAmerican Journal of Psychiatry\u003c/em\u003e 173, 607\u0026ndash;616 (2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHajek, T. \u003cem\u003eet al.\u003c/em\u003e Brain Age in Early Stages of Bipolar Disorders or Schizophrenia. \u003cem\u003eSchizophr Bull\u003c/em\u003e 45, 191\u0026ndash;198 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNenadić, I., Dietzek, M., Langbein, K., Sauer, H. \u0026amp; Gaser, C. BrainAGE score indicates accelerated brain aging in schizophrenia, but not bipolar disorder. \u003cem\u003ePsychiatry Res Neuroimaging\u003c/em\u003e 266, 86\u0026ndash;89 (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang, J. \u003cem\u003eet al.\u003c/em\u003e White matter brain aging in relationship to schizophrenia and its cognitive deficit. \u003cem\u003eSchizophr Res\u003c/em\u003e 230, 9\u0026ndash;16 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhu, J. D., Wu, Y. F., Tsai, S. J., Lin, C. P. \u0026amp; Yang, A. C. Investigating brain aging trajectory deviations in different brain regions of individuals with schizophrenia using multimodal magnetic resonance imaging and brain-age prediction: a multicenter study. \u003cem\u003eTransl Psychiatry\u003c/em\u003e 13, (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMillar, P. R. \u003cem\u003eet al.\u003c/em\u003e Predicting brain age from functional connectivity in symptomatic and preclinical Alzheimer disease. \u003cem\u003eNeuroimage\u003c/em\u003e 256, (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMonti, R. P. \u003cem\u003eet al.\u003c/em\u003e Interpretable brain age prediction using linear latent variable models of functional connectivity. \u003cem\u003ePLoS One\u003c/em\u003e 15, (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLund, M. J. \u003cem\u003eet al.\u003c/em\u003e Brain age prediction using fMRI network coupling in youths and associations with psychiatric symptoms. \u003cem\u003eNeuroimage Clin\u003c/em\u003e 33, (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSakoğlu, \u0026Uuml;. \u003cem\u003eet al.\u003c/em\u003e A method for evaluating dynamic functional network connectivity and task-modulation: Application to schizophrenia. \u003cem\u003eMagnetic Resonance Materials in Physics, Biology and Medicine\u003c/em\u003e 23, 351\u0026ndash;366 (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCalhoun, V. D., Miller, R., Pearlson, G. \u0026amp; Adali, T. The Chronnectome: Time-Varying Connectivity Networks as the Next Frontier in fMRI Data Discovery. \u003cem\u003eNeuron\u003c/em\u003e 84, 262\u0026ndash;274 (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLurie, D. J. et al. Questions and controversies in the study of time-varying functional connectivity in resting fMRI. Network Neuroscience vol. 4 30\u0026ndash;69 Preprint at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1162/netn_a_00116\u003c/span\u003e\u003cspan address=\"10.1162/netn_a_00116\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSendi, M. S. E. \u003cem\u003eet al.\u003c/em\u003e Multiple overlapping dynamic patterns of the visual sensory network in schizophrenia. \u003cem\u003eSchizophr Res\u003c/em\u003e 228, 103\u0026ndash;111 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSendi, M. S. E. \u003cem\u003eet al.\u003c/em\u003e The link between static and dynamic brain functional network connectivity and genetic risk of Alzheimer\u0026rsquo;s disease. \u003cem\u003eNeuroimage Clin\u003c/em\u003e 37, 103363 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSendi, M. S. et al. Brain dynamics reflecting an intra-network brain state is associated with increased posttraumatic stress symptoms in the early aftermath of trauma. Nature Mental Health (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSendi, M. S. E. \u003cem\u003eet al.\u003c/em\u003e Alzheimer\u0026rsquo;s Disease Projection From Normal to Mild Dementia Reflected in Functional Network Connectivity: A Longitudinal Study. \u003cem\u003eFront Neural Circuits\u003c/em\u003e 14, (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDini, H. \u003cem\u003eet al.\u003c/em\u003e Dynamic functional connectivity predicts treatment response to electroconvulsive therapy in major depressive disorder. \u003cem\u003eFront Hum Neurosci\u003c/em\u003e 15, 1\u0026ndash;11 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAllen, E. A. \u003cem\u003eet al.\u003c/em\u003e Tracking whole-brain connectivity dynamics in the resting state. \u003cem\u003eCerebral Cortex\u003c/em\u003e 24, 663\u0026ndash;676 (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMiller, K. L. \u003cem\u003eet al.\u003c/em\u003e Multimodal population brain imaging in the UK Biobank prospective epidemiological study. \u003cem\u003eNat Neurosci\u003c/em\u003e 19, 1523\u0026ndash;1536 (2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLittlejohns, T. J. et al. The UK Biobank imaging enhancement of 100,000 participants: rationale, data collection, management and future directions. Nature Communications vol. 11 Preprint at (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGlasser, M. F. \u003cem\u003eet al.\u003c/em\u003e The Human Connectome Project\u0026rsquo;s neuroimaging approach. \u003cem\u003eNat Neurosci\u003c/em\u003e 19, 1175\u0026ndash;1187 (2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVan Essen, D. C. \u003cem\u003eet al.\u003c/em\u003e The WU-Minn Human Connectome Project: An overview. \u003cem\u003eNeuroimage\u003c/em\u003e 80, 62\u0026ndash;79 (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBookheimer, S. Y. \u003cem\u003eet al.\u003c/em\u003e The Lifespan Human Connectome Project in Aging: An overview. \u003cem\u003eNeuroimage\u003c/em\u003e 185, 335\u0026ndash;348 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKeator, D. B. \u003cem\u003eet al.\u003c/em\u003e The Function Biomedical Informatics Research Network Data Repository. \u003cem\u003eNeuroimage\u003c/em\u003e 124, 1074\u0026ndash;1079 (2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDu, Y. \u003cem\u003eet al.\u003c/em\u003e NeuroMark: An automated and adaptive ICA based pipeline to identify reproducible fMRI markers of brain disorders. \u003cem\u003eNeuroimage Clin\u003c/em\u003e 28, 102375 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFilley, C. M. \u0026amp; Cullum, C. M. Attention and vigilance functions in normal aging. \u003cem\u003eAppl Neuropsychol\u003c/em\u003e 1, 29\u0026ndash;32 (1994).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNuechterlein, K. H. \u003cem\u003eet al.\u003c/em\u003e Attention/vigilance in schizophrenia: Performance results from a large multi-site study of the Consortium on the Genetics of Schizophrenia (COGS). \u003cem\u003eSchizophr Res\u003c/em\u003e 163, 38\u0026ndash;46 (2015).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCoull, J. T. fMRI studies of temporal attention: Allocating attention within, or towards, time. \u003cem\u003eCognitive Brain Research\u003c/em\u003e 21, 216\u0026ndash;226 (2004).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWilliams, J. H. G., Waiter, G. D., Perra, O., Perrett, D. I. \u0026amp; Whiten, A. An fMRI study of joint attention experience. \u003cem\u003eNeuroimage\u003c/em\u003e 25, 133\u0026ndash;140 (2005).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eForbes, N. F., Carrick, L. A., McIntosh, A. M. \u0026amp; Lawrie, S. M. Working memory in schizophrenia: A meta-analysis. Psychological Medicine vol. 39 889\u0026ndash;905 Preprint at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/S0033291708004558\u003c/span\u003e\u003cspan address=\"10.1017/S0033291708004558\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2009).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWingfield, A., Stine, E. A. L., Lahar, C. J. \u0026amp; Aberdeen, J. S. Does the capacity of working memory change with age? \u003cem\u003eExp Aging Res\u003c/em\u003e 14, 103\u0026ndash;107 (1988).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhu, J. \u003cem\u003eet al.\u003c/em\u003e Dynamic functional connectome predicts individual working memory performance across diagnostic categories. \u003cem\u003eNeuroimage Clin\u003c/em\u003e 30, (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFaghiri, A., Thomas, D., Oktay, A. \u0026amp; Calhoun, V. D. A unified approach to study the brain connectivity frequency profile. in International Symposium on Biomedical Imaging (ISBI) (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHsu, H. C. \u0026amp; Bai, C. H. Individual and environmental factors associated with cognitive function in older people: a longitudinal multilevel analysis. \u003cem\u003eBMC Geriatr\u003c/em\u003e 22, (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJeong, H. J. \u003cem\u003eet al.\u003c/em\u003e General and Specific Factors of Environmental Stress and Their Associations With Brain Structure and Dimensions of Psychopathology. \u003cem\u003eBiological Psychiatry Global Open Science\u003c/em\u003e 3, 480\u0026ndash;489 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFu, Z. \u003cem\u003eet al.\u003c/em\u003e Whole-Brain Functional Connectivity Dynamics Associated With Electroconvulsive Therapy Treatment Response. \u003cem\u003eBiol Psychiatry Cogn Neurosci Neuroimaging\u003c/em\u003e 7, 312\u0026ndash;322 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSendi, M. S. E., Salat, D. H., Miller, R. L. \u0026amp; Calhoun, V. D. Two-step clustering-based pipeline for big dynamic functional network connectivity data. \u003cem\u003eFront Neurosci\u003c/em\u003e 16, (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSendi, M. S. E. \u003cem\u003eet al.\u003c/em\u003e Aberrant dynamic functional connectivity of default mode network predicts symptom severity in major depressive disorder. \u003cem\u003eBrain Connect\u003c/em\u003e 11, 838\u0026ndash;849 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYao, D. \u003cem\u003eet al.\u003c/em\u003e A Mutual Multi-Scale Triplet Graph Convolutional Network for Classification of Brain Disorders Using Functional or Structural Connectivity. \u003cem\u003eIEEE Trans Med Imaging\u003c/em\u003e 40, 1279\u0026ndash;1289 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi, Y. \u003cem\u003eet al.\u003c/em\u003e Brain Connectivity Based Graph Convolutional Networks and Its Application to Infant Age Prediction. \u003cem\u003eIEEE Trans Med Imaging\u003c/em\u003e 41, 2764\u0026ndash;2776 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHochreiter, S. \u0026amp; Schmidhuber, J. Long short-term memory. \u003cem\u003eNeural Comput\u003c/em\u003e 9, (1997).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHochberg, Y. \u0026amp; Tamhane, A. C. Multiple Comparison Procedures. (John Wiley and Sons Inc., 1987).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7336363/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7336363/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSchizophrenia is characterized by deficits in attention and working memory. The brain age gap (BAG), the difference between brain-predicted and chronological age, has emerged as a biomarker of brain dysfunction, but its association with dynamic brain function remains unclear. We developed brain age models using static (sFNC) and dynamic (dFNC) functional network connectivity from a large resting-state fMRI dataset (\u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;22,569; UK Biobank, HCP-Young Adult, HCP-Aging) and validated them in an independent schizophrenia cohort (FBIRN; \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;153). Higher BAGs were significantly associated with lower attention and working memory performance (\u003cem\u003eFDR p\u0026thinsp;\u0026lt;\u0026thinsp;0.01\u003c/em\u003e), with dFNC-based models showing more potent effects than sFNC. Network-specific BAGs, particularly within cognitive control, default mode, and subcortical networks, were robust predictors of cognitive impairment. These findings establish dFNC-based BAG as a sensitive biomarker of cognitive dysfunction in schizophrenia and highlight the value of dynamic connectivity measures for advancing precision diagnostics and stratification.\u003c/p\u003e","manuscriptTitle":"Dynamic Brain Age Modeling Identifies Network-Specific Cognitive Deficits in Schizophrenia","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-10 21:13:14","doi":"10.21203/rs.3.rs-7336363/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"a3a51786-58de-4bdf-8f6d-542072e5b2b1","owner":[],"postedDate":"October 10th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":53822947,"name":"Health sciences/Biomarkers/Diagnostic markers"},{"id":53822948,"name":"Biological sciences/Computational biology and bioinformatics/Machine learning"},{"id":53822949,"name":"Biological sciences/Computational biology and bioinformatics/Computational models"}],"tags":[],"updatedAt":"2025-10-10T21:13:14+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-10 21:13:14","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7336363","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7336363","identity":"rs-7336363","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}