Dynamic Rest-to-Task Brain Network Reconfiguration Reveals Divergent Cognitive Changes in Aging

Dynamic Rest-to-Task Brain Network Reconfiguration Reveals Divergent Cognitive Changes in Aging | 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 Rest-to-Task Brain Network Reconfiguration Reveals Divergent Cognitive Changes in Aging Antao Chen, Zijin Liu, Haishuo Xia This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9298914/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Emerging research challenges the prevailing notion of uniform cognitive decline in aging, instead emphasizing the brain's adaptive ability that compensates for age-related deficits. Here we examined age-related differences in brain dynamics by focusing on rest-to-task reconfiguration, a process that reflects the brain’s ability to recruit neural resources for goal-directed behavior. Using hidden Markov modeling of whole-brain fMRI data, we characterized brain state transitions from rest to a Stroop task in young and older adults. Behaviorally, older adults exhibited generalized slowing yet preserved conflict resolution. Dynamically, both age groups could reconfigure from rest-specific towards a task-efficient brain state. However, while young adults preferentially engaged this state, older adults showed attenuated recruitment, instead maintaining a high-effort state from rest to task. Crucially, the functional benefit of this pattern was conditional: it supported conflict resolution only in high-accuracy older adults but presented detrimental in low performers. These findings demonstrate that cognitive preservation in aging depends not on preserving youthful brain reconfiguration dynamics, but on the effective recruitment of compensatory brain states. Biological sciences/Neuroscience/Cognitive ageing Health sciences/Biomarkers/Predictive markers Cognitive Aging Brain State Dynamics Rest-To-Task Reconfiguration Functional Compensation fMRI Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Significance statement Cognitive aging is often described as an inevitable decline, yet many older adults maintain high levels of cognitive performance. Using dynamic modeling of fMRI data, we show that aging alters how the brain reconfigures its states from rest to a Stroop task. Young adults preferentially engage a task-efficient brain state, whereas older adults rely on a high-effort, frontal-dominate state. Critically, this older-specific strategy supports conflict resolution only in high-accuracy older adults but is ineffective in low performers. These findings move beyond simple cognitive decline and demonstrate that cognitive preservation in aging depends on the conditional effectiveness of brain network reconfiguration. Introduction Aging challenges the brain ability to achieve goal-directed behavior, yet findings on the underlying neural mechanisms remain mixed ( 1 – 4 ). While prior research has extensively characterized age-related changes in either resting-state or task-evoked networks separately ( 5 – 9 ), this approach neglects a fundamental property of the brain: the ability to reconfigure network organization( 10 – 15 ) from a self-organized resting mode to a task-optimized state ( 16 – 19 ). Such reconfiguration reflects neural resource mobilization to meet cognitive demands, potentially supporting adaptive behavior ( 20 – 25 ). However, this dynamic reconfiguration remains largely unexplored in current aging literature. Behavioral findings on cognitive aging are inconsistent, ranging from declined to preserved ability( 26 – 31 ). This challenges the notion of uniform age-related cognitive decline and necessitates a deeper exploration of its underlying neural substrates. Two theoretical accounts of cognitive aging offer different interpretations for age-related neural changes. The neural inefficiency suggests that excessive recourse recruitment of aging brain yields no functional benefit under high cognitive demand ( 32 – 34 ). In comparison, the functional compensation proposes that older adults strategically engage alternative brain networks to maintain cognitive performance ( 35 – 37 ). It is yet to be clarified whether age-related alterations in rest-to-task reconfiguration represent efficiency failure or a compensatory strategy. For older adults who maintain cognitive performance, do they rely on distinct reconfiguration patterns that diverge from young adults? Quantifying the dynamic rest-to-task reconfiguration requires moving beyond static analysis. From a dynamical systems perspective, brain activity is modeled as discrete transitions through a state space of recurrent network configurations ( 38 – 43 , 43 ). In this view, task onset perturbs the intrinsic resting dynamics toward task-optimized brain state ( 10 , 20 , 44 – 46 ). Hidden Markov Modeling (HMM) operationalizes this framework by decoding latent brain states and their temporal transitions, thereby yielding a dynamic map of the rest-to-task network reconfiguration (Fig. 1 B). Unlike sliding-window approaches, HMM decodes brain states with high temporal sensitivity ( 38 , 39 , 41 – 43 , 47 – 49 ). Crucially, by fitting a shared state space across age groups and conditions, HMM enables direct quantification of how aging alters the reconfiguration of intrinsic dynamics in response to cognitive demand ( 39 , 49 – 51 ). Within this framework, we pursued four objectives to investigate the age-related changes. First, we used the classic color–word Stroop task as a probe of cognitive demand (Fig. 1 A) and assessed whether behavioral performance in older adults reflects preservation or decline (Fig. 1 C). Second, we computed the state transitions (Fig. 1 B) to test whether young and older adults utilize distinct rest-tot-task reconfiguration pattern (Fig. 1 D). Third, we characterized the spatial signature of each state by combining network activation with functional connectivity. Finally, we examined the functional significance of these dynamics by linking reconfiguration metrics to behavioral outcomes (Fig. 1 E). By shifting focus from static network properties to dynamic reconfiguration, this study aims to clarify whether the age-related changes in rest-to-task reconfiguration pattern represent a uniform decline or a compensatory adaptation essential for sustaining cognition. Results Stroop Performance: Age-Related Decline and Preservation Behavioral performance on the Stroop task was analyzed with linear mixed models (LMMs) for reaction time (RT), accuracy, and the Inverse Efficiency Score (IES; RT/accuracy). Fixed effects were Age Group (younger, older) and Condition (control, conflict); Gender and years of education were included as covariates and showed no significant effects. Accuracy was inverse–square-root transformed for modeling; figures display raw values. Across all three measures, LMMs revealed significant main effects of Age Group and Condition. Older adults had longer RTs and higher IES (lower efficiency) than younger adults in both Control and Conflict conditions (Fig. 2 A, C). For accuracy (Fig. 2 B), younger adults outperformed older adults in the Control condition ( p = 0.02), whereas no age difference was observed in the Conflict condition ( p = 0.86). As expected, both groups exhibited a robust conflict effect from Control to Conflict—RT and IES increased, and accuracy decreased (SI Appendix, Tables S1–S2). Critically, no significant Age Group x Condition interaction was observed for RT ( F (1, 102) = 1.13, p = 0.29) or IES ( F (1, 102) = 0.05, p = 0.83). To directly test age differences in the conflict effect (Conflict − Control), independent Welch’s t tests on difference scores indicated no age-group differences for RT Difference ( t = 1.08, p = 0.28) or IES Difference ( t = 0.24, p = 0.81). Thus, despite general age-related slowing and reduced efficiency, the magnitude of the conflict effect did not differ by age. This pattern suggests that older adults may engage distinct neural strategies to support conflict resolution, motivating the subsequent analyses of age-related differences in rest-to-task reconfiguration. Brain State Temporal Dynamics: Age-related changes from Rest to Task Based on the lowest free energy choice, 4 brain states were generated at the group level, which were shared among young and older adults across all conditions. We analyzed brain state in-degree with ART ANOVAs including Age Group (young, older) and Condition (rest: eyes-open, eyes-closed; task: Stroop control, Stroop conflict) while controlling for gender and years of education; all effects reported are covariate-controlled (Gender and education years). All four states exhibited significant main and interaction effects. All effects are covariate-adjusted for gender and years of education (SI Appendix, Table S3, S5). For S1, a robust Task effect was observed ( F (3,306) = 98.43, p < 0.001, ηp² = 0.49). The Age effect was also significant ( F (1,102) = 10.93, p < 0.001, ηp² = 0.10) accompanied by a significant Age × Task interaction ( F (3,306) = 3.59, p = 0.01, ηp² = 0.03). S2 demonstrated a very large Task modulation ( F (3,306) = 151.57, p < 0.001, ηp² = 0.60), a strong Age effect ( F (1,102) = 16.29, p < 0.001, ηp² = 0.14, and a substantial interaction ( F (3,306) = 10.43, p < 0.001, ηp² = 0.09). Despite having the lowest in-degree, S3 still showed small but significant main effects of Age ( F (1,102) = 4.32, p = 0.04, ηp²=.041), and Task ( F (3,306) = 3.98, p < 0.001, ηp² = 0.04), together with a sizable interaction, F (3,306) = 14.40, p < 0.001, ηp² = 0.12).S3 showed smaller yet significant main effects for Age ( F (1,102) = 4.32, p = 0.04, ηp² = 0.04, and Task ( F (3,306) = 3.98, p = 0.01, ηp² = 0.04, coupled with a sizeable interaction ( F (3,306) = 14.40, p < 0.001, ηp² = 0.12). Finally, S4 presented a large Task effect ( F (3,306) = 134.96, p < 0.001, ηp² = 0.57), a strong Age effect ( F (1,102) = 21.76, p < 0.001, ηp² = 0.18), and a medium interaction ( F (3,306) = 10.71, p < 0.001 ηp² = 0.10). Collectively, Task effects were large for S1, S2, and S4 (ηp² ≈ 0.49–0.60) and smaller for S3. Age effects spanned from small-to-medium (S3) to large (S4), while interactions were presented in all states, averaging medium and approaching medium-to-large in S3. Post hoc analysis showed that within-rest (eyes-open vs. eyes-closed) and within-task (control vs. conflict) comparisons were not significant (SI Appendix, Table S5). To maximize interpretability around the dominant rest–task contrast, we collapsed Task Conditions to eyes-open (rest) versus Stroop conflict (task) for confirmatory ART ANOVA analysis (Age Group: young, older; Conditions: eyes-open, Stroop conflict). Results replicated the main pattern (SI Appendix, Table S4, S6): S1 retained very large Task modulation ( F (1,102) = 154.89, p < 0.001, ηp² = 0.603), with Age ( F (1,102) = 13.74, p < 0.001, ηp² = 0.119), and a medium interaction ( F (1,102) = 9.87, p = 0.002, ηp² = 0.09); S2 showed a very large Task effect ( F (1,102) = 183.36, p < 0.001 ηp² = 0.64), a medium Age effect ( F (1,102) = 12.61, p < 0.001, ηp² = 0.11), and a medium interaction ( F (1,102) = 13.45, p < 0.001, ηp²=0.117); S3’s Task main effect was not significant ( F (1,102) = 0.79, p = 0.376, ηp²= 0.01), but age remained significant ( F (1,102) = 5.74, p = 0.02, ηp² = 0.05), and the interaction was large ( F (1,102) = 20.17, p < 0.001, ηp² = 0.17), indicating age-dependent divergence specifically under the rest-versus-conflict contrast; S4 kept a very large Task effect ( F (1,102) = 136.12, p < 0.001, ηp² = 0.57), a large Age effect ( F (1,102) = 23.19, p < 0.001, ηp²=0.19), and a large interaction ( F (1,102) = 18.94, p < 0.001, ηp² = 0.16). The convergent evidence supports classifying S1 and S4 as rest-specific states, S2 as a task-specific state, and S3 as a transient state (Fig. 3 ) whose dynamics are particularly age-sensitive under conflict demands. As shown by the significant Age × Task interactions ( SI Appendix, Table S4), older adults preserved the direction of the transition (↓S1/S4, ↑S2) but with differed magnitude and uniquely maintaining S3 from rest to task—consistent with the large S3 interaction despite a nonsignificant S3 Task main effect (F(1,102) = 0.79, p = 0.38, ηp² = 0.01). In contrast, young adults shifted more completely into S2 with a reduction of S3 ( Fig. 4 , 7 ). These patterns indicate that, upon task requests, older adults still transition out of rest-specific states but recruit S2 less and retain S3, whereas young adults engage S2 as the main task state. Brain State Spatial Signatures: Distinct Network Connectivity and Activation Patterns Building on the age-specific state temporal dynamics identified previously, we next investigated the spatial profiles of all states by examining their network connectivity (Fig. 5 A) and activation patterns (Fig. 5 B, C). This analysis first revealed a functional connectivity reconfiguration between rest and task conditions. The rest-specific states, S1 and S4, were characterized by cohesive, positive correlations among network nodes and robust intra-network coupling. In contrast, the task-specific states, S2 and S3, prompted a shift toward anti-correlated connections and reduced intra-network connectivity, reflecting the competitive interactions required for goal-directed processing. While functional connectivity distinguished rest-specific and task-specific states, network activation patterns revealed age-specific changes. For the rest-specific states, S1 displayed strong activity across DMN, MF and FP. S4, preferred by older adults, showed increased engagement of MOT and CB regions. A similar opposite activation pattern emerged for the task-specific states. For the young-preferred states, S2, the state most engaged by younger adults, was uniquely characterized by heightened BG activation. Conversely, S3, the state maintained by older adults from rest to task, showed reduced BG involvement but increased FP and MOT activation. These results revealed distinct spatial signatures for each brain state characterized by age- and task-specific temporal dynamics. Brain-behavior Association: S3 in-degree and conflict cost vary with accuracy in older adults Brain-behavior Association: S3 in-degree and conflict cost vary with accuracy in older adults Two linear regression models tested whether S2 and S3 in-degree (rank-transformed), conflict accuracy (logit-transformed), age group and their interactions predicted RT Difference (conflict cost), with accuracy included to address speed–accuracy trade-off. Gender and years of education were included as covariates and showed no significant effects. Only older adults showed an accuracy × in-degree interaction in both models. For S2, the interaction term was nominally significant (b = 0.91, t (96) = 2.28, p = 0.03) but not robust to BCa bootstrapping (95% CI [− 0.04, 1.82]). Given the non-robust CI crossing zero (SI Appendix, Table S8). For S3, the interaction was significant and robust (b = − 1.09, t (96) = − 2.77, p = 0.006; BCa 95% CI [(-2.04, -0.23]; ηp² = 0.08). Simple-slope estimates within older adults (Fig. 6 ) indicated that when accuracy was low (− 1 SD), higher S3 in-degree tended to predict larger RT Difference (b = 0.93, SE = 0.54, p = 0.09); at mean accuracy, the slope was null (b = − 0.05, SE = 0.38, p = 0.90); and when accuracy was high (+ 1 SD), higher S3 in-degree predicted smaller RT Difference (b = − 1.02, SE = 0.50, p = 0.05). No S3 in-degree × accuracy interaction was observed in young adults (SI Appendix, Tables S7-S9). As an exploratory illustration of this interaction in older adults, participants were divided into low- and high-accuracy groups based on median conflict accuracy. Spearman rank correlations suggested qualitatively opposite associations between S3 in-degree and RT difference across the two subgroups (SI Appendix, Fig. S1 ). Together, these results indicate that in older adults the association between S3 in-degree and conflict cost varies across accuracy levels: higher S3 in-degree is linked to smaller costs when accuracy is higher and to larger costs when accuracy is lower. Discussions The present study investigated whether age-related changes in rest-to-task brain state reconfiguration during the color–word Stroop task reflect functional decline or strategic compensation. We find that the aging brain supports goal-directed behavior not through a preservation of youth-like dynamics, but via a strategic reconfiguration of brain states. While older adults exhibited generalized behavioral slowing, they successfully maintained conflict resolution comparable to young adults. By leveraging HMM to decode the temporal evolution of brain states, we demonstrate that this behavioral preservation relies on a qualitative shift in reconfiguration: maintaining an effortful, frontal-dominant state (S3) from rest to task. Crucially, we identify the functional significance of this shift as conditional: the recruitment of S3 serves as a compensatory scaffold only for high-performing older adults but reflects neural inefficiency in those with lower accuracy. Age-related Slowing and Preservation of Conflict Resolution It is often assumed that cognition generally declines in older age, yet aging shows mixed trajectories in which decline, stability, and even growth co-exist ( 52 ). A growing body of behavioral work in cognitive control finds little evidence for a specific decline in conflict resolution with age—across large-sample lifespan analyses and meta-analyses ( 26 , 28 , 30 , 31 ). Notably, research field about attentional selection likewise report preserved focusing efficiency in older adults, arguing against a general decline in cognitive functions( 30 ). Our behavioral results align with this trend: despite generalized slowing, older adults showed no reliable age-group difference in the conflict resolution. While our fMRI Stroop used a block design, which can blur trial-wise dynamics (e.g., conflict adaptation), the preserved conflict resolution with generalized slowing suggests that cognitive maintenance among older adults may be not merely a design artifact but may reflect adaptive changes in how neural resources are recruited ( 53 – 55 ). It is intriguing to ask whether older adults employ distinct neural strategies to maintain effective conflict resolution in the context of generalized slowing. Common and Age-specific Brain state reconfigurations from rest to task Both age groups successfully reconfigured brain dynamics from rest to task, evidenced by the shared reduction in rest-specific states (S1/S4) and increase in a task-specific state (S2). Despite this common directional reconfiguration (↓S1/S4, ↑S2), group differences lay in its magnitude (Fig. 7 ). Specifically, while both groups showed a decrease in rest-specific states (S1/S4), the observed differential decline was largely attributable to higher rest baselines in older adults. Crucially, the task-induced reduction in state in-degrees upon task onset was comparable across age groups, suggesting a maintained capacity for disengaging from rest-specific processing in older adults, rather than a complete degradation. By contrast, although S2 increased in both groups, older adults showed a significantly smaller recruitment magnitude. This reduced S2 magnitude suggests declined efficiency in task-state engagement, rather than a complete loss of the reconfiguration pattern itself. These findings supports the view that healthy aging reflects a gradual change in brain functions rather than abrupt degradation, aligning with lifespan models that conceptualize development as a continuous process of gain and loss ( 56 – 60 ). Beyond these quantitative differences in state reconfiguration, our findings revealed a qualitatively distinct reconfiguration concerning S3 in older adults. While young adults downregulated S3, older adults uniquely maintained this state from rest to task. This implies that S3 is not merely a state to suppress during task but selectively retained in older adults, suggesting that beyond the aging brain may experience a divergent reconfiguration from rest to task. This observation fits with adaptive views of aging—such as the Scaffolding Theory of Aging and Cognition (STAC)—which propose that the aging brain recruits different neural resources to support cognitive function ( 32 , 61 , 62 ). The Shift from Efficient to Effortful Brain State for Task Execution During Aging Young adults preferentially engaged state S2, characterized by prominent BG activation and lower FP involvement. Given the role of the BG in automated processing ( 62 – 64 ), this suggests that young adults rely on a computationally efficient, subcortically-driven mode for task execution. For a relatively simple and well-practiced Stroop task with fMRI block design, this engagement with S2 may represent a highly efficient processing capability, potentially requiring less cognitive effort from young adults. In contrast, older adults recruited S2 to a lesser extent and instead maintained S3, which featured heightened FP and reduced BG involvement. The FP network is central to achieve goal-directed behavior, encompassing cognitive control functions such as working memory, inhibitory control, and cognitive flexibility ( 65 – 68 ). The heightened FP engagement in S3 may reflect increased cognitive effort or higher cortical arousal for task execution ( 66 , 69 – 71 ). This shift from a BG-dominate state to an FP-dominant state aligns with the Compensatory-Related Utilization of Neural Circuits Hypothesis (CRUNCH), suggesting an increased reliance on additional neural resources, particularly in frontal regions, to compensate for age-related neural inefficiencies and maintain cognitive performance ( 32 , 33 , 72 ). The contrasting activation of BG and FP networks of S2 and S3, underscore age-related differences in brain network recruitment and cognitive resource allocation during goal-directed processing. Conditional Functional Significance of The Older-specific State Reconfiguration The age-related decline in S2 and maintenance of S3 from rest to task s admits two possibilities. One hand, it may reflect a diminished capacity to engage automatic processing state, forcing older adults into S3 required of higher cognitive cost or arousal level, reflecting neural inefficiency ( 32 – 34 ). On the other hand, S2 alone proves insufficient for heightened task demands in aging, prompting the active recruitment of a state with greater neural effort to maintain function ( 72 – 74 ), reflecting functional compensation ( 35 – 37 ). The key distinction lies in whether the S3 engagement yield functional benefits for older adults to improve or sustain task performance. Our results suggest that neural inefficiency and functional compensation are not mutually exclusive phenomena but reflect individual differences in the efficacy of neural recruitment. S3's role in older adults is intricate: it operates as either a helpful or a costly brain state, depending on whether older adults could achieve a certain level of accuracy. In high-accuracy performers, the additional neural effort (S3 recruitment) is successfully translated into behavioral preservation (Compensation). In low-accuracy performers, the same neural strategy is deployed but fails to yield functional benefits, reflecting non-adaptive resource utilization (Inefficiency). Thus, the maintenance of cognitive function in aging depends not merely on recruiting additional neural resources, but on the capacity to effectively utilize high-effort brain state to support behavior. Limitations and Future Directions Our study has four primary limitations. First, the cross-sectional design restricts causal inferences regarding developmental changes; longitudinal designs are required to track intra-individual evolution of brain state dynamics. Second, the reported associations between brain states (e.g., S3 in-degree) and behavior are correlational. Future studies using causal manipulations, such as transcranial magnetic stimulation or neurofeedback, are needed to verify the functional significance of these transitions. Third, generalizability is currently limited to the specific tasks used here; broader validation across diverse cognitive domains is necessary. Finally, the sample size warrants caution; larger cohorts in future research will be essential to ensure statistical robustness. Despite these limitations, this study provides the first HMM-based mapping of dynamic rest-to-task reconfiguration in young and older adults, The conditional efficacy of this strategy underscores the heterogeneity of aging brain dynamics, offering crucial insights into why individual trajectories vary from delayed to accelerated decline ( 3 , 27 , 52 , 75 – 77 ). Future work should test whether specific interventions (e.g., cognitive training, neuromodulation, physical exercise) can shift S3 from a costly to a beneficial state. Determining the boundary conditions of successful reconfiguration will be key to strengthening adaptive dynamics and promoting cognitive resilience in aging. Conclusions This study challenges the perspective of uniform age-related cognitive decline, offering novel evidence via HMM that the aging brain supports goal-directed behavior through a distinct rest-to-task reconfiguration. We demonstrate that while young adults mainly transition to a task-efficient state (S2), older adults sustain an effortful, cortical-dominant state (S3) from rest to task. Crucially, this reconfiguration functions as a conditional compensatory strategy: the recruitment of S3 successfully scaffolds conflict resolution only in older adults who achieve high accuracy, while proving detrimental in low-accuracy performers. These findings suggest that cognitive maintenance in aging depends not on preserving youth-like rest-to-task reconfiguration, but on the capacity to effectively deploy effortful brain state to meet task demands. Materials and Methods Participants A total of 47 healthy young adults (19 females; mean age: 27 ± 2.7 years) and 85 healthy older adults (50 females; mean age: 66 ± 3.5 years) were recruited for the study. All participants completed both resting-state and Stroop task fMRI scans. Inclusion criteria were: ( 1 ) no history of neurological or psychiatric disorders or substance abuse; ( 2 ) self-reported no history of severe head injury; ( 3 ) all were right-handed with normal or corrected-to-normal vision and no color blindness. All participants had at least a middle-school education. Older adults were additionally screened for cognitive health using the Mini-Mental State Examination (MMSE), with a minimum score of 27 ( 78 – 81 ). The MMSE's established efficacy in detecting mild cognitive impairment and dementia made it a suitable screening tool ( 82 ). Following data quality control, 5 young and 18 older participants were excluded for excessive in-scanner head motion (the maximum absolute translation exceeded 3mm or the maximum absolute rotation exceeded 3° in any direction). 1 young and 3 older adults were excluded for failure to meet Stroop performance thresholds (accuracy 1500ms). The final sample for Hidden Markov Model (HMM) analysis comprised 41 young adults (18 females; mean age 25.2 ± 3.6 years) and 63 older adults (45 females; mean age 68.8 ± 4.4 years). Gender and years of education were included as covariates in subsequent statistical analysis. Years of education did not differ significantly between age groups (Welch’s t = − 1.56, p = 0.12, 95% CI [− 0.68, 0.08]). Gender distribution differed across groups (χ²( 1 ) = 12.35, p < 0.001). The data were sourced from a database currently being established to study cognitive control in aging, collected in Chongqing, China. The research was approved by the Institutional Review Board at the Brain Imaging Center of Southwest University. Written informed consents were provided by all participants before their involvement. Experimental Design and Stimuli This experiment employed a 2 (Age Group: young, older) × 4 (Condition: rest—eyes open; rest—eyes closed; task—Stroop control; task—Stroop conflict) mixed design, with Age Group as a between-subjects factor and Condition as a within-subjects factor (Fig. 1 A). The two resting conditions served as dual baselines to dissociate visual input from intrinsic processing( 83 ). Resting state preceded the task and comprised two stages of 7.5 min each. In the eyes-open stage, participants fixated a centrally presented cross; in the eyes-closed stage, they kept their eyes closed while remaining awake and still. These baselines were used to index activity with and without visual input, respectively (Fig. 1 A). The Stroop task was implemented in four blocks, each containing 40 trials (total = 160). Blocks lasted approximately 2 min and were separated by 20s rests, yielding a total task duration of ~ 9 min (Fig. 1 A). Each trial began with a 2000ms fixation, followed by a color–word stimulus displayed for 2500ms and a 500ms inter-stimulus interval (ISI). Stimuli were Chinese color words — “red,” “green,” “yellow,” and “blue” — presented in either control condition (congruent color–word pairings) and conflict condition (incongruent pairings). In the conflict condition, participants were instructed to ignore the word meaning and indicate the font color using a four-button response and required to respond as quickly and as accurately as possible. Before the formal task, participants completed a practice session and were required to reach ≥ 80% accuracy to proceed. Behavioral measures included reaction time (RT, ms) and accuracy (Accuracy, %). Participants with Accuracy 1500ms were excluded from subsequent analyses. Stimulus presentation and response logging were controlled with E-Prime 2.0. Data Acquisition and Preprocessing fMRI data were acquired on a 3.0T Siemens scanner equipped with a standard 12-channel head coil. Functional data were collected using an Echo Planar Imaging (EPI) sequence with the following parameters: repetition time (TR) = 2s; echo time (TE) = 30ms; flip angle = 90°; field of view (FOV) = 224 × 224 mm²; 62 slices; and a voxel size of 2×2×2 mm³. A total of 450 time points were acquired for resting-state scans, and 275 time points for task-state scans. For each participant, anatomical T1-weighted images were obtained (TR = 2.53s; TE = 2.98 ms; flip angle = 7; voxel size = 2×2×2 mm³, 192 slices). Resting-state fMRI data preprocessing was performed using the Gretna toolbox ( 84 ). The images were processed through the following steps: ( 1 ) removal of the first 10 time points; ( 2 ) slice-timing correction; ( 3 ) head motion correction via spatial realignment; ( 4 ) spatial normalization to standard MNI space by registering individual functional images to an Echo Planar Imaging (EPI) template; ( 5 ) spatial smoothing with a 6mm full-width at half-maximum (FWHM) Gaussian kernel; and ( 6 ) regression of nuisance covariates, including signals from the cerebrospinal fluid, white matter, and a 24-parameter motion model ( 85 ). Task-state fMRI data preprocessing was also performed using the Gretna toolbox ( 84 ). The preprocessing pipeline included: removal of the first 10 time points; slice-timing correction; head motion correction; spatial normalization to the standard MNI space by coregistering functional images to each participant's individual structural T1 image, which was then normalized to MNI space; spatial smoothing with a 6mm FWHM Gaussian kernel; and regression of nuisance covariates (cerebrospinal fluid signal, white matter signal, and the 24 head motion parameters). Hidden Markov Model (HMM) Analysis To formally test our hypotheses regarding adaptive brain reconfiguration, we inferred latent brain states by applying a Hidden Markov Model (HMM) using the HMM-MAR toolbox ( 43 , 86 , 87 ) in MATLAB. The HMM decodes discrete, latent brain states and their transitions directly from observed fMRI timeseries. This method aligns with metastability theory, positing rapid transitions between quasi-stable brain states reflecting distinct cognitive processes ( 46 , 47 , 88 , 89 ). fMRI time series were extracted from 268 regions defined by the Shen268 atlas( 90 ). To characterize the spatial organization of HMM states, these nodes were grouped into 10 functionally coherent networks (MF = medial frontal, DMN = default-mode, FP = frontoparietal, VA = visual associative, VI = visual I, VII = visual II, MOT = motor, BG = basal ganglia, LIM = limbic, CB = cerebellum) based on established functional boundaries ( 91 ). The original subcortical-cerebellar network was subdivided into three distinct networks (BG, LIM and CB) to refine neuroanatomical interpretation. A key challenge when comparing groups and conditions is ensuring that the generated brain states are defined consistently. To achieve this, we concatenated the data from all participants (both young and older adults) across all four conditions (rest—eyes open; rest—eyes closed; task—Stroop control; task—Stroop conflict) before fitting the HMM (Fig. 1 B). Specifically, we randomly sampled 128 time points from each resting-state condition (eyes-open, eyes-closed; original 450 time points each) and 128 time points from each Stroop condition (control, conflict), resulting in a concatenated matrix of 268 regions × 53,248 time points across 104 participants (Fig. 1 B). Mapping all data onto a shared state space ensured that the resulting brain states were identically defined across groups and conditions ( 51 ), enabling an unbiased comparison of rest-to-task reconfiguration of brain states. This approach allowed us to examine how external task demands reshape intrinsic state transitions and how aging alters these dynamic reconfigurations. Each HMM-derived state was modeled with a multivariate Gaussian observation model characterized by a mean activation vector (µ) and covariance matrix (Σ), capturing state-specific activation and functional connectivity (Fig. 1 B). We evaluated state numbers K = 1–20; for each K, 20 variational-Bayes optimizations were initialized randomly and the K value with the lowest free energy was retained (SI appendix, Fig. S2). We repeated this full model-selection procedure 10 times, and the final K used for inference was the median of the 20 per-run optimal K values. After estimating the group-level HMM on the concatenated data, we obtained subject-specific parameterizations via dual estimation. Dual estimation refits the previously learned group model to each individual’s time series so that the group-level parameters initialize and constrain the solution while being adapted to the individual, yielding subject-level state time courses and transition matrices suitable for group comparisons ( 43 , 92 ). From each subject-level HMM (separately for each condition), we derived dynamic metrics for statistical analysis. The transition probability (TP) matrix A contains elements \(\:{A}_{ij}=P({S}_{t+1}=j\mid\:{S}_{t}=i)\) , indexing the probability of moving from state i to state j; persistence probability refers to the diagonal entries \(\:{A}_{ii}\) (persistence probabilities), indexing the probability of transitions into the same state at the next time step ( \(\:i=j\) ). Accordingly, the in-degree for state j was defined as the mean of its incoming transition probability and its persistence probability, indexing brain state recruitment—capturing both state transition and persistence—and offers greater sensitivity to individual variability compared to fractional occupancy (FO) or mean dwell time (MDT)( 92 ). Brain state in-degree was computed per participant and per condition and used to quantify how external task demands perturbed intrinsic state dynamics and how aging modulates these rest-to-task reconfiguration patterns (Fig. 1 B). Data Analysis Behavior Analysis We fit linear mixed-effects models (LMMs) separately for reaction time (RT), accuracy (Accuracy), and the Inverse Efficiency Score (IES = RT/Accuracy). Age Group (young, older; between-subjects) and Stroop Condition (control, conflict; within-subjects) were entered as fixed effects with their interaction; Gender and Years of Education were included as covariates (continuous covariates z-standardized before modeling). Accuracy was inverse–square-root transformed for modeling to meet normality and homoscedasticity (figures display raw values). To account for between-subject variability, models included subject-specific random intercepts and by-condition random slopes. Models were estimated with restricted maximum likelihood (REML), and fixed effects were tested using F-tests (two-sided α = 0.05). All analysis were conducted in R environment. RT was calculated based on correct response trials only. Accuracy was defined as the number of correct responses divided by the total number of trials. The Inverse Efficiency Score (IES) was calculated as the mean RT (on correct trials) divided by the corresponding accuracy. Conceptually, IES can be understood as the average time required per correct response: a lower IES indicates higher overall response efficiency. Primary inferences rely on the separate RT and Accuracy models, with IES reported as a complementary efficiency index. To directly test whether the conflict effect (Conflict − Control) differed by age, we computed subject-level RT, accuracy and IES difference scores and compared young with older adults using independent Welch t tests (two-tailed, α = 0.05). HMM-decoded Brain State In-degree Comparison To examine differences in state temporal metrics across age groups (young, older) and conditions (rest: eyes-open; rest: eyes-closed; task: Stroop control; task: Stroop conflict), we employed a non-parametric Aligned Rank Transform (ART) ANOVA. This method addresses potential issues of non-normality in multi-factor mixed designs through rank transformation. Prior to the ART ANOVA, potential influences of gender and education years were evaluated. Given that ART ANOVA does not directly model between-subject covariates, a two-step adjustment was performed for each temporal metric. First, to remove between-subject baseline differences, individual subject-mean scores for each state were regressed onto gender and education years. The effects predicted by these covariates were then subtracted from the original observations, with respect to the overall grand mean. This procedure removes the linear effects of covariates on baseline while preserving task and age group effects, allowing subsequent analyses to focus on the experimental manipulations. Pre-analysis checks confirmed no significant covariate × task interactions across the four states (all p ≥ 0.1). Based on the adjusted data, separate ART models were fitted for each of the four states (S1–S4). The model included age group (between-subjects factor), conditions (within-subjects factor), and their interaction term. Type III sums of squares and sum-to-zero contrasts were used. BH correction was applied to control for false positives across multiple ART ANOVA tests (across 4 states). To address potential misinterpretation of main effects in the presence of significant interactions, post-hoc comparisons focused on simple effects. For each comparison, estimated differences, standard errors, t -statistics, p -values, and 95% confidence intervals were reported in SI Appendix. All ART ANOVA analyses and post-hoc comparisons were performed using the ARTool package in R( 93 ) with a statistical significance threshold set at α = 0.05. Results visualization was performed using the ggplot2 package in R. Brain-Behavior Analysis We tested whether S2 and S3 in-degree during conflict condition predicted the conflict cost (RT Difference) using two separate linear regression models. In each model, predictors were the brain state in-degree (rank-transformed), conflict accuracy (logit-transformed), Age Group (young, older), and their interactions; Gender and Years of Education were included as covariates. All continuous predictors were standardized after transformation. The model form was: "RT Difference"∼"In-degree"×"Accuracy"×"Age Group "+" Gender "+" Years of Education". Coefficient significance was evaluated with two-tailed tests at α = 0.05; 95% bias-corrected and accelerated (BCa) bootstrap confidence intervals were computed for interaction terms to assess robustness with 5000 bootstrap resamples (SI Appendix, Table S8). When probing interactions, simple slopes were estimated within each age group at − 1 SD, mean, and + 1 SD of Accuracy. To further characterize the interaction pattern observed in older adults, we conducted a complementary correlation analysis: older adults were split by the median of conflict Accuracy into Low- vs. High-accuracy sub-groups, and Spearman rank correlations were calculated between S3 in-degree and RT Difference within each subgroup. All analyses were conducted in R. Declarations Author contributions: Zijin Liu: Conceptualization, Methodology, Formal analysis, Investigation, Writing - original draft, Visulization. Haishuo Xia: Conceptualization, Methodology, Investigation, Data acquisition, Writing - review & editing. Antao Chen: Conceptualization, Investigation, Resources, Writing - review & editing, Supervision, Project Administration, Funding Acquisition. Conflict of Interest Disclosures: None. Funding/Support This work was supported by the National Natural Science Foundation of China [grant numbers: 32371105, 32541017]. Acknowledgements: We are grateful to Teng Gesi, Zhang Weikun, Ye Mingzhu, Xiao Yueyang, Peng Panyue, Xu Liang, Meng Zong, and Li Chaohui for their help in recruiting participants and acquiring the data reported in this article. We also thank all participants for their time and commitment to this research. Data, code, and materials availability: Data are not applicable as further research is still in progress. References Kolobaric, A., Andreescu, C., Gerlach, A.R., Jašarević, E., Aizenstein, H., Pascoal, T.A., Ferreira, P.C.L., Bellaver, B., Hong, C.H., Roh, H.W., Cho, Y.H., Hong, S., Nam, Y.J., Park, B., Lee, D.Y., Kim, N., Choi, J.W., Son, S.J., Karim, H.T.: Altered triple network model connectivity is associated with cognitive function and depressive symptoms in older adults. Alzheimer’s Dement. 21 , e14493 (2025) Shen, X., Wang, C., Zhou, X., Zhou, W., Hornburg, D., Wu, S., Snyder, M.P.: Nonlinear dynamics of multi-omics profiles during human aging. Nat. Aging. (2024). 10.1038/s43587-024-00692-2 Stine-Morrow, E.A.L., Manavbasi, I.E.: Beyond Use It or Lose It: The Impact of Engagement on Cognitive Aging. Annu. Rev. Dev. Psychol. 4 , 319–352 (2022) Yang, Z., Wen, J., Erus, G., Govindarajan, S.T., Melhem, R., Mamourian, E., Cui, Y., Srinivasan, D., Abdulkadir, A., Parmpi, P., Wittfeld, K., Grabe, H.J., Bülow, R., Frenzel, S., Tosun, D., Bilgel, M., An, Y., Yi, D., Marcus, D.S., LaMontagne, P., Benzinger, T.L.S., Heckbert, S.R., Austin, T.R., Waldstein, S.R., Evans, M.K., Zonderman, A.B., Launer, L.J., Sotiras, A., Espeland, M.A., Masters, C.L., Maruff, P., Fripp, J., Toga, A.W., O’Bryant, S., Chakravarty, M.M., Villeneuve, S., Johnson, S.C., Morris, J.C., Albert, M.S., Yaffe, K., Völzke, H., Ferrucci, L., Nick Bryan, R., Shinohara, R.T., Fan, Y., Habes, M., Lalousis, P.A., Koutsouleris, N., Wolk, D.A., Resnick, S.M., Shou, H., Nasrallah, I.M., Davatzikos, C.: Brain aging patterns in a large and diverse cohort of 49,482 individuals. Nat. Med. 30 , 3015–3026 (2024) Chan, M.Y., Han, L., Carreno, C.A., Zhang, Z., Rodriguez, R.M., LaRose, M., Hassenstab, J., Wig, G.S.: Long-term prognosis and educational determinants of brain network decline in older adult individuals. Nat. Aging. 1 , 1053–1067 (2021) Deery, H.A., Di Paolo, R., Moran, C., Egan, G.F., Jamadar, S.D.: The older adult brain is less modular, more integrated, and less efficient at rest: A systematic review of large-scale resting‐state functional brain networks in aging. Psychophysiology. 60 , e14159 (2023) Lipsitz, L.A.: Dynamics of Stability: The Physiologic Basis of Functional Health and Frailty. Journals Gerontol. Ser. A: Biol. Sci. Med. Sci. 57 , B115–B125 (2002) Saadon, N., Du, J., Kosakowski, H.L., Angeli, P.A., DiNicola, L.M., Eldaief, M.C., Buckner, R.L.: Within-individual organization of the human cognitive cerebellum: Evidence for closely juxtaposed, functionally specialized regions. S c i e n c e A d v an c e s (2024) L. Sun, T. Zhao, X. Liang, M. Xia, Q. Li, X. Liao, G. Gong, Q. Wang, C. Pang, Q.Yu, Y. Bi, P. Chen, R. Chen, Y. Chen, T. Chen, J. Cheng, Y. Cheng, Z. Cui, Z. Dai,Y. Deng, Y. Ding, Q. Dong, D. Duan, J.-H. Gao, Q. Gong, Y. Han, Z. Han, C.-C. Huang,R. Huang, R. Huo, L. Li, C.-P. Lin, Q. Lin, B. Liu, C. Liu, N. Liu, Y. Liu, Y. Liu,J. Lu, L. Ma, W. Men, S. Qin, J. Qiu, S. Qiu, T. Si, S. Tan, Y. Tang, S. Tao, D. Wang,F. Wang, J. Wang, P. Wang, X. Wang, Y. Wang, D. Wei, Y. Wu, P. Xie, X. Xu, Y. Xu,Z. Xu, L. Yang, H. Yuan, Z. Zeng, H. Zhang, X. Zhang, G. Zhao, Y. Zheng, S. Zhong,Alzheimer’s Disease Neuroimaging Initiative, M. Weiner, P. Aisen, R. Petersen, C.R. Jack, W. Jagust, J. Q. Trojanowki, A. W. Toga, L. Beckett, R. C. Green, A. J. Saykin,J. Morris, L. M. Shaw, E. Liu, T. Montine, R. G. Thomas, M. Donohue, S. Walter, D.Gessert, T. Sather, G. Jiminez, D. Harvey, M. Bernstein, N. Fox, P. Thompson, N. Schuff,C. DeCArli, B. Borowski, J. Gunter, M. Senjem, P. Vemuri, D. Jones, K. Kantarci, C.Ward, R. A. Koeppe, N. Foster, E. M. Reiman, K. Chen, C. Mathis, S. Landau, N. J.Cairns, E. Householder, L. T. Reinwald, V. Lee, M. Korecka, M. Figurski, K. Crawford,S. Neu, T. M. Foroud, S. Potkin, L. Shen, F. Kelley, S. Kim, K. Nho, Z. Kachaturian,R. Frank, P. J. Snyder, S. Molchan, J. Kaye, J. Quinn, B. Lind, R. Carter, S. Dolen,L. S. Schneider, S. Pawluczyk, M. Beccera, L. Teodoro, B. M. Spann, J. Brewer, H.Vanderswag, A. Fleisher, J. L. Heidebrink, J. L. Lord, S. S. Mason, C. S. Albers,D. Knopman, K. Johnson, R. S. Doody, J. V. Meyer, M. Chowdhury, S. Rountree, M. Dang,Y. Stern, L. S. Honig, K. L. Bell, B. Ances, J. C. Morris, M. Carroll, S. Leon, M.A. Mintun, S. Schneider, A. Oliver, R. Griffith, D. Clark, D. Geldmacher, J. Brockington,E. Roberson, H. Grossman, E. Mitsis, L. deToledo-Morrell, R. C. Shah, R. Duara, D.Varon, M. T. Greig, P. Roberts, M. Albert, C. Onyike, D. D’Agostino, S. Kielb, J.E. Galvin, D. M. Pogorelec, B. Cerbone, C. A. Michel, H. Rusinek, M. J. De Leon, L.Glodzik, S. D. Santi, P. M. Doraiswamy, J. R. Petrella, T. Z. Wong, S. E. Arnold,J. H. Karlawish, D. Wolk, C. D. Smith, G. Jicha, P. Hardy, P. Sinha, E. Oates, G.Conrad, O. L. Lopez, M. Oakley, D. M. Simpson, A. P. Porsteinsson, B. S. Goldstein,K. Martin, K. M. Makino, M. S. Ismail, C. Brand, R. A. Mulnard, G. Thai, C. M. Ortiz,K. Womack, D. Mathews, M. Quiceno, R. D. Arrastia, R. King, M. Weiner, K. M. Cook,M. DeVous, A. I. Levey, J. J. Lah, J. S. Cellar, J. M. Burns, H. S. Anderson, R. H.Swerdlow, L. Apostolova, K. Tingus, E. Woo, D. H. S. Silverman, P. H. Lu, G. Bartzokis,N. R. Graff Radford, F. Parfitt, T. Kendall, H. Johnson, M. R. Farlow, A. M. Hake,B. R. Matthews, S. Herring, C. Hunt, C. H. Van Dyck, R. E. Carson, M. G. MacAvoy,H. Chertkow, H. Bergman, C. Hosein, S. Black, B. Stefanovic, C. Caldwell, G. Y. R.Hsiung, H. Feldman, B. Mudge, M. A. Past, A. Kertesz, J. Rogers, D. Trost, C. Bernick,D. Munic, D. Kerwin, M. M. Mesulam, K. Lipowski, C. K. Wu, N. Johnson, C. Sadowsky,W. Martinez, T. Villena, R. S. Turner, K. Johnson, B. Reynolds, R. A. Sperling, K.A. Johnson, G. Marshall, M. Frey, J. Yesavage, J. L. Taylor, B. Lane, A. Rosen, J.Tinklenberg, M. N. Sabbagh, C. M. Belden, S. A. Jacobson, S. A. Sirrel, N. Kowall,R. Killiany, A. E. Budson, A. Norbash, P. L. Johnson, T. O. Obisesan, S. Wolday, J.Allard, A. Lerner, P. Ogrocki, L. Hudson, E. Fletcher, O. Carmichael, J. Olichney,C. DeCarli, S. Kittur, M. Borrie, T. Y. Lee, R. Bartha, S. Johnson, S. Asthana, C.M. Carlsson, S. G. Potkin, A. Preda, D. Nguyen, P. Tariot, S. Reeder, V. Bates, H.Capote, M. Rainka, D. W. Scharre, M. Kataki, A. Adeli, E. A. Zimmerman, D. Celmins,A. D. Brown, G. D. Pearlson, K. Blank, K. Anderson, R. B. Santulli, T. J. Kitzmiller,E. S. Schwartz, K. M. Sink, J. D. Williamson, P. Garg, F. Watkins, B. R. Ott, H. Querfurth,G. Tremont, S. Salloway, P. Malloy, S. Correia, H. J. Rosen, B. L. Miller, J. Mintzer,K. Spicer, D. Bachman, E. Finger, S. Pasternak, I. Rachinsky, D. Drost, N. Pomara,R. Hernando, A. Sarrael, S. K. Schultz, L. L. Boles Ponto, H. Shim, K. E. Smith, N.Relkin, G. Chaing, L. Raudin, A. Smith, K. Fargher, B. A. Raj, DIDA-MDD Working Group,MCADI, Y. Zhou, Y. He, Human lifespan changes in the brain’s functional connectome. Nat Neurosci 28, 891–901 (2025) Bolt, T., Anderson, M.L., Uddin, L.Q.: Beyond the evoked/intrinsic neural process dichotomy. Netw. Neurosci. 2 , 1–22 (2018) Capouskova, K., Kringelbach, M.L., Deco, G.: Modes of cognition: Evidence from metastable brain dynamics. NeuroImage. 260 , 119489 (2022) Honey, C.J., Newman, E.L., Schapiro, A.C.: Switching between internal and external modes: A multiscale learning principle. Netw. Neurosci. 1 , 339–356 (2017) Huang, Z.: Common and unique network basis for externally and internally driven flexibility in cognition: From a developmental perspective. Dev. Cogn. Neurosci. (2025) Weilnhammer, V., Stuke, H., Standvoss, K., Sterzer, P.: Sensory processing in humans and mice fluctuates between external and internal modes. PLoS Biol. 21 , e3002410 (2023) Xiong, B., Chen, C., Tian, Y., Zhang, S., Liu, C., Evans, T.M., Fernández, G., Wu, J., Qin, S.: Brain preparedness: The proactive role of the cortisol awakening response in hippocampal-prefrontal functional interactions. Prog. Neurobiol. 205 , 102127 (2021) Cole, M.W., Bassett, D.S., Power, J.D., Braver, T.S., Petersen, S.E.: Intrinsic and Task-Evoked Network Architectures of the Human Brain. Neuron. 83 , 238–251 (2014) Cole, M.W., Ito, T., Bassett, D.S., Schultz, D.H.: Activity flow over resting-state networks shapes cognitive task activations. Nat. Neurosci. 19 , 1718–1726 (2016) Finc, K., Bonna, K., Lewandowska, M., Wolak, T., Nikadon, J., Dreszer, J., Duch, W., Kühn, S.: Transition of the functional brain network related to increasing cognitive demands. Hum. Brain. Mapp. 38 , 3659–3674 (2017) Hearne, L.J., Cocchi, L., Zalesky, A., Mattingley, J.B.: Reconfiguration of Brain Network Architectures between Resting-State and Complexity-Dependent Cognitive Reasoning. J. Neurosci. 37 , 8399–8411 (2017) Bassett, D.S., Wymbs, N.F., Porter, M.A., Mucha, P.J., Carlson, J.M., Grafton, S.T.: Dynamic reconfiguration of human brain networks during learning. Proc. Natl. Acad. Sci. U.S.A. 108, 7641–7646 (2011) Bassett, D.S., Yang, M., Wymbs, N.F., Grafton, S.T.: Learning-induced autonomy of sensorimotor systems. Nat. Neurosci. 18 , 744–751 (2015) Kucyi, A., Esterman, M., Riley, C.S., Valera, E.M.: Spontaneous default network activity reflects behavioral variability independent of mind-wandering. Proc. Natl. Acad. Sci. U.S.A. 113, 13899–13904 (2016) Spreng, R.N., Stevens, W.D., Chamberlain, J.P., Gilmore, A.W., Schacter, D.L.: Default network activity, coupled with the frontoparietal control network, supports goal-directed cognition. NeuroImage. 53 , 303–317 (2010) Vatansever, D., Menon, D.K., Manktelow, A.E., Sahakian, B.J., Stamatakis, E.A.: Default Mode Dynamics for Global Functional Integration. J. Neurosci. 35 , 15254–15262 (2015) Vatansever, D., Menon, D.K., Stamatakis, E.A.: Default mode contributions to automated information processing. Proc. Natl. Acad. Sci. U.S.A. 114, 12821–12826 (2017) Erb, C.D., Germine, L., Hartshorne, J.K.: Cognitive control across the lifespan: Congruency effects reveal divergent developmental trajectories. J. Exp. Psychol. Gen. 152 , 3285–3291 (2023) LaPlume, A.A., Anderson, N.D., McKetton, L., Levine, B., Troyer, A.K.: When I’m 64: Age-Related Variability in Over 40,000 Online Cognitive Test Takers. Journals Gerontology: Ser. B. 77 , 104–117 (2022) Rey-Mermet, A., Gade, M.: Inhibition in aging: What is preserved? What declines? A meta-analysis. Psychon Bull. Rev. 25 , 1695–1716 (2018) Rey-Mermet, A., Gade, M.: Age-related deficits in the congruency sequence effect are task-specific: An investigation of nine tasks. Psychol. Aging. 35 , 744–764 (2020) Souza, A.S., Frischkorn, G.T., Oberauer, K.: Older yet sharp: No general age-related decline in focusing attention. J. Exp. Psychol. Gen. 153 , 2552–2572 (2024) Veríssimo, J., Verhaeghen, P., Goldman, N., Weinstein, M., Ullman, M.T.: Evidence that ageing yields improvements as well as declines across attention and executive functions. Nat. Hum. Behav. 6 , 97–110 (2021) Barulli, D., Stern, Y.: Efficiency, capacity, compensation, maintenance, plasticity: emerging concepts in cognitive reserve. Trends Cogn. Sci. 17 , 502–509 (2013) Grady, C.: The cognitive neuroscience of ageing. Nat. Rev. Neurosci. 13 , 491–505 (2012) Rieck, J.R., Baracchini, G., Nichol, D., Abdi, H., Grady, C.L.: Reconfiguration and dedifferentiation of functional networks during cognitive control across the adult lifespan. Neurobiol. Aging. 106 , 80–94 (2021) Barulli, D., Stern, Y.: Efficiency, capacity, compensation, maintenance, plasticity: emerging concepts in cognitive reserve. Trends Cogn. Sci. 17 , 502–509 (2013) Cabeza, R., Albert, M., Belleville, S., Craik, F.I.M., Duarte, A., Grady, C.L., Lindenberger, U., Nyberg, L., Park, D.C., Reuter-Lorenz, P.A., Rugg, M.D., Steffener, J.: Rajah, Maintenance, reserve and compensation: the cognitive neuroscience of healthy ageing. Nat. Rev. Neurosci. 19 , 701–710 (2018) Turner, G.R., Spreng, R.N.: Prefrontal Engagement and Reduced Default Network Suppression Co-occur and Are Dynamically Coupled in Older Adults: The Default–Executive Coupling Hypothesis of Aging. J. Cogn. Neurosci. 27 , 2462–2476 (2015) Liu, Z., Xia, H., Chen, A.: Impaired brain ability of older adults to transit and persist to latent states with well-organized structures at wakeful rest. GeroScience. (2024). 10.1007/s11357-024-01366-y van der Meer, J.N., Breakspear, M., Chang, L.J., Sonkusare, S., Cocchi, L.: Movie viewing elicits rich and reliable brain state dynamics. Nat. Commun. 11 , 5004 (2020) Quinn, A.J., Vidaurre, D., Abeysuriya, R., Becker, R., Nobre, A.C.: Woolrich, Task-Evoked Dynamic Network Analysis Through Hidden Markov Modeling. Front. Neurosci. 12 , 603 (2018) Stevner, A.B.A., Vidaurre, D., Cabral, J., Rapuano, K., Nielsen, S.F.V., Tagliazucchi, E., Laufs, H., Vuust, P., Deco, G., Woolrich, M.W., Van Someren, E., Kringelbach, M.L.: Discovery of key whole-brain transitions and dynamics during human wakefulness and non-REM sleep. Nat. Commun. 10 , 1035 (2019) Vidaurre, D.: A new model for simultaneous dimensionality reduction and time-varying functional connectivity estimation. PLoS Comput. Biol. 17 , e1008580 (2021) Vidaurre, D., Smith, S.M., Woolrich, M.W.: Brain network dynamics are hierarchically organized in time. Proc. Natl. Acad. Sci. U.S.A. 114, 12827–12832 (2017) Ito, T., Kulkarni, K.R., Schultz, D.H., Mill, R.D., Chen, R.H., Solomyak, L.I., Cole, M.W.: Cognitive task information is transferred between brain regions via resting-state network topology. Nat. Commun. 8 , 1027 (2017) Luppi, A.I., Mediano, P.A.M., Rosas, F.E., Holland, N., Fryer, T.D., O’Brien, J.T., Rowe, J.B., Menon, D.K., Bor, D., Stamatakis, E.A.: A synergistic core for human brain evolution and cognition. Nat. Neurosci. 25 , 771–782 (2022) Shine, J.M., Bissett, P.G., Bell, P.T., Koyejo, O., Balsters, J.H., Gorgolewski, K.J., Moodie, C.A., Poldrack, R.A.: The Dynamics of Functional Brain Networks: Integrated Network States during Cognitive Task Performance. Neuron. 92 , 544–554 (2016) Baker, A.P., Brookes, M.J., Rezek, I.A., Smith, S.M., Behrens, T., Probert Smith, P.J., Woolrich, M.: Fast transient networks in spontaneous human brain activity. eLife 3, e01867 (2014) Shappell, H., Caffo, B.S., Pekar, J.J., Lindquist, M.A.: Improved state change estimation in dynamic functional connectivity using hidden semi-Markov models. NeuroImage. 191 , 243–257 (2019) Whiteside, D.J., Malpetti, M., Jones, P.S., Ghosh, B.C.P., Coyle-Gilchrist, I., Van Swieten, J.C., Seelaar, H., Jiskoot, L., Borroni, B., Sanchez‐Valle, R., Moreno, F., Laforce, R., Graff, C., Synofzik, M., Galimberti, D., Masellis, M., Tartaglia, M.C., Finger, E., Vandenberghe, R., De Mendonça, A., Tagliavini, F., Butler, C.R., Santana, I., Ber, I.L., Gerhard, A., Ducharme, S., Levin, J., Danek, A., Otto, M., Sorbi, S., Pasquier, F., Bouzigues, A., Russell, L.L., Rohrer, J.D., Rowe, J.B., Rittman, T.: The GENFI consortium, Temporal dynamics predict symptom onset and cognitive decline in familial frontotemporal dementia. Alzheimer’s & Dementia 19, 1947–1962 (2023) Chen, K., Li, C., Sun, W., Tao, Y., Wang, R., Hou, W., Liu, D.-Q.: Hidden Markov Modeling Reveals Prolonged Baseline State and Shortened Antagonistic State across the Adult Lifespan. Cereb. Cortex. 32 , 439–453 (2022) Zeng, Y., Xiong, B., Gao, H., Liu, C., Chen, C., Wu, J., Qin, S.: Cortisol awakening response prompts dynamic reconfiguration of brain networks in emotional and executive functioning. Proc. Natl. Acad. Sci. U.S.A. 121, e2405850121 (2024) Loaiza, V.M.: An overview of the hallmarks of cognitive aging. Curr. Opin. Psychol. (2024) Finkel, D., Reynolds, C., Mcardle, J., Pedersen, N.: Age Changes in Processing Speed as a Leading Indicator of Cognitive Aging. Psychol. Aging. 22 , 558–568 (2007) Kane, M.J., Bleckley, M.K., Conway, A.R.A., Engle, R.W.: A controlled-attention view of working-memory capacity. J. Exp. Psychol. Gen. 130 , 169–183 (2001) Salthouse, T.A.: Aging and measures of processing speed. Biol. Psychol. 54 , 35–54 (2000) Baltes, P.B., Lindenberger, U., Staudinger, U.M.: Life Span Theory in Developmental Psychology in Handbook of Child Psychology , W. Damon, R. M. Lerner, Eds. (Wiley, ed. 1, ; https://onlinelibrary.wiley.com/doi/ (2007). 10.1002/9780470147658.chpsy0111 ) Sala-Llonch, R., Bartrés-Faz, D., Junqué, C.: Reorganization of brain networks in aging: a review of functional connectivity studies. Front. Psychol. 6 (2015) Spreng, R.N., Wojtowicz, M., Grady, C.L.: Reliable differences in brain activity between young and old adults: A quantitative meta-analysis across multiple cognitive domains. Neurosci. Biobehavioral Reviews. 34 , 1178–1194 (2010) Successful, Aging: : Perspectives from the Behavioral Sciences. Cambridge University Press, New York, NY, US, Successful aging: Perspectives from the behavioral sciences. (1990) Baltes, P.B.: Theoretical Propositions of Life-Span Developmental Psychology: On the Dynamics Between Growth and Decline Aron, L., Zullo, J., Yankner, B.A.: The adaptive aging brain. Curr. Opin. Neurobiol. 72 , 91–100 (2022) Park, D.C., Reuter-Lorenz, P.: The Adaptive Brain: Aging and Neurocognitive Scaffolding. Annu. Rev. Psychol. 60 , 173–196 (2009) Ashby, F.G., Turner, B.O., Horvitz, J.C.: Cortical and basal ganglia contributions to habit learning and automaticity. Trends Cogn. Sci. 14 , 208–215 (2010) Graybiel, A.M.: Habits, Rituals, and the Evaluative Brain. Annu. Rev. Neurosci. 31 , 359–387 (2008) Cole, M.W., Reynolds, J.R., Power, J.D., Repovs, G., Anticevic, A., Braver, T.S.: Multi-task connectivity reveals flexible hubs for adaptive task control. Nat. Neurosci. 16 , 1348–1355 (2013) Deco, G., Sanz Perl, Y., Ponce-Alvarez, A., Tagliazucchi, E., Whybrow, P.C., Fuster, J., Kringelbach, M.L.: One ring to rule them all: The unifying role of prefrontal cortex in steering task-related brain dynamics. Prog. Neurobiol. 227 , 102468 (2023) Finc, K., Bonna, K., He, X., Lydon-Staley, D.M., Kühn, S., Duch, W., Bassett, D.S.: Dynamic reconfiguration of functional brain networks during working memory training. Nat. Commun. 11 , 2435 (2020) Wang, X., Krieger-Redwood, K., Lyu, B., Lowndes, R., Wu, G., Souter, N.E., Wang, X., Kong, R., Shafiei, G., Bernhardt, B.C., Cui, Z., Smallwood, J., Du, Y., Jefferies, E.: The Brain’s Topographical Organization Shapes Dynamic Interaction Patterns That Support Flexible Behavior Based on Rules and Long-Term Knowledge. J. Neurosci. 44 , e2223232024 (2024) Braun, U., Schäfer, A., Walter, H., Erk, S., Romanczuk-Seiferth, N., Haddad, L., Schweiger, J.I., Grimm, O., Heinz, A., Tost, H., Meyer-Lindenberg, A., Bassett, D.S.: Dynamic reconfiguration of frontal brain networks during executive cognition in humans. Proc. Natl. Acad. Sci. U.S.A. 112, 11678–11683 (2015) Cole, M.W., Reynolds, J.R., Power, J.D., Repovs, G., Anticevic, A., Braver, T.S.: Multi-task connectivity reveals flexible hubs for adaptive task control. Nat. Neurosci. 16 , 1348–U247 (2013) Kitzbichler, M.G., Henson, R.N.A., Smith, M.L., Nathan, P.J., Bullmore, E.T.: Cognitive Effort Drives Workspace Configuration of Human Brain Functional Networks. J. Neurosci. 31 , 8259–8270 (2011) Cabeza, R., Albert, M., Belleville, S., Craik, F.I.M., Duarte, A., Grady, C.L., Lindenberger, U., Nyberg, L., Park, D.C., Reuter-Lorenz, P.A., Rugg, M.D., Steffener, J.: Rajah, Maintenance, reserve and compensation: the cognitive neuroscience of healthy ageing. Nat. Rev. Neurosci. 19 , 701–710 (2018) Knights, E., Henson, R.N., Morcom, A.M., Mitchell, D.J., Tsvetanov, K.A.: Neural Evidence of Functional Compensation for Fluid Intelligence in Healthy Ageing (preprint, 2024); https://doi.org/10.7554/eLife.93327.1 Vaci, N., Gula, B., Bilalić, M.: Is age really cruel to experts? Compensatory effects of activity. Psychol. Aging. 30 , 740–754 (2015) Yang, Y., Chen, Y., Sang, F., Zhao, S., Wang, J., Li, X., Chen, C., Chen, K., Zhang, Z.: Successful or pathological cognitive aging? Converging into a frontal preservation, temporal impairment (FPTI) hypothesis. Sci. Bull. 67 , 2285–2290 (2022) Zhang, L., Ross, B., Du, Y., Alain, C.: Long-term musical training can protect against age-related upregulation of neural activity in speech-in-noise perception. PLoS Biol. 23 , e3003247 (2025) Hernandez, H., Santamaria-Garcia, H., Moguilner, S., Farina, F.R., Legaz, A., Prado, P., Cuadros, J., Gonzalez, L., Gonzalez-Gomez, R., Migeot, J., Coronel-Oliveros, C., Tagliazucchi, E., Maito, M.A., Godoy, M.E., Cruzat, J., Shaheen, A., Farombi, T., Salazar, D., Da Ros, L.U., Borelli, W.V., Zimmer, E.R., Njamnshi, A.K., Bajpai, S., Dey, A.B., Mostert, C., Merali, Z., Salama, M., Moustafa, S.A., Fittipaldi, S., Altschuler, F., Medel, V., Huepe, D., Yaffe, K., Momoh, C., Eyre, H.A., Swieboda, P., Lawlor, B., Miranda, J.J., Duran-Aniotz, C., Baez, S., Ibanez, A.: The exposome of healthy and accelerated aging across 40 countries. Nat. Med. 31 , 3089–3100 (2025) Anguera, J.A., Reuter-Lorenz, P.A., Willingham, D.T., Seidler, R.D.: Failure to Engage Spatial Working Memory Contributes to Age-related Declines in Visuomotor Learning. J. Cogn. Neurosci. 23 , 11–25 (2011) Fernandez, N.B., Hars, M., Trombetti, A., Vuilleumier, P.: Age-related changes in attention control and their relationship with gait performance in older adults with high risk of falls. NeuroImage. 189 , 551–559 (2019) Nielson, K.A., Langenecker, S.A., Ross, T.J., Garavan, H., Rao, S.M., Stein, E.A.: Comparability of functional MRI response in young and old during inhibition. NeuroReport. 15 , 129–133 (2004) Townsend, J., Adamo, M., Haist, F.: Changing channels: An fMRI study of aging and cross-modal attention shifts. NeuroImage. 31 , 1682–1692 (2006) Tombaugh, T.N., McIntyre, N.J.: The Mini-Mental State Examination: A Comprehensive Review. J. Am. Geriatr. Soc. 40 , 922–935 (1992) Xu, P., Huang, R., Wang, J., Van Dam, N.T., Xie, T., Dong, Z., Chen, C., Gu, R., Zang, Y.-F., He, Y., Fan, J., Luo, Y.: Different topological organization of human brain functional networks with eyes open versus eyes closed. NeuroImage 90, 246–255 (2014) Wang, J., Wang, X., Xia, M., Liao, X., Evans, A., He, Y.: GRETNA: a graph theoretical network analysis toolbox for imaging connectomics. Front. Hum. Neurosci. 9 (2015) Yan, C.-G., Craddock, R.C., Zuo, X.-N., Zang, Y.-F., Milham, M.P.: Standardizing the intrinsic brain: Towards robust measurement of inter-individual variation in 1000 functional connectomes. NeuroImage 80, 246–262 (2013) Vidaurre, D., Quinn, A.J., Baker, A.P., Dupret, D., Tejero-Cantero, A., Woolrich, M.W.: Spectrally resolved fast transient brain states in electrophysiological data. NeuroImage. 126 , 81–95 (2016) Vidaurre, D., Hunt, L.T., Quinn, A.J., Hunt, B.A.E., Brookes, M.J., Nobre, A.C., Woolrich, M.W.: Spontaneous cortical activity transiently organises into frequency specific phase-coupling networks. Nat. Commun. 9 , 2987 (2018) Shine, J.M., Koyejo, O., Poldrack, R.A.: Temporal metastates are associated with differential patterns of time-resolved connectivity, network topology, and attention. Proc. Natl. Acad. Sci. U.S.A. 113, 9888–9891 (2016) Tognoli, E., Kelso, J.A.S.: Metastable Brain Neuron. 81 , 35–48 (2014) Shen, X., Finn, E.S., Scheinost, D., Rosenberg, M.D., Chun, M.M., Papademetris, X., Constable, R.T.: Using connectome-based predictive modeling to predict individual behavior from brain connectivity. Nat. Protoc. 12 , 506–518 (2017) Finn, E.S., Shen, X., Scheinost, D., Rosenberg, M.D., Huang, J., Chun, M.M., Papademetris, X., Constable, R.T.: Functional connectome fingerprinting: identifying individuals using patterns of brain connectivity. Nat. Neurosci. 18 , 1664–1671 (2015) Ahrends, C., Woolrich, M., Vidaurre, D.: Predicting individual traits from models of brain dynamics accurately and reliably using the Fisher kernel. [Preprint] (2024). https://doi.org/10.7554/eLife.95125.1 Wobbrock, J.O., Findlater, L., Gergle, D., Higgins, J.J.: The aligned rank transform for nonparametric factorial analyses using only anova procedures in Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (ACM, Vancouver BC Canada, ; (2011). https://dl.acm.org/doi/10.1145/1978942.1978963 ), pp. 143–146 Additional Declarations There is NO Competing Interest. Supplementary Files SupplementalInformation.docx Supplemental Materials RS441.pdf Reporting Summary Cite Share Download PDF Status: Under Review 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-9298914","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":619062620,"identity":"db0a65ea-02bb-4f6a-a593-e84a2d1ef71e","order_by":0,"name":"Antao Chen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA70lEQVRIie3RIQvCQBTA8TcOZhmsviHoVziLIIp+lY0DLVMEi59Ai2Id6IfQIjYnq9pfMAgDk+HAolh06mxu2gTvX+7C+/EODkCl+uFsAOYDxvfkWDym24D2d8Tgz+kUYo5HKymh2jKzg6MsnQIwMy6H8+I9wW3ALA9Ex5ps5oh2ANbgwLXhOmENCcga4DtTas4hIpxczrTee5EnwS4RWZIbyojU0ggnoT+2oAv3h3FMIQUSxZLHheNRvYhYbxi43rdXwwSSIyck2a06I0+ER6yUc2ZfzHbnBPJ83uNgt983ooufBl5p8uNRlUql+qeuYEVN0vnYxJgAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-9321-681X","institution":"Shanghai Jiao Tong University","correspondingAuthor":true,"prefix":"","firstName":"Antao","middleName":"","lastName":"Chen","suffix":""},{"id":619062621,"identity":"529cb614-cbbb-4e09-8293-fe3cc1cf1b3d","order_by":1,"name":"Zijin Liu","email":"","orcid":"","institution":"Shanghai University of Sport","correspondingAuthor":false,"prefix":"","firstName":"Zijin","middleName":"","lastName":"Liu","suffix":""},{"id":619062622,"identity":"c069ee6e-c2ac-4aef-8f97-96d9d8d74316","order_by":2,"name":"Haishuo Xia","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Haishuo","middleName":"","lastName":"Xia","suffix":""}],"badges":[],"createdAt":"2026-04-02 06:27:13","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9298914/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9298914/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107451313,"identity":"fc0d07d2-bc21-48c8-a61b-9d1bc3c2d2b6","added_by":"auto","created_at":"2026-04-21 15:19:55","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1893469,"visible":true,"origin":"","legend":"\u003cp\u003eDecoding age-related brain reconfiguration via HMM-based dynamic state analysis. (A) Experimental paradigm. A 2 (Age Group: young, older) × 4 (Condition: rest—eyes open; rest—eyes closed; task—Stroop control; task—Stroop conflict) mixed design. Young adults: 25.2 ± 3.6 years; older adults: 68.8 ± 4.4 years. Rest comprised two 7.5-min runs (eyes open with fixation; eyes closed with the instruction “please keep your eyes closed but try not to fall asleep”), providing dual baselines to dissociate visual input from intrinsic processing. During the Stroop task, Chinese color-word characters (“红” red, “绿” green, “黄” yellow, “蓝” blue) were presented in congruent (control) or incongruent (conflict) pairings. Each trial: 2000ms fixation, 2500ms stimulus, 500ms inter-stimulus interval (ISI); responses were made with a four-button box to report font color while ignoring word meaning. The task contained four blocks of 40 trials (160 total), ~2 min per block, with 20-s inter-block rest. A practice session required ≥80% accuracy before entering the scanner.\u003cstrong\u003e \u003c/strong\u003e(B) Dynamic brain state decoding. Preprocessed BOLD time series were extracted from Shen268 atlas (TR = 2 s) and concatenated across the four conditions for each participant. A Hidden Markov Model (HMM) was fit to the ROI time series to infer four latent brain states and their transition dynamics. The model yielded state temporal metrics—state in-degree derived from transition probabilities, and spatial metrics—the state-specific mean activation and covariance-based functional connectivity. These state properties were compared across age groups and conditions to characterize rest-to-task brain reconfiguration.\u003cstrong\u003e \u003c/strong\u003e(C) Behavioral comparison. A comparison of task performance between young and older adults, focusing on accuracy and reaction times during the Stroop task. (D) Brain state reconfiguration. Group comparisons of brain state in-degree to assess the age-related differences from rest to task mode. (E) Brain-behavior associations. Examination of how brain state reconfiguration is related to Stroop performance. Notes: Control = congruent color–word pairings; Conflict = incongruent color–word pairings; ISI = inter-stimulus interval. All on-screen instructions were presented in Chinese.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-9298914/v1/8b0e17a04ce09704b51694ee.png"},{"id":107490365,"identity":"406d9286-cf68-46ce-9dbb-156ffdc41d88","added_by":"auto","created_at":"2026-04-22 02:51:58","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":337045,"visible":true,"origin":"","legend":"\u003cp\u003eStroop task performance by age group and condition. (A) Reaction time (ms): Older adults were slower than young adults in both conditions; conflict increased RT in both groups. Linear mixed-effects models (fixed: Age, Condition; covariates: gender, education) showed main effects of Age and Condition, with no Age × Condition interaction. Welch’s t tests indicated no age group effect on RT Difference. (B) Accuracy (% correct): Conflict reduced accuracy in both groups. Young adults performed more accurate than older in control condition, but no group difference in conflict condition. Accuracy was modelled after an inverse square-root transformation; plotted values are raw. (C) Older adults showed higher (worse) IES across conditions; conflict increased IES in both groups; no significant Age × Condition interaction. Welch’s t tests indicated no age group effect on IES Difference. Bars show group means ± SEM with individual dots; the rightmost column shows boxplots of the conflict effect (Difference = Conflict − Control).\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-9298914/v1/3cb57469b6b576c38a379ab3.png"},{"id":107451317,"identity":"0f279109-80f1-4fc4-9c93-1d4c557cef17","added_by":"auto","created_at":"2026-04-21 15:19:55","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":257930,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification of Rest- and Task-specific Brain States. Boxplots showing the in-degree (quantile) for the four brain states (S1-S4) across different task conditions (Rest: Eyes-Open, Rest: Eyes-Closed, Stroop: Conflict, Stroop: Control). Significant task condition effects were observed for states S1, S2, and S4, indicating different engagement during rest versus task. Thus, S1 and S4 were named as rest-specific states, S2 as task-specific states. S3 showed the lowest in-degree and no significant difference between task conditions, was named as transient state.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-9298914/v1/3794370f19079b79b8b138cb.png"},{"id":107451318,"identity":"9642ac31-aa37-44f5-a3c7-5a84975fe7c3","added_by":"auto","created_at":"2026-04-21 15:19:55","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":320049,"visible":true,"origin":"","legend":"\u003cp\u003eAge × Task interactions in in-degree across four HMM-decoded brain states. Four brain states were identified: S1, a rest-specific state suppressed stronger in young adults during task; S4, a rest-specific state showing stronger suppression in older adults during task; S2, the task-specific state preferentially engaged by young adults; and S3, a state maintained exclusively by older adults from rest to task. Both age groups reconfigured from rest to task (↓S1/S4, ↑S2). Older adults preserved the direction but with reduced S2 recruitment and uniquely maintained S3 from rest to task, while young adults decreased S3 and transitioned more fully into S2.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-9298914/v1/ed14a7e24aa4b6a11ad401ab.png"},{"id":107868277,"identity":"6344441a-df81-4487-94b2-b150693f1717","added_by":"auto","created_at":"2026-04-27 07:09:47","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":949585,"visible":true,"origin":"","legend":"\u003cp\u003eComparisons of Network Spatial Signatures across Four Brain States. (A) Functional connectivity. Rest-specific states (S1, S4; green/blue) show cohesive, predominantly positive correlations and strong intra-network coupling. Task-specific states (S2, S3; red/purple) reconfigure toward increased anti-correlations and reduced intra-network coupling. (B) Spatial activation patterns. S1 exhibits robust activation of DMN and FP; S4 shows greater engagement of MOT and CB. Among task states, S2 is uniquely marked by heightened BG activation, whereas S3 shows reduced BG but increased FP. \u003cstrong\u003e(\u003c/strong\u003eC) Network-specific mean activation (DMN, FP, MOT, BG, CB). Bars summarize the state-wise network profiles, highlighting strong DMN/FP activity in S1, MOT/CB prominence in S4, BG elevation in S2, and BG suppression with FP enhancement in S3. Together, these results demonstrate distinct spatial signatures underlying the age- and task-specific temporal dynamics of each brain state. MF = medial frontal, DMN = default-mode, FP = frontoparietal, VA = visual associative, VI = visual I, VII = visual II, MOT = motor, BG = basal ganglia, LIM = limbic, CB = cerebellum.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-9298914/v1/41b2b1bc7ac78e329ec644a1.png"},{"id":107451320,"identity":"49d2ed56-7420-4cd5-9d9b-bbec4d2eb515","added_by":"auto","created_at":"2026-04-21 15:19:55","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":245604,"visible":true,"origin":"","legend":"\u003cp\u003eS3 in-degree × accuracy interaction by age group (RT Difference). Scatterplots of rank-transformed S3 in-degree vs. RT Difference, with regression lines at low (−1 SD), mean, and high (+1 SD) levels of logit-transformed conflict accuracy. Young adults (n = 41): no in-degree × accuracy interaction (p = 0.42). Older adults (n = 63): significant interaction (p = 0.006). Simple slopes within older adults: at low accuracy, higher S3 in-degree tended to predict larger RT Difference (b = 0.93, SE = 0.54, \u003cem\u003ep\u003c/em\u003e= 0.09); at mean accuracy, no significant correlation presented between S3 in-degree and RT Difference (b = −0.05, SE = 0.38, \u003cem\u003ep\u003c/em\u003e = 0.90); at high accuracy, higher S3 in-degree predicted smaller RT Difference (b = −1.02, SE = 0.50, \u003cem\u003ep\u003c/em\u003e = 0.05). Shaded ribbons denote 95% CIs. RT Difference indexes RT Difference.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-9298914/v1/96ed5b402ab9d1d2bac0443a.png"},{"id":107451321,"identity":"eebdf0f9-4190-4a1b-8058-43383d761134","added_by":"auto","created_at":"2026-04-21 15:19:55","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":270513,"visible":true,"origin":"","legend":"\u003cp\u003eSchematic for Age-related Brain State Reconfiguration from Rest to Task. The top row depicts the experimental context (rest instruction and Stroop stimuli). The panels summarize four HMM-decoded brain states: S1 (rest-specific; stronger task-related suppression in young adults), S4 (rest-specific state; stronger task-related suppression in older adults), S2(task-specific state; increased in both groups but more in young adults during task), and S3 (transient state; maintained from rest to task only in older adults, while young adults decrease). Blue arrows denote young adults, and gray arrows denote older adults; arrow direction indicates change from rest to task (↑ increase, ↓ decrease, → maintain), and relative arrow length/level sketches the magnitude. Asterisks mark interaction significance (*, **, *** for \u003cem\u003ep \u003c/em\u003e\u0026lt; 0.05, 0.01, 0.001). Across groups the direction of reconfiguration is shared (↓S1/S4, ↑S2). However, older adults show attenuated recruitment of S2 and uniquely maintain S3, whereas young adults shift more fully into S2 and down-regulate S3. This pattern indicates preserved transitions out of rest-specific states in aging, but reduced access to the core task state S2 with maintenance of S3.\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-9298914/v1/b373a7689cc74e7a1a05f2c8.png"},{"id":108006207,"identity":"a344ab55-eab6-4a06-a6f4-7554c899cdbc","added_by":"auto","created_at":"2026-04-28 12:54:33","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4707811,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9298914/v1/0104b46d-308f-4da8-948b-e351e3d7d579.pdf"},{"id":107451314,"identity":"746d79c0-ac8f-4b60-952c-394113601434","added_by":"auto","created_at":"2026-04-21 15:19:55","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":336687,"visible":true,"origin":"","legend":"Supplemental Materials","description":"","filename":"SupplementalInformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-9298914/v1/d81b0227a75954c70efc64f3.docx"},{"id":107489000,"identity":"c5a1e89e-f3bb-4640-b470-03eb6783fdd5","added_by":"auto","created_at":"2026-04-22 02:46:25","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":3842143,"visible":true,"origin":"","legend":"Reporting Summary","description":"","filename":"RS441.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9298914/v1/9b05d549bfe1dafbdd7f080b.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Dynamic Rest-to-Task Brain Network Reconfiguration Reveals Divergent Cognitive Changes in Aging","fulltext":[{"header":"Significance statement","content":"\u003cp\u003eCognitive aging is often described as an inevitable decline, yet many older adults maintain high levels of cognitive performance. Using dynamic modeling of fMRI data, we show that aging alters how the brain reconfigures its states from rest to a Stroop task. Young adults preferentially engage a task-efficient brain state, whereas older adults rely on a high-effort, frontal-dominate state. Critically, this older-specific strategy supports conflict resolution only in high-accuracy older adults but is ineffective in low performers. These findings move beyond simple cognitive decline and demonstrate that cognitive preservation in aging depends on the conditional effectiveness of brain network reconfiguration.\u003c/p\u003e"},{"header":"Introduction","content":"\u003cp\u003eAging challenges the brain ability to achieve goal-directed behavior, yet findings on the underlying neural mechanisms remain mixed (\u003cspan additionalcitationids=\"CR2 CR3\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). While prior research has extensively characterized age-related changes in either resting-state or task-evoked networks separately (\u003cspan additionalcitationids=\"CR6 CR7 CR8\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e), this approach neglects a fundamental property of the brain: the ability to reconfigure network organization(\u003cspan additionalcitationids=\"CR11 CR12 CR13 CR14\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e) from a self-organized resting mode to a task-optimized state (\u003cspan additionalcitationids=\"CR17 CR18\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). Such reconfiguration reflects neural resource mobilization to meet cognitive demands, potentially supporting adaptive behavior (\u003cspan additionalcitationids=\"CR21 CR22 CR23 CR24\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e). However, this dynamic reconfiguration remains largely unexplored in current aging literature.\u003c/p\u003e \u003cp\u003eBehavioral findings on cognitive aging are inconsistent, ranging from declined to preserved ability(\u003cspan additionalcitationids=\"CR27 CR28 CR29 CR30\" citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e). This challenges the notion of uniform age-related cognitive decline and necessitates a deeper exploration of its underlying neural substrates. Two theoretical accounts of cognitive aging offer different interpretations for age-related neural changes. The neural inefficiency suggests that excessive recourse recruitment of aging brain yields no functional benefit under high cognitive demand (\u003cspan additionalcitationids=\"CR33\" citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e). In comparison, the functional compensation proposes that older adults strategically engage alternative brain networks to maintain cognitive performance (\u003cspan additionalcitationids=\"CR36\" citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e). It is yet to be clarified whether age-related alterations in rest-to-task reconfiguration represent efficiency failure or a compensatory strategy. For older adults who maintain cognitive performance, do they rely on distinct reconfiguration patterns that diverge from young adults?\u003c/p\u003e \u003cp\u003eQuantifying the dynamic rest-to-task reconfiguration requires moving beyond static analysis. From a dynamical systems perspective, brain activity is modeled as discrete transitions through a state space of recurrent network configurations (\u003cspan additionalcitationids=\"CR39 CR40 CR41 CR42\" citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e). In this view, task onset perturbs the intrinsic resting dynamics toward task-optimized brain state (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan additionalcitationids=\"CR45\" citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e). Hidden Markov Modeling (HMM) operationalizes this framework by decoding latent brain states and their temporal transitions, thereby yielding a dynamic map of the rest-to-task network reconfiguration (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). Unlike sliding-window approaches, HMM decodes brain states with high temporal sensitivity (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan additionalcitationids=\"CR42\" citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan additionalcitationids=\"CR48\" citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e). Crucially, by fitting a shared state space across age groups and conditions, HMM enables direct quantification of how aging alters the reconfiguration of intrinsic dynamics in response to cognitive demand (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan additionalcitationids=\"CR50\" citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWithin this framework, we pursued four objectives to investigate the age-related changes. First, we used the classic color\u0026ndash;word Stroop task as a probe of cognitive demand (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA) and assessed whether behavioral performance in older adults reflects preservation or decline (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC). Second, we computed the state transitions (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB) to test whether young and older adults utilize distinct rest-tot-task reconfiguration pattern (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD). Third, we characterized the spatial signature of each state by combining network activation with functional connectivity. Finally, we examined the functional significance of these dynamics by linking reconfiguration metrics to behavioral outcomes (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE). By shifting focus from static network properties to dynamic reconfiguration, this study aims to clarify whether the age-related changes in rest-to-task reconfiguration pattern represent a uniform decline or a compensatory adaptation essential for sustaining cognition.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStroop Performance: Age-Related Decline and Preservation\u003c/h2\u003e \u003cp\u003eBehavioral performance on the Stroop task was analyzed with linear mixed models (LMMs) for reaction time (RT), accuracy, and the Inverse Efficiency Score (IES; RT/accuracy). Fixed effects were Age Group (younger, older) and Condition (control, conflict); Gender and years of education were included as covariates and showed no significant effects. Accuracy was inverse\u0026ndash;square-root transformed for modeling; figures display raw values.\u003c/p\u003e \u003cp\u003eAcross all three measures, LMMs revealed significant main effects of Age Group and Condition. Older adults had longer RTs and higher IES (lower efficiency) than younger adults in both Control and Conflict conditions (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, C). For accuracy (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB), younger adults outperformed older adults in the Control condition (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02), whereas no age difference was observed in the Conflict condition (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.86). As expected, both groups exhibited a robust conflict effect from Control to Conflict\u0026mdash;RT and IES increased, and accuracy decreased (SI Appendix, Tables S1\u0026ndash;S2).\u003c/p\u003e \u003cp\u003eCritically, no significant Age Group x Condition interaction was observed for RT (\u003cem\u003eF\u003c/em\u003e (1, 102)\u0026thinsp;=\u0026thinsp;1.13, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.29) or IES (\u003cem\u003eF\u003c/em\u003e (1, 102)\u0026thinsp;=\u0026thinsp;0.05, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.83). To directly test age differences in the conflict effect (Conflict\u0026thinsp;\u0026minus;\u0026thinsp;Control), independent Welch\u0026rsquo;s t tests on difference scores indicated no age-group differences for RT Difference (\u003cem\u003et\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.08, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.28) or IES Difference (\u003cem\u003et\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.24, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.81). Thus, despite general age-related slowing and reduced efficiency, the magnitude of the conflict effect did not differ by age. This pattern suggests that older adults may engage distinct neural strategies to support conflict resolution, motivating the subsequent analyses of age-related differences in rest-to-task reconfiguration.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eBrain State Temporal Dynamics: Age-related changes from Rest to Task\u003c/h3\u003e\n\u003cp\u003eBased on the lowest free energy choice, 4 brain states were generated at the group level, which were shared among young and older adults across all conditions. We analyzed brain state in-degree with ART ANOVAs including Age Group (young, older) and Condition (rest: eyes-open, eyes-closed; task: Stroop control, Stroop conflict) while controlling for gender and years of education; all effects reported are covariate-controlled (Gender and education years). All four states exhibited significant main and interaction effects. All effects are covariate-adjusted for gender and years of education (SI Appendix, Table S3, S5).\u003c/p\u003e \u003cp\u003eFor S1, a robust Task effect was observed (\u003cem\u003eF\u003c/em\u003e (3,306)\u0026thinsp;=\u0026thinsp;98.43, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2; = 0.49). The Age effect was also significant (\u003cem\u003eF\u003c/em\u003e (1,102)\u0026thinsp;=\u0026thinsp;10.93, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2; = 0.10) accompanied by a significant Age \u0026times; Task interaction (\u003cem\u003eF\u003c/em\u003e (3,306)\u0026thinsp;=\u0026thinsp;3.59, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.01, ηp\u0026sup2; = 0.03). S2 demonstrated a very large Task modulation (\u003cem\u003eF\u003c/em\u003e (3,306)\u0026thinsp;=\u0026thinsp;151.57, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2; = 0.60), a strong Age effect (\u003cem\u003eF\u003c/em\u003e (1,102)\u0026thinsp;=\u0026thinsp;16.29, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2; = 0.14, and a substantial interaction (\u003cem\u003eF\u003c/em\u003e (3,306)\u0026thinsp;=\u0026thinsp;10.43, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2; = 0.09). Despite having the lowest in-degree, S3 still showed small but significant main effects of Age (\u003cem\u003eF\u003c/em\u003e (1,102)\u0026thinsp;=\u0026thinsp;4.32, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.04, ηp\u0026sup2;=.041), and Task (\u003cem\u003eF\u003c/em\u003e(3,306)\u0026thinsp;=\u0026thinsp;3.98, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2; = 0.04), together with a sizable interaction, \u003cem\u003eF\u003c/em\u003e(3,306)\u0026thinsp;=\u0026thinsp;14.40, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2; = 0.12).S3 showed smaller yet significant main effects for Age (\u003cem\u003eF\u003c/em\u003e(1,102)\u0026thinsp;=\u0026thinsp;4.32, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.04, ηp\u0026sup2; = 0.04, and Task (\u003cem\u003eF\u003c/em\u003e(3,306)\u0026thinsp;=\u0026thinsp;3.98, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.01, ηp\u0026sup2; = 0.04, coupled with a sizeable interaction (\u003cem\u003eF\u003c/em\u003e(3,306)\u0026thinsp;=\u0026thinsp;14.40, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2; = 0.12). Finally, S4 presented a large Task effect (\u003cem\u003eF\u003c/em\u003e(3,306)\u0026thinsp;=\u0026thinsp;134.96, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2; = 0.57), a strong Age effect (\u003cem\u003eF\u003c/em\u003e(1,102)\u0026thinsp;=\u0026thinsp;21.76, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2; = 0.18), and a medium interaction (\u003cem\u003eF\u003c/em\u003e(3,306)\u0026thinsp;=\u0026thinsp;10.71, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001 ηp\u0026sup2; = 0.10). Collectively, Task effects were large for S1, S2, and S4 (ηp\u0026sup2; \u0026asymp; 0.49\u0026ndash;0.60) and smaller for S3. Age effects spanned from small-to-medium (S3) to large (S4), while interactions were presented in all states, averaging medium and approaching medium-to-large in S3.\u003c/p\u003e \u003cp\u003ePost hoc analysis showed that within-rest (eyes-open vs. eyes-closed) and within-task (control vs. conflict) comparisons were not significant (SI Appendix, Table S5). To maximize interpretability around the dominant rest\u0026ndash;task contrast, we collapsed Task Conditions to eyes-open (rest) versus Stroop conflict (task) for confirmatory ART ANOVA analysis (Age Group: young, older; Conditions: eyes-open, Stroop conflict). Results replicated the main pattern (SI Appendix, Table S4, S6): S1 retained very large Task modulation (\u003cem\u003eF\u003c/em\u003e(1,102)\u0026thinsp;=\u0026thinsp;154.89, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2; = 0.603), with Age (\u003cem\u003eF\u003c/em\u003e(1,102)\u0026thinsp;=\u0026thinsp;13.74, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2; = 0.119), and a medium interaction (\u003cem\u003eF\u003c/em\u003e(1,102)\u0026thinsp;=\u0026thinsp;9.87, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.002, ηp\u0026sup2; = 0.09); S2 showed a very large Task effect (\u003cem\u003eF\u003c/em\u003e(1,102)\u0026thinsp;=\u0026thinsp;183.36, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001 ηp\u0026sup2; = 0.64), a medium Age effect (\u003cem\u003eF\u003c/em\u003e(1,102)\u0026thinsp;=\u0026thinsp;12.61, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2; = 0.11), and a medium interaction (\u003cem\u003eF\u003c/em\u003e(1,102)\u0026thinsp;=\u0026thinsp;13.45, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2;=0.117); S3\u0026rsquo;s Task main effect was not significant (\u003cem\u003eF\u003c/em\u003e(1,102)\u0026thinsp;=\u0026thinsp;0.79, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.376, ηp\u0026sup2;= 0.01), but age remained significant (\u003cem\u003eF\u003c/em\u003e(1,102)\u0026thinsp;=\u0026thinsp;5.74, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02, ηp\u0026sup2; = 0.05), and the interaction was large (\u003cem\u003eF\u003c/em\u003e(1,102)\u0026thinsp;=\u0026thinsp;20.17, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2; = 0.17), indicating age-dependent divergence specifically under the rest-versus-conflict contrast; S4 kept a very large Task effect (\u003cem\u003eF\u003c/em\u003e(1,102)\u0026thinsp;=\u0026thinsp;136.12, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2; = 0.57), a large Age effect (\u003cem\u003eF\u003c/em\u003e(1,102)\u0026thinsp;=\u0026thinsp;23.19, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2;=0.19), and a large interaction (\u003cem\u003eF\u003c/em\u003e(1,102)\u0026thinsp;=\u0026thinsp;18.94, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ηp\u0026sup2; = 0.16).\u003c/p\u003e \u003cp\u003eThe convergent evidence supports classifying S1 and S4 as rest-specific states, S2 as a task-specific state, and S3 as a transient state (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) whose dynamics are particularly age-sensitive under conflict demands. As shown by the significant Age \u0026times; Task interactions \u003cb\u003e(\u003c/b\u003eSI Appendix, Table S4), older adults preserved the direction of the transition (\u0026darr;S1/S4, \u0026uarr;S2) but with differed magnitude and uniquely maintaining S3 from rest to task\u0026mdash;consistent with the large S3 interaction despite a nonsignificant S3 Task main effect (F(1,102)\u0026thinsp;=\u0026thinsp;0.79, p\u0026thinsp;=\u0026thinsp;0.38, ηp\u0026sup2; = 0.01). In contrast, young adults shifted more completely into S2 with a reduction of S3 \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). These patterns indicate that, upon task requests, older adults still transition out of rest-specific states but recruit S2 less and retain S3, whereas young adults engage S2 as the main task state.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eBrain State Spatial Signatures: Distinct Network Connectivity and Activation Patterns\u003c/h3\u003e\n\u003cp\u003eBuilding on the age-specific state temporal dynamics identified previously, we next investigated the spatial profiles of all states by examining their network connectivity (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA) and activation patterns (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB, C). This analysis first revealed a functional connectivity reconfiguration between rest and task conditions. The rest-specific states, S1 and S4, were characterized by cohesive, positive correlations among network nodes and robust intra-network coupling. In contrast, the task-specific states, S2 and S3, prompted a shift toward anti-correlated connections and reduced intra-network connectivity, reflecting the competitive interactions required for goal-directed processing.\u003c/p\u003e \u003cp\u003eWhile functional connectivity distinguished rest-specific and task-specific states, network activation patterns revealed age-specific changes. For the rest-specific states, S1 displayed strong activity across DMN, MF and FP. S4, preferred by older adults, showed increased engagement of MOT and CB regions. A similar opposite activation pattern emerged for the task-specific states. For the young-preferred states, S2, the state most engaged by younger adults, was uniquely characterized by heightened BG activation. Conversely, S3, the state maintained by older adults from rest to task, showed reduced BG involvement but increased FP and MOT activation. These results revealed distinct spatial signatures for each brain state characterized by age- and task-specific temporal dynamics.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eBrain-behavior Association: S3 in-degree and conflict cost vary with accuracy in older adults\u003c/h3\u003e\n\u003cdiv class=\"Heading\"\u003eBrain-behavior Association: S3 in-degree and conflict cost vary with accuracy in older adults\u003c/div\u003e \u003cp\u003eTwo linear regression models tested whether S2 and S3 in-degree (rank-transformed), conflict accuracy (logit-transformed), age group and their interactions predicted RT Difference (conflict cost), with accuracy included to address speed\u0026ndash;accuracy trade-off. Gender and years of education were included as covariates and showed no significant effects.\u003c/p\u003e \u003cp\u003eOnly older adults showed an accuracy \u0026times; in-degree interaction in both models. For S2, the interaction term was nominally significant (b\u0026thinsp;=\u0026thinsp;0.91, \u003cem\u003et\u003c/em\u003e(96)\u0026thinsp;=\u0026thinsp;2.28, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.03) but not robust to BCa bootstrapping (95% CI [\u0026minus;\u0026thinsp;0.04, 1.82]). Given the non-robust CI crossing zero (SI Appendix, Table S8). For S3, the interaction was significant and robust (b\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;1.09, \u003cem\u003et\u003c/em\u003e(96)\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;2.77, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.006; BCa 95% CI [(-2.04, -0.23]; ηp\u0026sup2; = 0.08). Simple-slope estimates within older adults (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e) indicated that when accuracy was low (\u0026minus;\u0026thinsp;1 SD), higher S3 in-degree tended to predict larger RT Difference (b\u0026thinsp;=\u0026thinsp;0.93, SE\u0026thinsp;=\u0026thinsp;0.54, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.09); at mean accuracy, the slope was null (b\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.05, SE\u0026thinsp;=\u0026thinsp;0.38, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.90); and when accuracy was high (+\u0026thinsp;1 SD), higher S3 in-degree predicted smaller RT Difference (b\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;1.02, SE\u0026thinsp;=\u0026thinsp;0.50, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.05). No S3 in-degree \u0026times; accuracy interaction was observed in young adults (SI Appendix, Tables S7-S9).\u003c/p\u003e \u003cp\u003eAs an exploratory illustration of this interaction in older adults, participants were divided into low- and high-accuracy groups based on median conflict accuracy. Spearman rank correlations suggested qualitatively opposite associations between S3 in-degree and RT difference across the two subgroups (SI Appendix, Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTogether, these results indicate that in older adults the association between S3 in-degree and conflict cost varies across accuracy levels: higher S3 in-degree is linked to smaller costs when accuracy is higher and to larger costs when accuracy is lower.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Discussions","content":"\u003cp\u003eThe present study investigated whether age-related changes in rest-to-task brain state reconfiguration during the color\u0026ndash;word Stroop task reflect functional decline or strategic compensation. We find that the aging brain supports goal-directed behavior not through a preservation of youth-like dynamics, but via a strategic reconfiguration of brain states. While older adults exhibited generalized behavioral slowing, they successfully maintained conflict resolution comparable to young adults. By leveraging HMM to decode the temporal evolution of brain states, we demonstrate that this behavioral preservation relies on a qualitative shift in reconfiguration: maintaining an effortful, frontal-dominant state (S3) from rest to task. Crucially, we identify the functional significance of this shift as conditional: the recruitment of S3 serves as a compensatory scaffold only for high-performing older adults but reflects neural inefficiency in those with lower accuracy.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eAge-related Slowing and Preservation of Conflict Resolution\u003c/h2\u003e \u003cp\u003eIt is often assumed that cognition generally declines in older age, yet aging shows mixed trajectories in which decline, stability, and even growth co-exist (\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e). A growing body of behavioral work in cognitive control finds little evidence for a specific decline in conflict resolution with age\u0026mdash;across large-sample lifespan analyses and meta-analyses (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e). Notably, research field about attentional selection likewise report preserved focusing efficiency in older adults, arguing against a general decline in cognitive functions(\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e). Our behavioral results align with this trend: despite generalized slowing, older adults showed no reliable age-group difference in the conflict resolution. While our fMRI Stroop used a block design, which can blur trial-wise dynamics (e.g., conflict adaptation), the preserved conflict resolution with generalized slowing suggests that cognitive maintenance among older adults may be not merely a design artifact but may reflect adaptive changes in how neural resources are recruited (\u003cspan additionalcitationids=\"CR54\" citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e). It is intriguing to ask whether older adults employ distinct neural strategies to maintain effective conflict resolution in the context of generalized slowing.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eCommon and Age-specific Brain state reconfigurations from rest to task\u003c/h3\u003e\n\u003cp\u003eBoth age groups successfully reconfigured brain dynamics from rest to task, evidenced by the shared reduction in rest-specific states (S1/S4) and increase in a task-specific state (S2). Despite this common directional reconfiguration (\u0026darr;S1/S4, \u0026uarr;S2), group differences lay in its magnitude (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). Specifically, while both groups showed a decrease in rest-specific states (S1/S4), the observed differential decline was largely attributable to higher rest baselines in older adults. Crucially, the \u003cb\u003etask-induced reduction in state in-degrees\u003c/b\u003e upon task onset was comparable across age groups, suggesting a maintained capacity for disengaging from rest-specific processing in older adults, rather than a complete degradation. By contrast, although S2 increased in both groups, older adults showed a significantly smaller recruitment magnitude. This reduced S2 magnitude suggests declined efficiency in task-state engagement, rather than a complete loss of the reconfiguration pattern itself. These findings supports the view that healthy aging reflects a gradual change in brain functions rather than abrupt degradation, aligning with lifespan models that conceptualize development as a continuous process of gain and loss (\u003cspan additionalcitationids=\"CR57 CR58 CR59\" citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eBeyond these quantitative differences in state reconfiguration, our findings revealed a qualitatively distinct reconfiguration concerning S3 in older adults. While young adults downregulated S3, older adults uniquely maintained this state from rest to task. This implies that S3 is not merely a state to suppress during task but selectively retained in older adults, suggesting that beyond the aging brain may experience a divergent reconfiguration from rest to task. This observation fits with adaptive views of aging\u0026mdash;such as the Scaffolding Theory of Aging and Cognition (STAC)\u0026mdash;which propose that the aging brain recruits different neural resources to support cognitive function (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e, \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eThe Shift from Efficient to Effortful Brain State for Task Execution During Aging\u003c/h3\u003e\n\u003cp\u003eYoung adults preferentially engaged state S2, characterized by prominent BG activation and lower FP involvement. Given the role of the BG in automated processing (\u003cspan additionalcitationids=\"CR63\" citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e), this suggests that young adults rely on a computationally efficient, subcortically-driven mode for task execution. For a relatively simple and well-practiced Stroop task with fMRI block design, this engagement with S2 may represent a highly efficient processing capability, potentially requiring less cognitive effort from young adults. In contrast, older adults recruited S2 to a lesser extent and instead maintained S3, which featured heightened FP and reduced BG involvement. The FP network is central to achieve goal-directed behavior, encompassing cognitive control functions such as working memory, inhibitory control, and cognitive flexibility (\u003cspan additionalcitationids=\"CR66 CR67\" citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e). The heightened FP engagement in S3 may reflect increased cognitive effort or higher cortical arousal for task execution (\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e, \u003cspan additionalcitationids=\"CR70\" citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e). This shift from a BG-dominate state to an FP-dominant state aligns with the Compensatory-Related Utilization of Neural Circuits Hypothesis (CRUNCH), suggesting an increased reliance on additional neural resources, particularly in frontal regions, to compensate for age-related neural inefficiencies and maintain cognitive performance (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e). The contrasting activation of BG and FP networks of S2 and S3, underscore age-related differences in brain network recruitment and cognitive resource allocation during goal-directed processing.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eConditional Functional Significance of The Older-specific State Reconfiguration\u003c/h2\u003e \u003cp\u003eThe age-related decline in S2 and maintenance of S3 from rest to task s admits two possibilities. One hand, it may reflect a diminished capacity to engage automatic processing state, forcing older adults into S3 required of higher cognitive cost or arousal level, reflecting neural inefficiency (\u003cspan additionalcitationids=\"CR33\" citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e). On the other hand, S2 alone proves insufficient for heightened task demands in aging, prompting the active recruitment of a state with greater neural effort to maintain function (\u003cspan additionalcitationids=\"CR73\" citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e), reflecting functional compensation (\u003cspan additionalcitationids=\"CR36\" citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e). The key distinction lies in whether the S3 engagement yield functional benefits for older adults to improve or sustain task performance. Our results suggest that neural inefficiency and functional compensation are not mutually exclusive phenomena but reflect individual differences in the efficacy of neural recruitment. S3's role in older adults is intricate: it operates as either a helpful or a costly brain state, depending on whether older adults could achieve a certain level of accuracy. In high-accuracy performers, the additional neural effort (S3 recruitment) is successfully translated into behavioral preservation (Compensation). In low-accuracy performers, the same neural strategy is deployed but fails to yield functional benefits, reflecting non-adaptive resource utilization (Inefficiency). Thus, the maintenance of cognitive function in aging depends not merely on recruiting additional neural resources, but on the capacity to effectively utilize high-effort brain state to support behavior.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eLimitations and Future Directions\u003c/h2\u003e \u003cp\u003eOur study has four primary limitations. First, the cross-sectional design restricts causal inferences regarding developmental changes; longitudinal designs are required to track intra-individual evolution of brain state dynamics. Second, the reported associations between brain states (e.g., S3 in-degree) and behavior are correlational. Future studies using causal manipulations, such as transcranial magnetic stimulation or neurofeedback, are needed to verify the functional significance of these transitions. Third, generalizability is currently limited to the specific tasks used here; broader validation across diverse cognitive domains is necessary. Finally, the sample size warrants caution; larger cohorts in future research will be essential to ensure statistical robustness.\u003c/p\u003e \u003cp\u003eDespite these limitations, this study provides the first HMM-based mapping of dynamic rest-to-task reconfiguration in young and older adults, The conditional efficacy of this strategy underscores the heterogeneity of aging brain dynamics, offering crucial insights into why individual trajectories vary from delayed to accelerated decline (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e, \u003cspan additionalcitationids=\"CR76\" citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e). Future work should test whether specific interventions (e.g., cognitive training, neuromodulation, physical exercise) can shift S3 from a costly to a beneficial state. Determining the boundary conditions of successful reconfiguration will be key to strengthening adaptive dynamics and promoting cognitive resilience in aging.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusions","content":"\u003cp\u003eThis study challenges the perspective of uniform age-related cognitive decline, offering novel evidence via HMM that the aging brain supports goal-directed behavior through a distinct rest-to-task reconfiguration. We demonstrate that while young adults mainly transition to a task-efficient state (S2), older adults sustain an effortful, cortical-dominant state (S3) from rest to task. Crucially, this reconfiguration functions as a conditional compensatory strategy: the recruitment of S3 successfully scaffolds conflict resolution only in older adults who achieve high accuracy, while proving detrimental in low-accuracy performers. These findings suggest that cognitive maintenance in aging depends not on preserving youth-like rest-to-task reconfiguration, but on the capacity to effectively deploy effortful brain state to meet task demands.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eParticipants\u003c/h2\u003e \u003cp\u003eA total of 47 healthy young adults (19 females; mean age: 27\u0026thinsp;\u0026plusmn;\u0026thinsp;2.7 years) and 85 healthy older adults (50 females; mean age: 66\u0026thinsp;\u0026plusmn;\u0026thinsp;3.5 years) were recruited for the study. All participants completed both resting-state and Stroop task fMRI scans. Inclusion criteria were: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) no history of neurological or psychiatric disorders or substance abuse; (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) self-reported no history of severe head injury; (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) all were right-handed with normal or corrected-to-normal vision and no color blindness. All participants had at least a middle-school education. Older adults were additionally screened for cognitive health using the Mini-Mental State Examination (MMSE), with a minimum score of 27 (\u003cspan additionalcitationids=\"CR79 CR80\" citationid=\"CR78\" class=\"CitationRef\"\u003e78\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e81\u003c/span\u003e). The MMSE's established efficacy in detecting mild cognitive impairment and dementia made it a suitable screening tool (\u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e82\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFollowing data quality control, 5 young and 18 older participants were excluded for excessive in-scanner head motion (the maximum absolute translation exceeded 3mm or the maximum absolute rotation exceeded 3\u0026deg; in any direction). 1 young and 3 older adults were excluded for failure to meet Stroop performance thresholds (accuracy\u0026thinsp;\u0026lt;\u0026thinsp;60% and RT\u0026thinsp;\u0026gt;\u0026thinsp;1500ms). The final sample for Hidden Markov Model (HMM) analysis comprised 41 young adults (18 females; mean age 25.2\u0026thinsp;\u0026plusmn;\u0026thinsp;3.6 years) and 63 older adults (45 females; mean age 68.8\u0026thinsp;\u0026plusmn;\u0026thinsp;4.4 years). Gender and years of education were included as covariates in subsequent statistical analysis. Years of education did not differ significantly between age groups (Welch\u0026rsquo;s \u003cem\u003et\u003c/em\u003e\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;1.56, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.12, 95% CI [\u0026minus;\u0026thinsp;0.68, 0.08]). Gender distribution differed across groups (χ\u0026sup2;(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e)\u0026thinsp;=\u0026thinsp;12.35, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e \u003cp\u003eThe data were sourced from a database currently being established to study cognitive control in aging, collected in Chongqing, China. The research was approved by the Institutional Review Board at the Brain Imaging Center of Southwest University. Written informed consents were provided by all participants before their involvement.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eExperimental Design and Stimuli\u003c/h2\u003e \u003cp\u003eThis experiment employed a 2 (Age Group: young, older) \u0026times; 4 (Condition: rest\u0026mdash;eyes open; rest\u0026mdash;eyes closed; task\u0026mdash;Stroop control; task\u0026mdash;Stroop conflict) mixed design, with Age Group as a between-subjects factor and Condition as a within-subjects factor (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). The two resting conditions served as dual baselines to dissociate visual input from intrinsic processing(\u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e83\u003c/span\u003e). Resting state preceded the task and comprised two stages of 7.5 min each. In the eyes-open stage, participants fixated a centrally presented cross; in the eyes-closed stage, they kept their eyes closed while remaining awake and still. These baselines were used to index activity with and without visual input, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA).\u003c/p\u003e \u003cp\u003eThe Stroop task was implemented in four blocks, each containing 40 trials (total\u0026thinsp;=\u0026thinsp;160). Blocks lasted approximately 2 min and were separated by 20s rests, yielding a total task duration of ~\u0026thinsp;9 min (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). Each trial began with a 2000ms fixation, followed by a color\u0026ndash;word stimulus displayed for 2500ms and a 500ms inter-stimulus interval (ISI). Stimuli were Chinese color words \u0026mdash; \u0026ldquo;red,\u0026rdquo; \u0026ldquo;green,\u0026rdquo; \u0026ldquo;yellow,\u0026rdquo; and \u0026ldquo;blue\u0026rdquo; \u0026mdash; presented in either control condition (congruent color\u0026ndash;word pairings) and conflict condition (incongruent pairings). In the conflict condition, participants were instructed to ignore the word meaning and indicate the font color using a four-button response and required to respond as quickly and as accurately as possible.\u003c/p\u003e \u003cp\u003e Before the formal task, participants completed a practice session and were required to reach\u0026thinsp;\u0026ge;\u0026thinsp;80% accuracy to proceed. Behavioral measures included reaction time (RT, ms) and accuracy (Accuracy, %). Participants with Accuracy\u0026thinsp;\u0026lt;\u0026thinsp;60% and RT\u0026thinsp;\u0026gt;\u0026thinsp;1500ms were excluded from subsequent analyses. Stimulus presentation and response logging were controlled with E-Prime 2.0.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eData Acquisition and Preprocessing\u003c/h2\u003e \u003cp\u003efMRI data were acquired on a 3.0T Siemens scanner equipped with a standard 12-channel head coil. Functional data were collected using an Echo Planar Imaging (EPI) sequence with the following parameters: repetition time (TR) = 2s; echo time (TE) = 30ms; flip angle\u0026thinsp;=\u0026thinsp;90\u0026deg;; field of view (FOV)\u0026thinsp;=\u0026thinsp;224 \u0026times; 224 mm\u0026sup2;; 62 slices; and a voxel size of 2\u0026times;2\u0026times;2 mm\u0026sup3;. A total of 450 time points were acquired for resting-state scans, and 275 time points for task-state scans. For each participant, anatomical T1-weighted images were obtained (TR\u0026thinsp;=\u0026thinsp;2.53s; TE\u0026thinsp;=\u0026thinsp;2.98 ms; flip angle\u0026thinsp;=\u0026thinsp;7; voxel size\u0026thinsp;=\u0026thinsp;2\u0026times;2\u0026times;2 mm\u0026sup3;, 192 slices).\u003c/p\u003e \u003cp\u003eResting-state fMRI data preprocessing was performed using the Gretna toolbox (\u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e84\u003c/span\u003e). The images were processed through the following steps: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) removal of the first 10 time points; (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) slice-timing correction; (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) head motion correction via spatial realignment; (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e) spatial normalization to standard MNI space by registering individual functional images to an Echo Planar Imaging (EPI) template; (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e) spatial smoothing with a 6mm full-width at half-maximum (FWHM) Gaussian kernel; and (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e) regression of nuisance covariates, including signals from the cerebrospinal fluid, white matter, and a 24-parameter motion model (\u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e85\u003c/span\u003e). Task-state fMRI data preprocessing was also performed using the Gretna toolbox (\u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e84\u003c/span\u003e). The preprocessing pipeline included: removal of the first 10 time points; slice-timing correction; head motion correction; spatial normalization to the standard MNI space by coregistering functional images to each participant's individual structural T1 image, which was then normalized to MNI space; spatial smoothing with a 6mm FWHM Gaussian kernel; and regression of nuisance covariates (cerebrospinal fluid signal, white matter signal, and the 24 head motion parameters).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eHidden Markov Model (HMM) Analysis\u003c/h2\u003e \u003cp\u003eTo formally test our hypotheses regarding adaptive brain reconfiguration, we inferred latent brain states by applying a Hidden Markov Model (HMM) using the HMM-MAR toolbox (\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e86\u003c/span\u003e, \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e87\u003c/span\u003e) in MATLAB. The HMM decodes discrete, latent brain states and their transitions directly from observed fMRI timeseries. This method aligns with metastability theory, positing rapid transitions between quasi-stable brain states reflecting distinct cognitive processes (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e, \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e88\u003c/span\u003e, \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e89\u003c/span\u003e). fMRI time series were extracted from 268 regions defined by the Shen268 atlas(\u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e90\u003c/span\u003e). To characterize the spatial organization of HMM states, these nodes were grouped into 10 functionally coherent networks (MF\u0026thinsp;=\u0026thinsp;medial frontal, DMN\u0026thinsp;=\u0026thinsp;default-mode, FP\u0026thinsp;=\u0026thinsp;frontoparietal, VA\u0026thinsp;=\u0026thinsp;visual associative, VI\u0026thinsp;=\u0026thinsp;visual I, VII\u0026thinsp;=\u0026thinsp;visual II, MOT\u0026thinsp;=\u0026thinsp;motor, BG\u0026thinsp;=\u0026thinsp;basal ganglia, LIM\u0026thinsp;=\u0026thinsp;limbic, CB\u0026thinsp;=\u0026thinsp;cerebellum) based on established functional boundaries (\u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e91\u003c/span\u003e). The original subcortical-cerebellar network was subdivided into three distinct networks (BG, LIM and CB) to refine neuroanatomical interpretation.\u003c/p\u003e \u003cp\u003eA key challenge when comparing groups and conditions is ensuring that the generated brain states are defined consistently. To achieve this, we concatenated the data from all participants (both young and older adults) across all four conditions (rest\u0026mdash;eyes open; rest\u0026mdash;eyes closed; task\u0026mdash;Stroop control; task\u0026mdash;Stroop conflict) before fitting the HMM (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). Specifically, we randomly sampled 128 time points from each resting-state condition (eyes-open, eyes-closed; original 450 time points each) and 128 time points from each Stroop condition (control, conflict), resulting in a concatenated matrix of 268 regions \u0026times; 53,248 time points across 104 participants (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). Mapping all data onto a shared state space ensured that the resulting brain states were identically defined across groups and conditions (\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e), enabling an unbiased comparison of rest-to-task reconfiguration of brain states. This approach allowed us to examine how external task demands reshape intrinsic state transitions and how aging alters these dynamic reconfigurations.\u003c/p\u003e \u003cp\u003eEach HMM-derived state was modeled with a multivariate Gaussian observation model characterized by a mean activation vector (\u0026micro;) and covariance matrix (Σ), capturing state-specific activation and functional connectivity (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). We evaluated state numbers K\u0026thinsp;=\u0026thinsp;1\u0026ndash;20; for each K, 20 variational-Bayes optimizations were initialized randomly and the K value with the lowest free energy was retained (SI appendix, Fig. S2). We repeated this full model-selection procedure 10 times, and the final K used for inference was the median of the 20 per-run optimal K values. After estimating the group-level HMM on the concatenated data, we obtained subject-specific parameterizations via dual estimation. Dual estimation refits the previously learned group model to each individual\u0026rsquo;s time series so that the group-level parameters initialize and constrain the solution while being adapted to the individual, yielding subject-level state time courses and transition matrices suitable for group comparisons (\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e92\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFrom each subject-level HMM (separately for each condition), we derived dynamic metrics for statistical analysis. The transition probability (TP) matrix A contains elements \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{A}_{ij}=P({S}_{t+1}=j\\mid\\:{S}_{t}=i)\\)\u003c/span\u003e\u003c/span\u003e, indexing the probability of moving from state i to state j; persistence probability refers to the diagonal entries \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{A}_{ii}\\)\u003c/span\u003e\u003c/span\u003e(persistence probabilities), indexing the probability of transitions into the same state at the next time step (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:i=j\\)\u003c/span\u003e\u003c/span\u003e). Accordingly, the in-degree for state j was defined as the mean of its incoming transition probability and its persistence probability, indexing brain state recruitment\u0026mdash;capturing both state transition and persistence\u0026mdash;and offers greater sensitivity to individual variability compared to fractional occupancy (FO) or mean dwell time (MDT)(\u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e92\u003c/span\u003e). Brain state in-degree was computed per participant and per condition and used to quantify how external task demands perturbed intrinsic state dynamics and how aging modulates these rest-to-task reconfiguration patterns (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eData Analysis\u003c/h2\u003e \u003cdiv id=\"Sec20\" class=\"Section3\"\u003e \u003ch2\u003eBehavior Analysis\u003c/h2\u003e \u003cp\u003eWe fit linear mixed-effects models (LMMs) separately for reaction time (RT), accuracy (Accuracy), and the Inverse Efficiency Score (IES\u0026thinsp;=\u0026thinsp;RT/Accuracy). Age Group (young, older; between-subjects) and Stroop Condition (control, conflict; within-subjects) were entered as fixed effects with their interaction; Gender and Years of Education were included as covariates (continuous covariates z-standardized before modeling). Accuracy was inverse\u0026ndash;square-root transformed for modeling to meet normality and homoscedasticity (figures display raw values). To account for between-subject variability, models included subject-specific random intercepts and by-condition random slopes. Models were estimated with restricted maximum likelihood (REML), and fixed effects were tested using F-tests (two-sided α\u0026thinsp;=\u0026thinsp;0.05). All analysis were conducted in R environment.\u003c/p\u003e \u003cp\u003eRT was calculated based on correct response trials only. Accuracy was defined as the number of correct responses divided by the total number of trials. The Inverse Efficiency Score (IES) was calculated as the mean RT (on correct trials) divided by the corresponding accuracy. Conceptually, IES can be understood as the average time required per correct response: a lower IES indicates higher overall response efficiency. Primary inferences rely on the separate RT and Accuracy models, with IES reported as a complementary efficiency index. To directly test whether the conflict effect (Conflict\u0026thinsp;\u0026minus;\u0026thinsp;Control) differed by age, we computed subject-level RT, accuracy and IES difference scores and compared young with older adults using independent Welch t tests (two-tailed, α\u0026thinsp;=\u0026thinsp;0.05).\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eHMM-decoded Brain State In-degree Comparison\u003c/h2\u003e \u003cp\u003eTo examine differences in state temporal metrics across age groups (young, older) and conditions (rest: eyes-open; rest: eyes-closed; task: Stroop control; task: Stroop conflict), we employed a non-parametric Aligned Rank Transform (ART) ANOVA. This method addresses potential issues of non-normality in multi-factor mixed designs through rank transformation.\u003c/p\u003e \u003cp\u003ePrior to the ART ANOVA, potential influences of gender and education years were evaluated. Given that ART ANOVA does not directly model between-subject covariates, a two-step adjustment was performed for each temporal metric. First, to remove between-subject baseline differences, individual subject-mean scores for each state were regressed onto gender and education years. The effects predicted by these covariates were then subtracted from the original observations, with respect to the overall grand mean. This procedure removes the linear effects of covariates on baseline while preserving task and age group effects, allowing subsequent analyses to focus on the experimental manipulations. Pre-analysis checks confirmed no significant covariate \u0026times; task interactions across the four states (all \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026ge;\u0026thinsp;0.1).\u003c/p\u003e \u003cp\u003eBased on the adjusted data, separate ART models were fitted for each of the four states (S1\u0026ndash;S4). The model included age group (between-subjects factor), conditions (within-subjects factor), and their interaction term. Type III sums of squares and sum-to-zero contrasts were used. BH correction was applied to control for false positives across multiple ART ANOVA tests (across 4 states). To address potential misinterpretation of main effects in the presence of significant interactions, post-hoc comparisons focused on simple effects. For each comparison, estimated differences, standard errors, \u003cem\u003et\u003c/em\u003e-statistics, \u003cem\u003ep\u003c/em\u003e-values, and 95% confidence intervals were reported in SI Appendix.\u003c/p\u003e \u003cp\u003eAll ART ANOVA analyses and post-hoc comparisons were performed using the ARTool package in R(\u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e93\u003c/span\u003e) with a statistical significance threshold set at \u003cem\u003eα\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.05. Results visualization was performed using the ggplot2 package in R.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eBrain-Behavior Analysis\u003c/h2\u003e \u003cp\u003eWe tested whether S2 and S3 in-degree during conflict condition predicted the conflict cost (RT Difference) using two separate linear regression models. In each model, predictors were the brain state in-degree (rank-transformed), conflict accuracy (logit-transformed), Age Group (young, older), and their interactions; Gender and Years of Education were included as covariates. All continuous predictors were standardized after transformation. The model form was:\u003c/p\u003e\n\u003cp\u003e\"RT Difference\"∼\"In-degree\"×\"Accuracy\"×\"Age Group \"+\" Gender \"+\" Years of Education\".\u003c/p\u003e\n\u003cp\u003eCoefficient significance was evaluated with two-tailed tests at α\u0026thinsp;=\u0026thinsp;0.05; 95% bias-corrected and accelerated (BCa) bootstrap confidence intervals were computed for interaction terms to assess robustness with 5000 bootstrap resamples (SI Appendix, Table S8). When probing interactions, simple slopes were estimated within each age group at \u0026minus;\u0026thinsp;1 SD, mean, and +\u0026thinsp;1 SD of Accuracy. To further characterize the interaction pattern observed in older adults, we conducted a complementary correlation analysis: older adults were split by the median of conflict Accuracy into Low- vs. High-accuracy sub-groups, and Spearman rank correlations were calculated between S3 in-degree and RT Difference within each subgroup. All analyses were conducted in R.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor contributions:\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eZijin Liu: Conceptualization, Methodology, Formal analysis, Investigation, Writing - original draft, Visulization. Haishuo Xia: Conceptualization, Methodology, Investigation, Data acquisition, Writing - review \u0026amp; editing. Antao Chen: Conceptualization, Investigation, Resources, Writing - review \u0026amp; editing, Supervision, Project Administration, Funding Acquisition.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of Interest Disclosures:\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNone.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding/Support\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the National Natural Science Foundation of China [grant numbers: 32371105, 32541017].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe are grateful to Teng Gesi, Zhang Weikun, Ye Mingzhu, Xiao Yueyang, Peng Panyue, Xu Liang, Meng Zong, and Li Chaohui for their help in recruiting participants and acquiring the data reported in this article. We also thank all participants for their time and commitment to this research.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData, code, and materials availability:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData are not applicable as further research is still in progress.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eKolobaric, A., Andreescu, C., Gerlach, A.R., Jašarević, E., Aizenstein, H., Pascoal, T.A., Ferreira, P.C.L., Bellaver, B., Hong, C.H., Roh, H.W., Cho, Y.H., Hong, S., Nam, Y.J., Park, B., Lee, D.Y., Kim, N., Choi, J.W., Son, S.J., Karim, H.T.: Altered triple network model connectivity is associated with cognitive function and depressive symptoms in older adults. Alzheimer\u0026rsquo;s Dement. \u003cb\u003e21\u003c/b\u003e, e14493 (2025)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShen, X., Wang, C., Zhou, X., Zhou, W., Hornburg, D., Wu, S., Snyder, M.P.: Nonlinear dynamics of multi-omics profiles during human aging. Nat. Aging. (2024). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s43587-024-00692-2\u003c/span\u003e\u003cspan address=\"10.1038/s43587-024-00692-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStine-Morrow, E.A.L., Manavbasi, I.E.: Beyond Use It or Lose It: The Impact of Engagement on Cognitive Aging. Annu. Rev. Dev. Psychol. \u003cb\u003e4\u003c/b\u003e, 319\u0026ndash;352 (2022)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang, Z., Wen, J., Erus, G., Govindarajan, S.T., Melhem, R., Mamourian, E., Cui, Y., Srinivasan, D., Abdulkadir, A., Parmpi, P., Wittfeld, K., Grabe, H.J., B\u0026uuml;low, R., Frenzel, S., Tosun, D., Bilgel, M., An, Y., Yi, D., Marcus, D.S., LaMontagne, P., Benzinger, T.L.S., Heckbert, S.R., Austin, T.R., Waldstein, S.R., Evans, M.K., Zonderman, A.B., Launer, L.J., Sotiras, A., Espeland, M.A., Masters, C.L., Maruff, P., Fripp, J., Toga, A.W., O\u0026rsquo;Bryant, S., Chakravarty, M.M., Villeneuve, S., Johnson, S.C., Morris, J.C., Albert, M.S., Yaffe, K., V\u0026ouml;lzke, H., Ferrucci, L., Nick Bryan, R., Shinohara, R.T., Fan, Y., Habes, M., Lalousis, P.A., Koutsouleris, N., Wolk, D.A., Resnick, S.M., Shou, H., Nasrallah, I.M., Davatzikos, C.: Brain aging patterns in a large and diverse cohort of 49,482 individuals. Nat. Med. \u003cb\u003e30\u003c/b\u003e, 3015\u0026ndash;3026 (2024)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChan, M.Y., Han, L., Carreno, C.A., Zhang, Z., Rodriguez, R.M., LaRose, M., Hassenstab, J., Wig, G.S.: Long-term prognosis and educational determinants of brain network decline in older adult individuals. Nat. Aging. \u003cb\u003e1\u003c/b\u003e, 1053\u0026ndash;1067 (2021)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeery, H.A., Di Paolo, R., Moran, C., Egan, G.F., Jamadar, S.D.: The older adult brain is less modular, more integrated, and less efficient at rest: A systematic review of large-scale resting‐state functional brain networks in aging. Psychophysiology. \u003cb\u003e60\u003c/b\u003e, e14159 (2023)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLipsitz, L.A.: Dynamics of Stability: The Physiologic Basis of Functional Health and Frailty. Journals Gerontol. Ser. A: Biol. Sci. Med. Sci. \u003cb\u003e57\u003c/b\u003e, B115\u0026ndash;B125 (2002)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaadon, N., Du, J., Kosakowski, H.L., Angeli, P.A., DiNicola, L.M., Eldaief, M.C., Buckner, R.L.: Within-individual organization of the human cognitive cerebellum: Evidence for closely juxtaposed, functionally specialized regions. \u003cem\u003eS c i e n c e A d v an c e s\u003c/em\u003e (2024)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eL. Sun, T. Zhao, X. Liang, M. Xia, Q. Li, X. Liao, G. Gong, Q. Wang, C. Pang, Q.Yu, Y. Bi, P. Chen, R. Chen, Y. Chen, T. Chen, J. Cheng, Y. Cheng, Z. Cui, Z. Dai,Y. Deng, Y. Ding, Q. Dong, D. Duan, J.-H. Gao, Q. Gong, Y. Han, Z. Han, C.-C. Huang,R. Huang, R. Huo, L. Li, C.-P. Lin, Q. Lin, B. Liu, C. Liu, N. Liu, Y. Liu, Y. Liu,J. Lu, L. Ma, W. Men, S. Qin, J. Qiu, S. Qiu, T. Si, S. Tan, Y. Tang, S. Tao, D. Wang,F. Wang, J. Wang, P. Wang, X. Wang, Y. Wang, D. Wei, Y. Wu, P. Xie, X. Xu, Y. Xu,Z. Xu, L. Yang, H. Yuan, Z. Zeng, H. Zhang, X. Zhang, G. Zhao, Y. Zheng, S. Zhong,Alzheimer\u0026rsquo;s Disease Neuroimaging Initiative, M. Weiner, P. Aisen, R. Petersen, C.R. Jack, W. Jagust, J. Q. Trojanowki, A. W. Toga, L. Beckett, R. C. Green, A. J. Saykin,J. Morris, L. M. Shaw, E. Liu, T. Montine, R. G. Thomas, M. Donohue, S. Walter, D.Gessert, T. Sather, G. Jiminez, D. Harvey, M. Bernstein, N. Fox, P. Thompson, N. Schuff,C. DeCArli, B. Borowski, J. Gunter, M. Senjem, P. Vemuri, D. Jones, K. Kantarci, C.Ward, R. A. Koeppe, N. Foster, E. M. Reiman, K. Chen, C. Mathis, S. Landau, N. J.Cairns, E. Householder, L. T. Reinwald, V. Lee, M. Korecka, M. Figurski, K. Crawford,S. Neu, T. M. Foroud, S. Potkin, L. Shen, F. Kelley, S. Kim, K. Nho, Z. Kachaturian,R. Frank, P. J. Snyder, S. Molchan, J. Kaye, J. Quinn, B. Lind, R. Carter, S. Dolen,L. S. Schneider, S. Pawluczyk, M. Beccera, L. Teodoro, B. M. Spann, J. Brewer, H.Vanderswag, A. Fleisher, J. L. Heidebrink, J. L. Lord, S. S. Mason, C. S. Albers,D. Knopman, K. Johnson, R. S. Doody, J. V. Meyer, M. Chowdhury, S. Rountree, M. Dang,Y. Stern, L. S. Honig, K. L. Bell, B. Ances, J. C. Morris, M. Carroll, S. Leon, M.A. Mintun, S. Schneider, A. Oliver, R. Griffith, D. Clark, D. Geldmacher, J. Brockington,E. Roberson, H. Grossman, E. Mitsis, L. deToledo-Morrell, R. C. Shah, R. Duara, D.Varon, M. T. Greig, P. Roberts, M. Albert, C. Onyike, D. D\u0026rsquo;Agostino, S. Kielb, J.E. Galvin, D. M. Pogorelec, B. Cerbone, C. A. Michel, H. Rusinek, M. J. De Leon, L.Glodzik, S. D. Santi, P. M. Doraiswamy, J. R. Petrella, T. Z. Wong, S. E. Arnold,J. H. Karlawish, D. Wolk, C. D. Smith, G. Jicha, P. Hardy, P. Sinha, E. Oates, G.Conrad, O. L. Lopez, M. Oakley, D. M. Simpson, A. P. Porsteinsson, B. S. Goldstein,K. Martin, K. M. Makino, M. S. Ismail, C. Brand, R. A. Mulnard, G. Thai, C. M. Ortiz,K. Womack, D. Mathews, M. Quiceno, R. D. Arrastia, R. King, M. Weiner, K. M. Cook,M. DeVous, A. I. Levey, J. J. Lah, J. S. Cellar, J. M. Burns, H. S. Anderson, R. H.Swerdlow, L. Apostolova, K. Tingus, E. Woo, D. H. S. Silverman, P. H. Lu, G. Bartzokis,N. R. Graff Radford, F. Parfitt, T. Kendall, H. Johnson, M. R. Farlow, A. M. Hake,B. R. Matthews, S. Herring, C. Hunt, C. H. Van Dyck, R. E. Carson, M. G. MacAvoy,H. Chertkow, H. Bergman, C. Hosein, S. Black, B. Stefanovic, C. Caldwell, G. Y. R.Hsiung, H. Feldman, B. Mudge, M. A. Past, A. Kertesz, J. Rogers, D. Trost, C. Bernick,D. Munic, D. Kerwin, M. M. Mesulam, K. Lipowski, C. K. Wu, N. Johnson, C. Sadowsky,W. Martinez, T. Villena, R. S. Turner, K. Johnson, B. Reynolds, R. A. Sperling, K.A. Johnson, G. Marshall, M. Frey, J. Yesavage, J. L. Taylor, B. Lane, A. Rosen, J.Tinklenberg, M. N. Sabbagh, C. M. Belden, S. A. Jacobson, S. A. Sirrel, N. Kowall,R. Killiany, A. E. Budson, A. Norbash, P. L. Johnson, T. O. Obisesan, S. Wolday, J.Allard, A. Lerner, P. Ogrocki, L. Hudson, E. Fletcher, O. Carmichael, J. Olichney,C. DeCarli, S. Kittur, M. Borrie, T. Y. Lee, R. Bartha, S. Johnson, S. Asthana, C.M. Carlsson, S. G. Potkin, A. Preda, D. Nguyen, P. Tariot, S. Reeder, V. Bates, H.Capote, M. Rainka, D. W. Scharre, M. Kataki, A. Adeli, E. A. Zimmerman, D. Celmins,A. D. Brown, G. D. Pearlson, K. Blank, K. Anderson, R. B. Santulli, T. J. Kitzmiller,E. S. Schwartz, K. M. Sink, J. D. Williamson, P. Garg, F. Watkins, B. R. Ott, H. Querfurth,G. Tremont, S. Salloway, P. Malloy, S. Correia, H. J. Rosen, B. L. Miller, J. Mintzer,K. Spicer, D. Bachman, E. Finger, S. Pasternak, I. Rachinsky, D. Drost, N. Pomara,R. Hernando, A. Sarrael, S. K. Schultz, L. L. Boles Ponto, H. Shim, K. E. Smith, N.Relkin, G. Chaing, L. Raudin, A. Smith, K. Fargher, B. A. Raj, DIDA-MDD Working Group,MCADI, Y. Zhou, Y. He, Human lifespan changes in the brain\u0026rsquo;s functional connectome.\u003cem\u003eNat Neurosci\u003c/em\u003e 28, 891\u0026ndash;901 (2025)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBolt, T., Anderson, M.L., Uddin, L.Q.: Beyond the evoked/intrinsic neural process dichotomy. Netw. Neurosci. \u003cb\u003e2\u003c/b\u003e, 1\u0026ndash;22 (2018)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCapouskova, K., Kringelbach, M.L., Deco, G.: Modes of cognition: Evidence from metastable brain dynamics. NeuroImage. \u003cb\u003e260\u003c/b\u003e, 119489 (2022)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoney, C.J., Newman, E.L., Schapiro, A.C.: Switching between internal and external modes: A multiscale learning principle. Netw. Neurosci. \u003cb\u003e1\u003c/b\u003e, 339\u0026ndash;356 (2017)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang, Z.: Common and unique network basis for externally and internally driven flexibility in cognition: From a developmental perspective. Dev. Cogn. Neurosci. (2025)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeilnhammer, V., Stuke, H., Standvoss, K., Sterzer, P.: Sensory processing in humans and mice fluctuates between external and internal modes. PLoS Biol. \u003cb\u003e21\u003c/b\u003e, e3002410 (2023)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXiong, B., Chen, C., Tian, Y., Zhang, S., Liu, C., Evans, T.M., Fern\u0026aacute;ndez, G., Wu, J., Qin, S.: Brain preparedness: The proactive role of the cortisol awakening response in hippocampal-prefrontal functional interactions. Prog. Neurobiol. \u003cb\u003e205\u003c/b\u003e, 102127 (2021)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCole, M.W., Bassett, D.S., Power, J.D., Braver, T.S., Petersen, S.E.: Intrinsic and Task-Evoked Network Architectures of the Human Brain. Neuron. \u003cb\u003e83\u003c/b\u003e, 238\u0026ndash;251 (2014)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCole, M.W., Ito, T., Bassett, D.S., Schultz, D.H.: Activity flow over resting-state networks shapes cognitive task activations. Nat. Neurosci. \u003cb\u003e19\u003c/b\u003e, 1718\u0026ndash;1726 (2016)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFinc, K., Bonna, K., Lewandowska, M., Wolak, T., Nikadon, J., Dreszer, J., Duch, W., K\u0026uuml;hn, S.: Transition of the functional brain network related to increasing cognitive demands. Hum. Brain. Mapp. \u003cb\u003e38\u003c/b\u003e, 3659\u0026ndash;3674 (2017)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHearne, L.J., Cocchi, L., Zalesky, A., Mattingley, J.B.: Reconfiguration of Brain Network Architectures between Resting-State and Complexity-Dependent Cognitive Reasoning. J. Neurosci. \u003cb\u003e37\u003c/b\u003e, 8399\u0026ndash;8411 (2017)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBassett, D.S., Wymbs, N.F., Porter, M.A., Mucha, P.J., Carlson, J.M., Grafton, S.T.: Dynamic reconfiguration of human brain networks during learning. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e 108, 7641\u0026ndash;7646 (2011)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBassett, D.S., Yang, M., Wymbs, N.F., Grafton, S.T.: Learning-induced autonomy of sensorimotor systems. Nat. Neurosci. \u003cb\u003e18\u003c/b\u003e, 744\u0026ndash;751 (2015)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKucyi, A., Esterman, M., Riley, C.S., Valera, E.M.: Spontaneous default network activity reflects behavioral variability independent of mind-wandering. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e 113, 13899\u0026ndash;13904 (2016)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSpreng, R.N., Stevens, W.D., Chamberlain, J.P., Gilmore, A.W., Schacter, D.L.: Default network activity, coupled with the frontoparietal control network, supports goal-directed cognition. NeuroImage. \u003cb\u003e53\u003c/b\u003e, 303\u0026ndash;317 (2010)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVatansever, D., Menon, D.K., Manktelow, A.E., Sahakian, B.J., Stamatakis, E.A.: Default Mode Dynamics for Global Functional Integration. J. Neurosci. \u003cb\u003e35\u003c/b\u003e, 15254\u0026ndash;15262 (2015)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVatansever, D., Menon, D.K., Stamatakis, E.A.: Default mode contributions to automated information processing. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e 114, 12821\u0026ndash;12826 (2017)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eErb, C.D., Germine, L., Hartshorne, J.K.: Cognitive control across the lifespan: Congruency effects reveal divergent developmental trajectories. J. Exp. Psychol. Gen. \u003cb\u003e152\u003c/b\u003e, 3285\u0026ndash;3291 (2023)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLaPlume, A.A., Anderson, N.D., McKetton, L., Levine, B., Troyer, A.K.: When I\u0026rsquo;m 64: Age-Related Variability in Over 40,000 Online Cognitive Test Takers. Journals Gerontology: Ser. B. \u003cb\u003e77\u003c/b\u003e, 104\u0026ndash;117 (2022)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRey-Mermet, A., Gade, M.: Inhibition in aging: What is preserved? What declines? A meta-analysis. Psychon Bull. Rev. \u003cb\u003e25\u003c/b\u003e, 1695\u0026ndash;1716 (2018)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRey-Mermet, A., Gade, M.: Age-related deficits in the congruency sequence effect are task-specific: An investigation of nine tasks. Psychol. Aging. \u003cb\u003e35\u003c/b\u003e, 744\u0026ndash;764 (2020)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSouza, A.S., Frischkorn, G.T., Oberauer, K.: Older yet sharp: No general age-related decline in focusing attention. J. Exp. Psychol. Gen. \u003cb\u003e153\u003c/b\u003e, 2552\u0026ndash;2572 (2024)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVer\u0026iacute;ssimo, J., Verhaeghen, P., Goldman, N., Weinstein, M., Ullman, M.T.: Evidence that ageing yields improvements as well as declines across attention and executive functions. Nat. Hum. Behav. \u003cb\u003e6\u003c/b\u003e, 97\u0026ndash;110 (2021)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarulli, D., Stern, Y.: Efficiency, capacity, compensation, maintenance, plasticity: emerging concepts in cognitive reserve. Trends Cogn. Sci. \u003cb\u003e17\u003c/b\u003e, 502\u0026ndash;509 (2013)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrady, C.: The cognitive neuroscience of ageing. Nat. Rev. Neurosci. \u003cb\u003e13\u003c/b\u003e, 491\u0026ndash;505 (2012)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRieck, J.R., Baracchini, G., Nichol, D., Abdi, H., Grady, C.L.: Reconfiguration and dedifferentiation of functional networks during cognitive control across the adult lifespan. Neurobiol. Aging. \u003cb\u003e106\u003c/b\u003e, 80\u0026ndash;94 (2021)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarulli, D., Stern, Y.: Efficiency, capacity, compensation, maintenance, plasticity: emerging concepts in cognitive reserve. Trends Cogn. Sci. \u003cb\u003e17\u003c/b\u003e, 502\u0026ndash;509 (2013)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCabeza, R., Albert, M., Belleville, S., Craik, F.I.M., Duarte, A., Grady, C.L., Lindenberger, U., Nyberg, L., Park, D.C., Reuter-Lorenz, P.A., Rugg, M.D., Steffener, J.: Rajah, Maintenance, reserve and compensation: the cognitive neuroscience of healthy ageing. Nat. Rev. Neurosci. \u003cb\u003e19\u003c/b\u003e, 701\u0026ndash;710 (2018)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTurner, G.R., Spreng, R.N.: Prefrontal Engagement and Reduced Default Network Suppression Co-occur and Are Dynamically Coupled in Older Adults: The Default\u0026ndash;Executive Coupling Hypothesis of Aging. J. Cogn. Neurosci. \u003cb\u003e27\u003c/b\u003e, 2462\u0026ndash;2476 (2015)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, Z., Xia, H., Chen, A.: Impaired brain ability of older adults to transit and persist to latent states with well-organized structures at wakeful rest. GeroScience. (2024). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s11357-024-01366-y\u003c/span\u003e\u003cspan address=\"10.1007/s11357-024-01366-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan der Meer, J.N., Breakspear, M., Chang, L.J., Sonkusare, S., Cocchi, L.: Movie viewing elicits rich and reliable brain state dynamics. Nat. Commun. \u003cb\u003e11\u003c/b\u003e, 5004 (2020)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQuinn, A.J., Vidaurre, D., Abeysuriya, R., Becker, R., Nobre, A.C.: Woolrich, Task-Evoked Dynamic Network Analysis Through Hidden Markov Modeling. Front. Neurosci. \u003cb\u003e12\u003c/b\u003e, 603 (2018)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStevner, A.B.A., Vidaurre, D., Cabral, J., Rapuano, K., Nielsen, S.F.V., Tagliazucchi, E., Laufs, H., Vuust, P., Deco, G., Woolrich, M.W., Van Someren, E., Kringelbach, M.L.: Discovery of key whole-brain transitions and dynamics during human wakefulness and non-REM sleep. Nat. Commun. \u003cb\u003e10\u003c/b\u003e, 1035 (2019)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVidaurre, D.: A new model for simultaneous dimensionality reduction and time-varying functional connectivity estimation. PLoS Comput. Biol. \u003cb\u003e17\u003c/b\u003e, e1008580 (2021)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVidaurre, D., Smith, S.M., Woolrich, M.W.: Brain network dynamics are hierarchically organized in time. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e 114, 12827\u0026ndash;12832 (2017)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIto, T., Kulkarni, K.R., Schultz, D.H., Mill, R.D., Chen, R.H., Solomyak, L.I., Cole, M.W.: Cognitive task information is transferred between brain regions via resting-state network topology. Nat. Commun. \u003cb\u003e8\u003c/b\u003e, 1027 (2017)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLuppi, A.I., Mediano, P.A.M., Rosas, F.E., Holland, N., Fryer, T.D., O\u0026rsquo;Brien, J.T., Rowe, J.B., Menon, D.K., Bor, D., Stamatakis, E.A.: A synergistic core for human brain evolution and cognition. Nat. Neurosci. \u003cb\u003e25\u003c/b\u003e, 771\u0026ndash;782 (2022)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShine, J.M., Bissett, P.G., Bell, P.T., Koyejo, O., Balsters, J.H., Gorgolewski, K.J., Moodie, C.A., Poldrack, R.A.: The Dynamics of Functional Brain Networks: Integrated Network States during Cognitive Task Performance. Neuron. \u003cb\u003e92\u003c/b\u003e, 544\u0026ndash;554 (2016)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaker, A.P., Brookes, M.J., Rezek, I.A., Smith, S.M., Behrens, T., Probert Smith, P.J., Woolrich, M.: Fast transient networks in spontaneous human brain activity. \u003cem\u003eeLife\u003c/em\u003e 3, e01867 (2014)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShappell, H., Caffo, B.S., Pekar, J.J., Lindquist, M.A.: Improved state change estimation in dynamic functional connectivity using hidden semi-Markov models. NeuroImage. \u003cb\u003e191\u003c/b\u003e, 243\u0026ndash;257 (2019)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhiteside, D.J., Malpetti, M., Jones, P.S., Ghosh, B.C.P., Coyle-Gilchrist, I., Van Swieten, J.C., Seelaar, H., Jiskoot, L., Borroni, B., Sanchez‐Valle, R., Moreno, F., Laforce, R., Graff, C., Synofzik, M., Galimberti, D., Masellis, M., Tartaglia, M.C., Finger, E., Vandenberghe, R., De Mendon\u0026ccedil;a, A., Tagliavini, F., Butler, C.R., Santana, I., Ber, I.L., Gerhard, A., Ducharme, S., Levin, J., Danek, A., Otto, M., Sorbi, S., Pasquier, F., Bouzigues, A., Russell, L.L., Rohrer, J.D., Rowe, J.B., Rittman, T.: The GENFI consortium, Temporal dynamics predict symptom onset and cognitive decline in familial frontotemporal dementia. \u003cem\u003eAlzheimer\u0026rsquo;s \u0026amp; Dementia\u003c/em\u003e 19, 1947\u0026ndash;1962 (2023)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, K., Li, C., Sun, W., Tao, Y., Wang, R., Hou, W., Liu, D.-Q.: Hidden Markov Modeling Reveals Prolonged Baseline State and Shortened Antagonistic State across the Adult Lifespan. Cereb. Cortex. \u003cb\u003e32\u003c/b\u003e, 439\u0026ndash;453 (2022)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZeng, Y., Xiong, B., Gao, H., Liu, C., Chen, C., Wu, J., Qin, S.: Cortisol awakening response prompts dynamic reconfiguration of brain networks in emotional and executive functioning. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e 121, e2405850121 (2024)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLoaiza, V.M.: An overview of the hallmarks of cognitive aging. Curr. Opin. Psychol. (2024)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFinkel, D., Reynolds, C., Mcardle, J., Pedersen, N.: Age Changes in Processing Speed as a Leading Indicator of Cognitive Aging. Psychol. Aging. \u003cb\u003e22\u003c/b\u003e, 558\u0026ndash;568 (2007)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKane, M.J., Bleckley, M.K., Conway, A.R.A., Engle, R.W.: A controlled-attention view of working-memory capacity. J. Exp. Psychol. Gen. \u003cb\u003e130\u003c/b\u003e, 169\u0026ndash;183 (2001)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSalthouse, T.A.: Aging and measures of processing speed. Biol. Psychol. \u003cb\u003e54\u003c/b\u003e, 35\u0026ndash;54 (2000)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaltes, P.B., Lindenberger, U., Staudinger, U.M.: Life Span Theory in Developmental Psychology in \u003cem\u003eHandbook of Child Psychology\u003c/em\u003e, W. Damon, R. M. Lerner, Eds. (Wiley, ed. 1, ; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://onlinelibrary.wiley.com/doi/\u003c/span\u003e\u003cspan address=\"https://onlinelibrary.wiley.com/doi/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2007). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/9780470147658.chpsy0111\u003c/span\u003e\u003cspan address=\"10.1002/9780470147658.chpsy0111\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSala-Llonch, R., Bartr\u0026Atilde;\u0026copy;s-Faz, D., Junqu\u0026Atilde;\u0026copy;, C.: Reorganization of brain networks in aging: a review of functional connectivity studies. Front. Psychol. 6 (2015)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSpreng, R.N., Wojtowicz, M., Grady, C.L.: Reliable differences in brain activity between young and old adults: A quantitative meta-analysis across multiple cognitive domains. Neurosci. Biobehavioral Reviews. \u003cb\u003e34\u003c/b\u003e, 1178\u0026ndash;1194 (2010)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuccessful, Aging: : \u003cem\u003ePerspectives from the Behavioral Sciences.\u003c/em\u003eCambridge University Press, New York, NY, US, \u003cem\u003eSuccessful aging: Perspectives from the behavioral sciences.\u003c/em\u003e (1990)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaltes, P.B.: Theoretical Propositions of Life-Span Developmental Psychology: On the Dynamics Between Growth and Decline\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAron, L., Zullo, J., Yankner, B.A.: The adaptive aging brain. Curr. Opin. Neurobiol. \u003cb\u003e72\u003c/b\u003e, 91\u0026ndash;100 (2022)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePark, D.C., Reuter-Lorenz, P.: The Adaptive Brain: Aging and Neurocognitive Scaffolding. Annu. Rev. Psychol. \u003cb\u003e60\u003c/b\u003e, 173\u0026ndash;196 (2009)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAshby, F.G., Turner, B.O., Horvitz, J.C.: Cortical and basal ganglia contributions to habit learning and automaticity. Trends Cogn. Sci. \u003cb\u003e14\u003c/b\u003e, 208\u0026ndash;215 (2010)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGraybiel, A.M.: Habits, Rituals, and the Evaluative Brain. Annu. Rev. Neurosci. \u003cb\u003e31\u003c/b\u003e, 359\u0026ndash;387 (2008)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCole, M.W., Reynolds, J.R., Power, J.D., Repovs, G., Anticevic, A., Braver, T.S.: Multi-task connectivity reveals flexible hubs for adaptive task control. Nat. Neurosci. \u003cb\u003e16\u003c/b\u003e, 1348\u0026ndash;1355 (2013)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeco, G., Sanz Perl, Y., Ponce-Alvarez, A., Tagliazucchi, E., Whybrow, P.C., Fuster, J., Kringelbach, M.L.: One ring to rule them all: The unifying role of prefrontal cortex in steering task-related brain dynamics. Prog. Neurobiol. \u003cb\u003e227\u003c/b\u003e, 102468 (2023)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFinc, K., Bonna, K., He, X., Lydon-Staley, D.M., K\u0026uuml;hn, S., Duch, W., Bassett, D.S.: Dynamic reconfiguration of functional brain networks during working memory training. Nat. Commun. \u003cb\u003e11\u003c/b\u003e, 2435 (2020)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, X., Krieger-Redwood, K., Lyu, B., Lowndes, R., Wu, G., Souter, N.E., Wang, X., Kong, R., Shafiei, G., Bernhardt, B.C., Cui, Z., Smallwood, J., Du, Y., Jefferies, E.: The Brain\u0026rsquo;s Topographical Organization Shapes Dynamic Interaction Patterns That Support Flexible Behavior Based on Rules and Long-Term Knowledge. J. Neurosci. \u003cb\u003e44\u003c/b\u003e, e2223232024 (2024)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBraun, U., Sch\u0026auml;fer, A., Walter, H., Erk, S., Romanczuk-Seiferth, N., Haddad, L., Schweiger, J.I., Grimm, O., Heinz, A., Tost, H., Meyer-Lindenberg, A., Bassett, D.S.: Dynamic reconfiguration of frontal brain networks during executive cognition in humans. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e 112, 11678\u0026ndash;11683 (2015)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCole, M.W., Reynolds, J.R., Power, J.D., Repovs, G., Anticevic, A., Braver, T.S.: Multi-task connectivity reveals flexible hubs for adaptive task control. Nat. Neurosci. \u003cb\u003e16\u003c/b\u003e, 1348\u0026ndash;U247 (2013)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKitzbichler, M.G., Henson, R.N.A., Smith, M.L., Nathan, P.J., Bullmore, E.T.: Cognitive Effort Drives Workspace Configuration of Human Brain Functional Networks. J. Neurosci. \u003cb\u003e31\u003c/b\u003e, 8259\u0026ndash;8270 (2011)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCabeza, R., Albert, M., Belleville, S., Craik, F.I.M., Duarte, A., Grady, C.L., Lindenberger, U., Nyberg, L., Park, D.C., Reuter-Lorenz, P.A., Rugg, M.D., Steffener, J.: Rajah, Maintenance, reserve and compensation: the cognitive neuroscience of healthy ageing. Nat. Rev. Neurosci. \u003cb\u003e19\u003c/b\u003e, 701\u0026ndash;710 (2018)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKnights, E., Henson, R.N., Morcom, A.M., Mitchell, D.J., Tsvetanov, K.A.: Neural Evidence of Functional Compensation for Fluid Intelligence in Healthy Ageing (preprint, 2024); \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.7554/eLife.93327.1\u003c/span\u003e\u003cspan address=\"10.7554/eLife.93327.1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVaci, N., Gula, B., Bilalić, M.: Is age really cruel to experts? Compensatory effects of activity. Psychol. Aging. \u003cb\u003e30\u003c/b\u003e, 740\u0026ndash;754 (2015)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang, Y., Chen, Y., Sang, F., Zhao, S., Wang, J., Li, X., Chen, C., Chen, K., Zhang, Z.: Successful or pathological cognitive aging? Converging into a frontal preservation, temporal impairment (FPTI) hypothesis. Sci. Bull. \u003cb\u003e67\u003c/b\u003e, 2285\u0026ndash;2290 (2022)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang, L., Ross, B., Du, Y., Alain, C.: Long-term musical training can protect against age-related upregulation of neural activity in speech-in-noise perception. PLoS Biol. \u003cb\u003e23\u003c/b\u003e, e3003247 (2025)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHernandez, H., Santamaria-Garcia, H., Moguilner, S., Farina, F.R., Legaz, A., Prado, P., Cuadros, J., Gonzalez, L., Gonzalez-Gomez, R., Migeot, J., Coronel-Oliveros, C., Tagliazucchi, E., Maito, M.A., Godoy, M.E., Cruzat, J., Shaheen, A., Farombi, T., Salazar, D., Da Ros, L.U., Borelli, W.V., Zimmer, E.R., Njamnshi, A.K., Bajpai, S., Dey, A.B., Mostert, C., Merali, Z., Salama, M., Moustafa, S.A., Fittipaldi, S., Altschuler, F., Medel, V., Huepe, D., Yaffe, K., Momoh, C., Eyre, H.A., Swieboda, P., Lawlor, B., Miranda, J.J., Duran-Aniotz, C., Baez, S., Ibanez, A.: The exposome of healthy and accelerated aging across 40 countries. Nat. Med. \u003cb\u003e31\u003c/b\u003e, 3089\u0026ndash;3100 (2025)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAnguera, J.A., Reuter-Lorenz, P.A., Willingham, D.T., Seidler, R.D.: Failure to Engage Spatial Working Memory Contributes to Age-related Declines in Visuomotor Learning. J. Cogn. Neurosci. \u003cb\u003e23\u003c/b\u003e, 11\u0026ndash;25 (2011)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFernandez, N.B., Hars, M., Trombetti, A., Vuilleumier, P.: Age-related changes in attention control and their relationship with gait performance in older adults with high risk of falls. NeuroImage. \u003cb\u003e189\u003c/b\u003e, 551\u0026ndash;559 (2019)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNielson, K.A., Langenecker, S.A., Ross, T.J., Garavan, H., Rao, S.M., Stein, E.A.: Comparability of functional MRI response in young and old during inhibition. NeuroReport. \u003cb\u003e15\u003c/b\u003e, 129\u0026ndash;133 (2004)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTownsend, J., Adamo, M., Haist, F.: Changing channels: An fMRI study of aging and cross-modal attention shifts. NeuroImage. \u003cb\u003e31\u003c/b\u003e, 1682\u0026ndash;1692 (2006)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTombaugh, T.N., McIntyre, N.J.: The Mini-Mental State Examination: A Comprehensive Review. J. Am. Geriatr. Soc. \u003cb\u003e40\u003c/b\u003e, 922\u0026ndash;935 (1992)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu, P., Huang, R., Wang, J., Van Dam, N.T., Xie, T., Dong, Z., Chen, C., Gu, R., Zang, Y.-F., He, Y., Fan, J., Luo, Y.: Different topological organization of human brain functional networks with eyes open versus eyes closed. \u003cem\u003eNeuroImage\u003c/em\u003e 90, 246\u0026ndash;255 (2014)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, J., Wang, X., Xia, M., Liao, X., Evans, A., He, Y.: GRETNA: a graph theoretical network analysis toolbox for imaging connectomics. Front. Hum. Neurosci. \u003cb\u003e9\u003c/b\u003e (2015)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYan, C.-G., Craddock, R.C., Zuo, X.-N., Zang, Y.-F., Milham, M.P.: Standardizing the intrinsic brain: Towards robust measurement of inter-individual variation in 1000 functional connectomes. \u003cem\u003eNeuroImage\u003c/em\u003e 80, 246\u0026ndash;262 (2013)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVidaurre, D., Quinn, A.J., Baker, A.P., Dupret, D., Tejero-Cantero, A., Woolrich, M.W.: Spectrally resolved fast transient brain states in electrophysiological data. NeuroImage. \u003cb\u003e126\u003c/b\u003e, 81\u0026ndash;95 (2016)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVidaurre, D., Hunt, L.T., Quinn, A.J., Hunt, B.A.E., Brookes, M.J., Nobre, A.C., Woolrich, M.W.: Spontaneous cortical activity transiently organises into frequency specific phase-coupling networks. Nat. Commun. \u003cb\u003e9\u003c/b\u003e, 2987 (2018)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShine, J.M., Koyejo, O., Poldrack, R.A.: Temporal metastates are associated with differential patterns of time-resolved connectivity, network topology, and attention. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e 113, 9888\u0026ndash;9891 (2016)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTognoli, E., Kelso, J.A.S.: Metastable Brain Neuron. \u003cb\u003e81\u003c/b\u003e, 35\u0026ndash;48 (2014)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShen, X., Finn, E.S., Scheinost, D., Rosenberg, M.D., Chun, M.M., Papademetris, X., Constable, R.T.: Using connectome-based predictive modeling to predict individual behavior from brain connectivity. Nat. Protoc. \u003cb\u003e12\u003c/b\u003e, 506\u0026ndash;518 (2017)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFinn, E.S., Shen, X., Scheinost, D., Rosenberg, M.D., Huang, J., Chun, M.M., Papademetris, X., Constable, R.T.: Functional connectome fingerprinting: identifying individuals using patterns of brain connectivity. Nat. Neurosci. \u003cb\u003e18\u003c/b\u003e, 1664\u0026ndash;1671 (2015)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAhrends, C., Woolrich, M., Vidaurre, D.: Predicting individual traits from models of brain dynamics accurately and reliably using the Fisher kernel. [Preprint] (2024). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.7554/eLife.95125.1\u003c/span\u003e\u003cspan address=\"10.7554/eLife.95125.1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWobbrock, J.O., Findlater, L., Gergle, D., Higgins, J.J.: The aligned rank transform for nonparametric factorial analyses using only anova procedures in \u003cem\u003eProceedings of the SIGCHI Conference on Human Factors in Computing Systems\u003c/em\u003e (ACM, Vancouver BC Canada, ; (2011). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://dl.acm.org/doi/10.1145/1978942.1978963\u003c/span\u003e\u003cspan address=\"https://dl.acm.doi/10.1145/1978942.1978963\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), pp. 143\u0026ndash;146\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Cognitive Aging, Brain State Dynamics, Rest-To-Task Reconfiguration, Functional Compensation, fMRI","lastPublishedDoi":"10.21203/rs.3.rs-9298914/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9298914/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eEmerging research challenges the prevailing notion of uniform cognitive decline in aging, instead emphasizing the brain's adaptive ability that compensates for age-related deficits. Here we examined age-related differences in brain dynamics by focusing on rest-to-task reconfiguration, a process that reflects the brain\u0026rsquo;s ability to recruit neural resources for goal-directed behavior. Using hidden Markov modeling of whole-brain fMRI data, we characterized brain state transitions from rest to a Stroop task in young and older adults. Behaviorally, older adults exhibited generalized slowing yet preserved conflict resolution. Dynamically, both age groups could reconfigure from rest-specific towards a task-efficient brain state. However, while young adults preferentially engaged this state, older adults showed attenuated recruitment, instead maintaining a high-effort state from rest to task. Crucially, the functional benefit of this pattern was conditional: it supported conflict resolution only in high-accuracy older adults but presented detrimental in low performers. These findings demonstrate that cognitive preservation in aging depends not on preserving youthful brain reconfiguration dynamics, but on the effective recruitment of compensatory brain states.\u003c/p\u003e","manuscriptTitle":"Dynamic Rest-to-Task Brain Network Reconfiguration Reveals Divergent Cognitive Changes in Aging","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-21 15:19:51","doi":"10.21203/rs.3.rs-9298914/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"communications-biology","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"commsbio","sideBox":"Learn more about [Communications Biology](http://www.nature.com/commsbio/)","snPcode":"","submissionUrl":"","title":"Communications Biology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Communications Series","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"44e8157a-42e9-4684-be94-99da5cb5530a","owner":[],"postedDate":"April 21st, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"This content is not available.","date":"2026-05-05T05:15:27+00:00","index":1,"fulltext":"This content is not available."},{"type":"editorInvitedReview","content":"This content is not available.","date":"2026-04-30T09:23:41+00:00","index":2,"fulltext":"This content is not available."}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":65875415,"name":"Biological sciences/Neuroscience/Cognitive ageing"},{"id":65875416,"name":"Health sciences/Biomarkers/Predictive markers"}],"tags":[],"updatedAt":"2026-04-21T15:19:51+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-21 15:19:51","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9298914","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9298914","identity":"rs-9298914","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}