Distinct Effects of Data Occupational Complexity On Cognitive Aging: Evidence for Dual Brain–Cognitive Reserve Pathways

Distinct Effects of Data Occupational Complexity On Cognitive Aging: Evidence for Dual Brain–Cognitive Reserve Pathways | 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 Research Article Distinct Effects of Data Occupational Complexity On Cognitive Aging: Evidence for Dual Brain–Cognitive Reserve Pathways Lingruina Xu, Lei Yu, Jiawen Liu, Ting Li, Lixia Yang, Kewei Chen, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9143855/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Introduction : Occupational complexity is a major source of long-term cognitive stimulation across adulthood, yet its multidimensional effects on cognitive aging and their neural mechanisms remain unclear. This study examined how three dimensions of occupational complexity—data, people, and things—shape late-life cognitive performance, and whether brain reserve(BR) and cognitive reserve(CR) mediate these associations. Methods : A total of 3,754 retirees meeting sex-specific age eligibility criteria were recruited from the Beijing Aging Brain Rejuvenation Initiative completed a battery of multidomain neuropsychological assessments and standardized occupational history assessments coded using O*NET-based ratings.. A subsample of 851 participants also underwent structural MRI. Linear models, structural equation modeling, and voxel-/surface-based morphometry were used to test (1) dimension-specific associations with cognitive domains, (2) links to global and regional brain structure, and (3) dual-reserve mediation pathways. Results : Data complexity emerged as the primary protective dimension, independently predicting higher reasoning (β = 0.058), attention (β = 0.041), and reduced mild cognitive impairment(MCI) risk (− 5.7% after adjustment for education, p < 0.05; −14.3% unadjusted, p < 0.001). After accounting for data complexity, higher people complexity was associated with poorer working memory and language performance.. Things complexity, referring to demands related to tools and physical objects, showed no direct associations but demonstrated compensation via late-life leisure activities. Neuroimaging showed that data complexity was uniquely associated with larger gray matter volume in frontotemporal–limbic regions and higher CR. Mediation models revealed that data complexity protected cognition via both BR and CR, with mediation effects observed across multiple cognitive domains.. Conclusions : Occupational complexity, particularly data complexity, is associated with enhanced cognitive aging outcomes, including improved reasoning, attention, and reduced mild cognitive impairment (MCI) risk. Neuroimaging revealed that data complexity predicts greater frontotemporal–limbic gray matter volume and higher cognitive reserve. Mediation analyses suggested dual reserve pathways, with cognitive reserve mediating multiple cognitive domains, while brain reserve influenced hippocampal and temporal regions. These findings underscore the role of occupational environments in promoting cognitive health and mitigating late-life MCI risk. Occupational complexity Cognitive aging BR CR Structural MRI Life-course exposures Retirement Cognitive resilience Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1 Introduction The continuously accelerated aging of global population has made healthy cognitive aging a major public health challenge. Cognitive impairment is one of the most devastating health problems associated with aging and poses a particularly severe burden in China, where the absolute number of individuals living with dementia is the highest worldwide, exceeding 15 million and accounting for nearly one quarter of the global dementia population [1] . It becomes a crucial topic in healthy aging research to identify modifiable protective factors for cognitive aging. In recent years, occupation has attracted a growing attention as a potential protective factor, on its own or as a result of education, for healthy cognitive aging [2, 3] . Because work takes a large proportion of time across lifespan, particularly in adulthood, occupational complexity may play a predominant role in cognitive functions, including cognitive trajectories well after retirement. Theoretically, the “use it or lose it” hypothesis emphasized the promoting effect of continuous cognitive engagement on neuroplasticity [4] . Similarly, the theory of environmental complexity highlighted the cumulative effects of cognitive reserve (CR) as assessed by the characteristics of professional tasks (such as decision variations, conflicting information resolution) [5] . Most researchers use metrics of occupational complexity to examine the impact of occupations on cognitive abilities in older adults. Occupational complexity (OC), as a multi-dimensional concept to quantify cognitive stimulation, has advantages in two aspects: (1) it quantifies the complexity through objective characteristics of tasks; (2) it systematically covers a three-dimension concept of “data-people-things” [6] , thus enabling the analysis of the effects of each dimension independently and their interactions on cognition. Briefly, data complexity refers to cognitive demands related to information processing, reasoning, and decision-making; people complexity captures the extent of interpersonal interaction, communication, and coordination at work; whereas things complexity reflects demands associated with handling tools, machines, or physical objects. Most longitudinal studies have shown that high people complexity (e.g., management or education) is associated with a lower risk of dementia and better cognitive function, and the protective effect remains significant after controlling for education [7, 8] . The role of data complexity remains controversial: Some studies show that it can enhance processing speed and memory [9] , while others suggest that prolonged high-intensity work may actually increase the risk of Alzheimer's disease due to chronic stress [10] . The influence of things complexity is generally weak and unstable, which may depend on the task novelty rather than the complexity itself [11] . Furthermore, the cognitive effect of occupational complexity on cognitive function is moderated by the retirement transition and leisure activities in old age, suggesting that the effect is context-dependency [12, 13] . Overall, high occupational complexity is generally associated with better cognitive ability in old age, with significant heterogeneity across dimensions. Neuroimaging studies provide crucial evidence supporting the mechanisms of brain reserve (BR) and CR in relation to occupational complexity. BR (i.e., a neuroanatomic resource reflecting structural properties that afford surplus capacity to tolerate neuropathology) and CR (i.e., the acquired skills and flexible processing strategies that allow more efficient use of remaining neural resources) together can explain why individuals with similar brain pathology often show different clinical outcomes [14] . Research on BR and CR clarifies interindividual variability in cognitive aging trajectory, informs prevention strategies for dementia, and guides interventions to promote resilient cognitive aging [15, 16] . Experimental and observational studies further indicate that life-course activities — including mentally stimulating work — modulate both reserves via neuroplastic mechanisms [17] . Furthermore, the relative contributions of BR and CR may differ across pathologies and cognitive domains [18] . Consistent with with reserve-based theories of cognitive aging rooted in environmental complexity, evidence suggests that occupational complexity contributes to brain reserve accumulation across the life course. Longitudinal and cross-sectional studies in older adults demonstrate that supervisory and managerial experience in midlife is associated with larger hippocampal and medial temporal gray matter volumes and substantially slower rates of late-life hippocampal atrophy, independent of demographic, genetic, and late-life lifestyle factors [19] . Extending these findings to earlier stages of adulthood, occupational cognitive complexity during early to midadulthood has also been linked to higher white matter integrity and superior executive functioning in midlife, indicating that work-related cognitive demands may shape neural substrates of reserve decades before old age [20] . These findings suggest that complex professional experiences promote BR accumulation, in light of the "brain maintenance" theory [21] . Furthermore, even when experiencing brain atrophy, individuals with high occupational complexity can maintain relatively preserved cognitive function [22] , through CR by which the brain responds to aging and pathological changes by more efficient and flexible resource utilization. Moreover, different occupational complexity dimensions showed distinct effects on different brain regions. For example, high physical demands correlate with cortical thickness in pre- and postcentral gyri [23] , whereas high-pressure working environment may be associated with the degeneration of the prefrontal cortex and the limbic structures [24, 25] . However, there remained several key issues understudied. First, it is unclear about the specific mechanisms through which different dimensions of occupational complexity affect cognitive function. In particular, the relative contributions of retirement timing and post-retirement lifestyle require further clarifications. Second, it remains unclear whether the effects might be mediated by BR, CR, or both. Finally, further research is needed to elucidate how occupational complexity contributes to post-retirement cognitive engagement and brain structural maintenance, and whether late-life activities reinforce or compensate for the reserve mechanisms established during one’s occupational years. Therefore, this study used a multi-modal data analysis approach to systematically examine the differential effects of different occupational complexity dimensions on cognitive function in old age. We aim to (1) analyze the heterogeneous effects of data, people, and things complexity on specific cognitive domains; (2) reveal the association patterns between different dimensions of occupational complexity and brain structural changes; and (3) construct a mediation model of “occupational complexity–BR/CR–cognitive function” to elucidate the associated cognitive protection mechanism. This study will provide a theoretical and empirical basis for developing cognitive intervention strategies based on occupational characteristics 2. Methods 2.1 Participants A total of 3,754 community-dwelling retirees aged 50 years and older participated in the Beijing Aging Brain Rejuvenation Initiative (BABRI), an ongoing community-based prospective cohort study in China focusing on higher-order cognitive aging and its neural mechanisms (Fig. 1 ).The study protocol was approved by the Institutional Review Boards of the Beijing Normal University Imaging Center for Brain Research and Beijing Tiantan Hospital and was conducted in accordance with the Declaration of Helsinki. All participants provided informed consent in written format. Participants were screened with the following inclusion criteria: (1) retired individuals (≥ 50 years for women, ≥ 55 years for men); (2) native Chinese speakers; (3) right-handed; and (4) with normal or corrected-to-normal vision and hearing, as well as the following exclusion criteria: (1) a history of major neurological disorders (e.g., stroke, intracranial lesions); (2) a current clinical diagnosis or a history of neuropsychiatric conditions (e.g., depression, schizophrenia, or epilepsy); and (3) a history of substance abuse (e.g., alcohol dependence or sedative abuse). Among these participants, a sub-sample of 851 MRI eligible participants voluntarily underwent MRI scanning. All participants met the MRI safety requirements, with no metallic implants (e.g., dental prostheses, cardiac stents, orthopedic plates or screws), and had no history of claustrophobia, Ménière’s disease, or other ear-related disorders that might interfere with MRI procedures. All participants provided written informed consent prior to scanning. No participant withdrew during image acquisition, and all scans were successfully completed with high image quality and minimal head motion artifacts. Participants completed T1-weighted structural MRI and an extensive cognitive battery within a month. Mild cognitive impairment (MCI) was diagnosed according to Petersen’s criteria [26] . Specifically, participants were classified as having MCI if they met the following conditions: (1) the presence of subjective cognitive complaints reported by the participant and/or an informant; (2) preserved general cognitive functioning, defined as a Mini-Mental State Examination (MMSE) score greater than 23; (3) intact activities of daily living, indicated by a score of 0 (no functional impairment) on both the Activities of Daily Living (ADL) and Instrumental Activities of Daily Living (IADL) scales; and (4) objective cognitive impairment in at least one cognitive domain, operationalized as performance 1.5 standard deviations below the normative mean on at least one neuropsychological test [27] . Participants who did not meet the diagnostic criteria for MCI were classified as cognitively normal, with neuropsychological test scores within 1.5 standard deviations. 2.2 Measures 2.2.1 Occupational Complexity Occupational complexity was evaluated based on the matrix of U.S. occupations developed for the 1970 U.S. Census [28, 29] and standardized using the O*NET database to ensure cross-cultural comparability. Each participant’s primary occupation—defined as the occupation held for at least five years—was rated along three dimensions: people complexity(0–8), data complexity(0–6), and things complexity(0–7). In the original coding scheme, higher scores indicated lower occupational demands. For ease of interpretation, we reversed the scores so that higher values reflected greater occupational demands. The three dimensions were then summed to yield a total occupational complexity score ranging from 0 to 21, with higher scores indicating greater overall complexity [30] . 2.2.2 Retirement Duration and Leisure Activities after Retirement As the original dataset did not directly record the retirement duration, it was operationalized through retrospective calculation: Retirement Duration = Baseline Survey Year − Official Retirement Year. Leisure activities after retirement were assessed using a personal information questionnaire consisting of 23 items covering cognitive, social, and physical leisure activities. Participants were asked to recall their engagement in leisure activities during the past year, including reading, writing, attending senior university courses, playing chess, poker, or mahjong, and engaging in handcrafts. The frequency of each activity was defined as either “frequent” if being engaged several times per week or “rare” if being engaged for less than once per week. A weighted composite score across all 23 items was calculated, with higher scores indicating greater engagement in leisure activities in later life [31] . 2.2.3 Neuropsychological Test and Personal Information Questionnaire All participants completed a comprehensive neuropsychological battery to assess eight cognitive domains, including: (1) general cognition, assessed by the Mini-Mental State Examination (MMSE) total score; (2) episodic memory, assessed by the Auditory Verbal Learning Test (AVLT)—covering immediate recall (N1–N3), short-delay recall (N4), long-delay recall (N5), and total recall (N1–N5), and the Rey–Osterrieth Complex Figure Test (ROCF) delayed recall; (3) working memory, measured by the Digit Span Test (DST)—backward; (4) reasoning, assessed by the Similarities Test (SIM); (5) visuospatial processing, assessed by ROCF copy and the Clock Drawing Test (CDT); (6) language, evaluated by the Category Verbal Fluency Test (CVFT)—animal, vegetable, and fruit categories, and the Boston Naming Test (BNT); (7) attention, assessed by the Symbol Digit Modalities Test (SDMT) and the Trail Making Test A (TMT-A); and (8) executive function, assessed by the Trail Making Test B (TMT-B) and the Stroop Color-Word Test C (SCWT-C). Demographic and health information were collected using a standardized personal information questionnaire, including age, sex, and years of education as demographic variables, and chronic conditions such as hypertension, hyperlipidemia, diabetes, coronary heart disease, and cerebral small vessel disease. The number of cardiometabolic conditions, together with age and sex, were used as covariates in subsequent statistical analyses. 2.3 Statistic Analysis Analyses were conducted for both behavioral and neuroimaging data. All statistical analyses were performed using SPSS version 22.0 and R version 4.0. MRI morphological analyses were implemented using the Computational Anatomy Toolbox (CAT12; http://www.neuro.uni-jena.de/cat/ ) within the Statistical Parametric Mapping software (SPM12; https://www.fil.ion.ucl.ac.uk/spm/ ) running in the MATLAB (R2020a) environment. To improve interpretability of effect estimates, key continuous predictors were Z-standardized prior to analysis. Statistical significance was set at two-tailed p 3) were removed, and missing values were imputed using a linear regression method with 20 iterative estimations to ensure data completeness and robustness. Cognitive performance was evaluated across eight domains (general cognition, episodic memory, working memory, reasoning, visuospatial processing, language, attention, and executive function). Domain scores were Z-standardized and averaged to obtain composite indices; reverse-coded measures (e.g., TMT-A) were reciprocally transformed prior to standardization. Descriptive statistics were used to summarize demographic and cognitive characteristics. Group differences across levels of occupational complexity (people, data, and things) were examined using independent-sample t tests and chi-square tests. To assess the independent effects of occupational complexity on cognitive domains, multivariable linear regression models were constructed, in which people, data, and things complexity were entered simultaneously, with sex, age, education years, retirement duration, and cardiometabolic conditions entered hierarchically as covariates. Further, within a path analysis framework, the PROCESS macro (Model 1 and Model 4) was applied to examine the mediating role of late-life leisure activities and the moderating effect of retirement duration. Indirect effects were estimated using bias-corrected bootstrap procedures (5,000 resamples), and significance regions were determined via the Johnson–Neyman technique, distinguishing between “enhancement” and “compensation” patterns of association. 2.3.2 Neuroimaging To investigate the neural correlates of occupational complexity and to test hypothesized brain reserve pathways, primary neuroimaging analyses focused on whole-brain morphometric measures using voxel-based morphometry (VBM) and surface-based morphometry (SBM). VBM analyses involved whole-brain voxel-wise regression models and two-sample t tests to examine associations between occupational complexity dimensions and local gray matter volume, controlling for sex, age, retirement duration, cardiometabolic conditions, and total intracranial volume (TIV). SBM analyses were conducted using cortical thickness measures derived from the Desikan–Killiany atlas, with regression and group comparison models assessing associations between occupational complexity dimensions and regional cortical thickness. Covariates in SBM analyses were identical to those used in the VBM models, excluding TIV. As supporting analyses, associations between occupational complexity dimensions and global brain structural indices were examined, and exploratory region-of-interest (ROI) analyses based on the Automated Anatomical Labeling (AAL) atlas were performed to provide complementary regional validation of whole-brain findings. All whole-brain analyses were corrected for multiple comparisons using the Family-Wise Error (FWE) or FDR method, as appropriate. 2.3.3 CR Quantification and Mediation Modeling CR was quantified using the residual-based approach, which operationalizes cognitive reserve as the portion of cognitive performance not explained by structural brain integrity, an approach that has been well established in prior work [32–34] . Conceptually, this framework defines CR as better-than-expected cognitive functioning given measurable brain structure and relevant covariates. First, overall cognitive performance (observed value, \(\:{Y}_{cog}\) )was treated as the dependent variable, and total gray matter volume ( \(\:{X}_{GMV}\) )as the independent variable. GMV was selected because it represents a global macrostructural marker of brain integrity that has been explicitly incorporated into residual-based neuroimaging models of cognitive reserve, particularly in aging and neurodegenerative contexts [35] .A multiple linear regression model was constructed, controlling for sex( \(\:{X}_{sex}\) ), age( \(\:{X}_{age}\) ), retirement duration( \(\:{X}_{retire}\) ), cardiometabolic conditions( \(\:{X}_{disease}\) ), and total intracranial volume ( \(\:{X}_{TIV}\) ): $$\:{Y}_{cog}={\beta\:}_{0}+{\beta\:}_{1}{X}_{GMV}+{\beta\:}_{2}{X}_{sex}+{\beta\:}_{3}{X}_{age}+{\beta\:}_{4}{X}_{retire}+{\beta\:}_{5}{X}_{disease}+{\beta\:}_{6}{X}_{TIV}+ϵ$$ The predicted cognitive score was then computed as: $$\\hat :{{Y}_{cog}}=\hat {{\beta\:}}_{0}+\hat {{\beta\:}}_{1}{X}_{GMV}+\hat {{\beta\:}}_{2}{X}_{sex}+\hat {{\beta\:}}_{3}{X}_{age\:}+\hat {{\beta\:}}_{4}{X}_{retire}+\hat {{\beta\:}}_{5}{X}_{disease}+\hat {{\beta\:}}_{6}{X}_{TIV}$$ CR was defined as the residual difference between the observed and predicted cognitive scores: $$\:CR\:=\:{Y}_{cog}\hat -{{Y}_{cog}}$$ Higher CR values indicate better-than-expected cognitive performance given an individual’s level of brain structure. Differences in CR between occupational complexity groups were examined using t tests, and associations between continuous measures of occupational complexity and CR were assessed using partial correlations and multivariable regression models controlling for the same covariates. To evaluate whether CR functioned as a pathway linking occupational complexity to cognitive performance, mediation models were constructed using the PROCESS macro with 5,000 bootstrap resamples. These models estimated the indirect effect of occupational complexity on cognition through CR while adjusting for demographic and health variables. Similar mediation analyses were then conducted to examine BR as an alternative pathway, employing the same modeling strategy and covariates. 2.3.4 Model Robustness and Validation All regression models were evaluated for residual normality and multicollinearity among predictors, including the three occupational complexity dimensions and covariates, with variance inflation factors (VIFs) below 5, indicating acceptable model fit. The findings remained consistent across different sampling strategies and covariate adjustments, supporting the robustness of the results. 3. Results 3.1 Demographic and Cognitive Characteristics In the total sample (N = 3,754), participants in the high occupational complexity group were older, more often male, and had more years of education. They also engaged more frequently in cognitive, social, and physical activities, and exhibited lower levels of depressive symptoms, body mass index, and a lower prevalence of cardiometabolic conditions (all p < .05). 3.2 Associations Between Occupational Complexity Dimensions and Cognitive Function Across multiple cognitive domains, data complexity demonstrated a robust and consistent positive association with cognitive performance. Before controlling for education, data complexity predicted performance across all domains except for executive function (standardized β = 0.057–0.109, all p < .001) and was associated with a 14.3% reduction in MCI risk ( p < .001). After controlling for education, associations with global cognition, episodic memory, working memory, reasoning, visuospatial processing, and attention remained significant (β = 0.016–0.058, all p < .05), corresponding to a 5.7% reduction in MCI risk ( p < .05). The people complexity was only associated with reasoning (β = 0.034, p < .001) and attention (β = 0.018, p < .05), whereas the things complexity showed no significant effects. Moderation analyses indicated that retirement duration significantly moderated the association between occupational complexity and cognitive performance. Specifically, the positive effect of the data complexity on visuospatial processing was stronger with increasing retirement duration. (interaction β = 0.002, p = .029) and enhanced the predictive effect of the people complexity on visuospatial processing (interaction β = 0.003, p = .001), suggesting that the benefits of occupational complexity may accumulate over time.The moderating effects of retirement duration on the associations between occupational complexity dimensions and visuospatial processing are visually depicted in Fig. 2 . Mediation analyses further suggested that late-life activities constituted key behavioral pathways linking occupational complexity to cognitive performance. Specifically, data complexity was indirectly associated with attention through engagement in intelligent activities. This indirect pathway showed conditional effects ranging from 0.0071 to 0.0080 across low, medium, and high retirement-duration strata and was consistently stronger than the corresponding pathways through social activities (0.0028–0.0036) and physical activities (0.0019–0.0024), indicating that intelligent activity represented the strongest and most stable behavioral mediator linking data complexity to attention..In contrast, people complexity demonstrated a weaker but significant indirect association with visuospatial processing through social activities, with effects varying across retirement duration (0.0001–0.0018) . Moderated mediation models further revealed compensatory effects among individuals with low things complexity. Specifically, these participants showed greater cognitive gains in multiple domains, including episodic memory and attention, through more frequent intelligent (β = 0.19–0.24, p < .01) and social activities (β = 0.15–0.18, p < .05). Enhancing effects were observed only for reasoning (β = 0.12, p < .05) and executive function (β = 0.09, p < .05).The moderated mediation models illustrating the pathways from occupational complexity to cognitive outcomes through intelligent activity are presented in Fig. 3 . 3.3 BR as a Pathway Linking Occupational Complexity to Cognition To establish the structural basis for mediation analyses, we first examined the associations between occupational complexity and brain morphology in the MRI subsample. After adjusting for demographic and health covariates, regression analyses on global brain measures indicated that data complexity significantly predicted larger intracranial volume (β = 4.93, p < .05), whereas its associations with total gray matter and white matter volumes were not statistically significant. Region-specific analyses revealed robust structural correlates. After FDR correction, several associations remained significant (Fig. 4 ), including greater cortical thickness in the right inferior temporal gyrus for data complexity ( p = .011), and significant associations of things complexity with the right ventrolateral thalamic nucleus and the left middle frontal gyrus (both FDR-corrected p < .05). People complexity showed negative associations with cortical thickness in both the left anterior cingulate cortex (FDR-corrected p < .05) and the left inferior orbital frontal gyrus (FDR-corrected p = .015). These findings indicate that data-related occupational demands provide the strongest and most spatially extensive neural correlates relevant for subsequent brain-reserve analyses. Building on these structural associations, we then tested whether occupational complexity influenced cognitive outcomes via BR, indexed by hippocampal gray matter and regional cortical thickness. Hippocampal volume was selected as it reflects episodic memory and global cognition, and larger volume has been linked to experiences that increase brain reserve [36–38] . Cortical thickness was included as it captures local neuronal integrity, with greater thickness associated with higher reserve and better tolerance to cortical loss [15, 39, 40] . As shown in Fig. 5 , mediation models revealed that data complexity exerted significant indirect effects on multiple cognitive domains through structural markers of BR. Specifically, higher data complexity predicted greater left hippocampal gray matter volume and increased cortical thickness in the right inferior temporal gyrus, which in turn mediated performance in general cognition (β = 0.005–0.006, p < .05), episodic memory (β = 0.004, p < .05), language (β = 0.004, p < .05), and attention (β = 0.004, p < .05). A small negative indirect effect was observed for executive function (β = − 0.005, p < .05), suggesting domain-specific differences. While data complexity–related structural adaptations generally benefited memory, language, and attention, their influence on executive function was limited, consistent with prior findings that cognitive and brain reserve often show weaker effects on executive domains compared to memory or attention [18, 41, 42] . 3.4 CR as a Pathway Linking Occupational Complexity to Cognition CR was quantified using a residual-based modeling approach (see Methods) and examined as a potential mediator between occupational complexity and cognitive function. After adjusting for sex, age, years of education, and cardiometabolic conditions, data complexity remained an independent predictor (β = 0.19, p = .018). Mediation analyses demonstrated that CR played a central role in linking data complexity to cognitive performance. The detailed mediation effects of data complexity on cognitive domains through CR are summarized in Fig. 5 . CR significantly mediated the associations between data complexity and global cognition, reasoning, episodic memory, visuospatial processing, language, and executive function, with indirect effects ranging from β = 0.01 to 0.03 for most domains, and all 95% bootstrap confidence intervals excluding zero. A negative indirect effect was observed for executive function. Partial mediation effects were observed for working memory and attention. The largest total effect was observed for attention (β = 0.068, p < .001). Exploratory, uncorrected analyses further suggested potential contributions of the right inferior temporal gyrus and left hippocampus; however, these effects did not survive FDR correction. Together, these findings indicate that CR serves as a key mechanism through which data complexity supports cognitive performance in later life. (b) Forest plot displaying the standardized indirect effects (β) of occupational data complexity on eight cognitive domains. Estimates and 95% confidence intervals are shown for three specific pathways, distinguished by color and shape: Light blue circles: Indirect effects via the CR pathway (residual-based score), representing functional reserve. Royal blue squares: Indirect effects via the BR pathway mediated specifically by the left hippocampus. Navy blue triangles: Indirect effects via the BR pathway mediated specifically by the right inferior temporal gyrus. Error bars indicate 95% bias-corrected bootstrap confidence intervals. Intervals that do not cross the vertical dashed line ( x = 0) are statistically significant ( p < 0.05). The right-hand column lists the precise standardized point estimates and [95% CI] for each path. Note that while CR mediates effects across most domains, structural BR mediation is domain-specific, and a negative indirect effect is observed for executive function. 4 Discussion In the present sample, occupational complexity demonstrated clear dimension-specific associations with later-life cognition and brain structure. Data complexity showed the most robust and generalized benefits, predicting higher global cognition, reasoning, and attention after adjustment for education, demographic covariates, and cardiometabolic conditions. It was also associated with a lower risk of MCI and exhibited the broadest indirect pathways through CR and structural markers. In contrast, people complexity showed selective associations with reasoning and attention but was negatively related to cortical thickness in the left anterior cingulate cortex (ACC-L), whereas things complexity displayed only modest effects largely restricted to sensorimotor regions. Late-life cognitive and social engagement partially mediated several of these associations, and retirement duration amplified specific pathways. The cognitive and neural protective effects of occupational complexity exhibited marked dimensional heterogeneity. Data complexity emerged as the most robust protective factor in this study. Even after controlling for education, it independently contributed to global cognition, reasoning, and attention, and lower risk of MCI, consistent with previous findings that occupations with high cognitive demands delay cognitive decline [7, 9] . Mechanistically, data complexity was associated with larger gray matter volumes and greater cortical thickness in frontotemporal–limbic regions, including the hippocampus and inferior temporal gyrus, and was the only dimension showing a significant positive correlation with CR quantified via the residual-based approach. One plausible explanation for these advantages lies in the core cognitive features of data-intensive occupations, which emphasize relational processing, abstract reasoning, and the integration of multiple information streams. Tasks characterized by high data complexity require individuals to simultaneously coordinate multiple variables and rules, a form of relational complexity that has been shown to recruit large-scale cognitive control networks involving frontoparietal and cingulo-opercular systems, with the dorsolateral prefrontal cortex supporting flexible, trial-by-trial control and integration of task sets [43] . In addition, data-intensive work is associated with information-rich yet efficient neural processing. Neuroimaging evidence suggests that higher-order cognitive engagement is supported by brain activity patterns that are both highly informative and highly compressible, reflecting optimized representational efficiency rather than increased neural redundancy [44] . Such information-rich yet compressible neural activity patterns may reflect an efficient and flexible mode of processing during high-level cognitive engagement. This characterization is conceptually aligned with descriptions of neural efficiency in the cognitive reserve framework. Finally, the sustained and cumulative nature of data-related cognitive demands may facilitate long-term neuroplastic adaptation. Continuous engagement in abstract problem solving and decision-making has been linked to synaptic plasticity mechanisms that support adaptive reorganization of neural systems across multiple temporal scales [45] . Consistent with a life-course perspective, higher occupational cognitive complexity earlier in adulthood has been associated with better white matter integrity and executive function in midlife, capturing a potential pathway through which prolonged exposure to data complexity contributes to reserve accumulation over time [20] . These results indicate that data-intensive occupations can simultaneously enhance brain and CRs, optimizing both brain structure and neural efficiency, thereby providing broad protection across multiple cognitive domains [19, 21, 22] . We observed that people complexity did not uniformly confer cognitive benefits: although it modestly predicted better reasoning, it was also associated with poorer working memory and language performance, as well as reduced cortical thickness in the left anterior cingulate cortex (ACC-L). This pattern contrasts with several studies reporting protective effects of “people-oriented” work. For example, higher lifetime people complexity has been linked to better executive and episodic memory performance in the KHANDLE cohort [46] , and to lower risks of MCI/dementia and greater MRI-based BR in the SNAD study [47] . Yet other community-based or cross-cultural studies have failed to detect consistent cognitive benefits of people complexity after accounting for education, socioeconomic gradients, or heterogeneity in work environments [48] , suggesting that the effects of this dimension are not universally positive. Three sources of heterogeneity may help reconcile the discrepancies between our findings and prior reports suggesting protective effects of people complexity. One important source lies in cultural and occupational-context differences. Studies reporting strong benefits of people complexity typically draw samples with higher education, stable professional careers, and limited involvement in manual or informal labor [9, 47, 49] . In these contexts, interpersonal demands often reflect cognitively stimulating managerial or supervisory responsibilities. In contrast, our community-based cohort likely includes a broader range of people-oriented occupations—particularly service and caregiving roles characterized by low autonomy and high emotional demands. Such roles may transform “social engagement” into chronic stress rather than cognitive stimulation, echoing recent concerns regarding occupational generalizability [50, 51] . Beyond contextual variation, methodological heterogeneity in the operationalization of people complexity constitutes a further source of inconsistency across studies. The present study employs the Dictionary of Occupational Titles (DOT), which quantifies interpersonal complexity based on the frequency and intensity of interactions with others. Other studies have also used later occupational coding frameworks such as O*NET; however, both systems share a key limitation: they collapse qualitatively distinct forms of interpersonal demands into a single composite dimension. Methodological reviews have noted that such frameworks often fail to differentiate cognitively stimulating social tasks from emotionally taxing or low-autonomy interactional roles [52, 53] . This conflation reduces the ability of occupational complexity measures to capture the critical distinction between cognitively stimulating and emotionally draining forms of interpersonal work, potentially contributing to the mixed pattern of cognitive gains and structural costs observed for the interpersonal dimension in our cohort. Finally, differences in the motivational and regulatory demands of interpersonal work offer a plausible mechanistic account for the observed trade-offs. High-autonomy interpersonal roles typically support goal-directed cognitive engagement and recruit prefrontal systems associated with reserve accrual, including the frontal pole and dorsolateral prefrontal cortex [54, 55] . In contrast, service- and caregiving-oriented roles impose sustained emotional labor, requiring continuous emotion regulation, conflict management, and behavioral compliance, which is associated with elevated allostatic load, reduced cognitive flexibility, and executive-resource competition [56, 57] . The magnitude of these trade-offs may further be shaped by cultural norms regarding social roles and relational expectations. In collectivistic contexts that emphasize interdependence and social harmony, interpersonal occupations often carry stricter demands for emotional regulation, role compliance, and collaboration, potentially intensifying cognitive and physiological costs [58–60] . By contrast, in individualistic contexts valuing autonomy and self-expression, similar interpersonal demands may be experienced as less taxing, allowing more open communication and active engagement [58, 61] . Collectively, these mechanisms are consistent with the “resource competition” hypothesis aligns with the profile observed in our cohort, wherein modest improvements in reasoning coexisted with deficits in working memory and language. The protective effects of things complexity were the most limited, being associated only with gray matter volume in motor-related regions and failing to contribute to CR, which may be because occupations high in things complexity predominantly involve repetitive, routine tasks with limited innovation or problem-solving, providing minimal stimulation to neural circuits critical for higher-order cognitive processes [62, 63] , reflecting the constraints of mechanized tasks in driving higher-order neuroplasticity [64] . However, individuals with low things complexity could achieve cross-domain cognitive compensation through engagement in late-life cognitive and social activities, highlighting the value of alternative environmental stimulation [13] . Furthermore, distinguishing between BR and CR enabled a more precise delineation of the neural pathways through which occupational complexity contributes to cognitive aging. Data complexity was the only dimension simultaneously associated with preserved structural markers and higher CR quantified using the residual-based approach (Methods). Together, these patterns support a dual-pathway model, in which structural preservation provides an anatomical buffer (BR), whereas the residual cognitive advantage reflects more efficient or flexible neural processing (CR). Importantly, the CR index captures variance in cognitive performance unexplained by contemporaneous brain structure, allowing it to be empirically separable from BR. Whereas brain reserve captures the absolute capacity of neural resources, including their structural quantity and integrity, cognitive reserve reflects the efficiency and adaptability of the neural processes that support performance despite age-related structural decline. This methodological distinction permits the dissociation of structural preservation from processing-level compensation, which is essential for isolating the unique contribution of occupational experiences. At the regional level, the hippocampus—central to episodic encoding, retrieval, and attentional allocation—has been consistently associated with superior cognitive performance in aging cohorts and frequently identified as a structural substrate of occupational or lifestyle reserve proxies [65, 66] . The inferior temporal gyrus (ITG), which supports high-level visual–semantic representation and object/word recognition, may further contribute to cross-domain transfer effects involving language, reasoning, and visuospatial processing. Recent evidence indicates that occupation-related reserve proxies correspond to increased ITG volume and to CR-related moderation of pathology–performance associations [67] . By contrast, people- and things-related complexity showed more restricted and domain-specific neural correlates. People complexity was linked to a localized reduction in cortical thickness within the left anterior cingulate cortex (ACC-L) and did not coincide with higher residual cognitive reserve. Considering the ACC’s well-established involvement in conflict monitoring, emotion regulation, and effort allocation—as well as its documented susceptibility to prolonged emotional and social demands—this focal reduction in thickness is more plausibly interpreted as reflecting chronic functional load on this region [68] . This neural profile is consistent with the selective pattern of cognitive effects observed in the current study, in which slight gains in reasoning ability coexisted with lower performance in working memory and language. In contrast, things complexity was primarily associated with structural variation in sensorimotor cortices, which support motor and perceptual functions, suggesting that object- or skill-focused occupations may enhance localized BR without generating cross-domain CR benefits [23, 69] . Together, these dissociable patterns highlight the complementary roles of BR and CR in the context of the present study, in which BR was operationalized through region-specific structural markers, including left hippocampal volume and right inferior temporal gyrus cortical thickness, and CR was quantified using a residual-based modeling approach capturing cognitive performance beyond what could be predicted from brain structure. This methodological distinction likely contributes to the observed dissociation: BR reflects structural reinforcement shaped by data-related occupational demands, whereas CR captures broader, task-independent processing advantages, allowing individuals to maintain cognitive performance despite age-related or localized structural decline. Moreover, from a life-course perspective, this study confirmed that the protective effects of occupational complexity can persist and be reinforced after retirement through engagement in late-life activities. Path analyses indicated that high occupational complexity, particularly in the data dimension, served as an important predictor of sustained participation in cognitively stimulating activities post-retirement. Notably, retirement duration positively moderated the pathway from “occupational complexity → cognitive activity → cognitive function.” This suggests that cognitive and BR accumulated during the occupational period provide a solid foundation for cognitive health in later life, while leisure activities after retirement play a role in activating and maintaining these reserves. As retirement duration increases, the protective effects, grounded in occupational experience and driven by late-life engagement, become increasingly pronounced, forming a synergistic effect across the lifespan [70] . Importantly, the observation that these protective effects become more evident with longer retirement duration may reflect an adaptive process during the post-retirement transition. The initial phase after retirement can involve a sudden loss of structured mental demands, which may temporarily attenuate cognitive benefits, as suggested by longitudinal findings showing accelerated decline in some cognitive domains immediately post-retirement [71] . Over time, however, individuals with higher occupational complexity are likely better equipped to navigate this transition, gradually re-establishing cognitively stimulating routines and engaging in leisure activities that reinforce previously accumulated cognitive and brain reserves [72] . These findings emphasize the long-term value of cognitive stimulation during the occupational period and integrate it with lifestyle factors in retirement to form a coherent model of healthy cognitive aging. The present study makes several key contributions to the literature on occupational complexity and cognitive aging. By disentangling the heterogeneous effects of the three occupational complexity dimensions, our findings refine and extend the theoretical framework of environmental complexity. Building on this distinction, we further propose and empirically validate a dual-pathway model linking occupational complexity to both brain structure and cognitive reserve, thereby elucidating their synergistic yet dissociable mechanisms. Importantly, by situating these pathways within a life-course perspective, the study highlights the dynamic interplay between midlife occupational experience and late-life engagement, offering an integrative account of cognitive aging trajectories.Taken together, these findings not only consolidate previously fragmented evidence but also yield actionable implications for the development of targeted cognitive interventions and policy strategies aimed at promoting healthy aging. Nonetheless, certain limitations should be noted. The cross-sectional and short-term follow-up design restricts causal inference. Although the O*NET indicators offer objective quantification, they cannot fully capture within-occupation heterogeneity. In addition, the cognitive reserve index derived in the present study reflects a model-dependent construct that is contingent on the specific cognitive and neuroanatomical variables included in the estimation procedure. Because CR was inferred from individual differences in cognitive performance relative to structural brain measures, its magnitude may vary with the choice of cognitive domains assessed and the extent to which the selected brain markers capture relevant neurobiological variation. As a result, the residual-based CR measure may incorporate sources of variance that are not uniquely attributable to reserve-related processes, and its comparability across studies employing different variable sets may be limited. Finally, potential biases in self-reported late-life activities warrant caution. Future studies should adopt longitudinal designs with harmonized cognitive and neural assessments to further refine reserve quantification and to validate the dynamic effects of occupational complexity on cognitive aging, while exploring occupation-specific intervention strategies tailored to diverse occupational trajectories. 5 Conclusion This study provides large-scale behavioral and neuroimaging evidence that occupational complexity constitutes an important life-course determinant of late-life cognitive health. Data complexity stands out as the primary driver, conferring broad cognitive benefits through coordinated enhancement of BR and CR. In contrast, people and things complexity exert narrower or context-dependent effects, suggesting substantial heterogeneity in how occupational environments shape cognitive aging. By delineating these domain-specific pathways and identifying reserve-based mechanisms, our findings highlight occupational complexity as a modifiable and quantifiable target for dementia prevention strategies. Interventions that enrich cognitive stimulation during working life or reinforce reserve accumulation after retirement may yield long-term benefits for maintaining cognitive resilience in aging populations. Abbreviations BR Brain Reserve CR Cognitive Reserve MCI Mild Cognitive Impairment MRI Magnetic Resonance Imaging BABRI Beijing Aging Brain Rejuvenation Initiative ADL Activities of Daily Living IADL Instrumental Activities of Daily Living MMSE Mini-Mental State Examination AVLT Auditory Verbal Learning Test ROCF Rey–Osterrieth Complex Figure Test DST Digit Span Test SIM Similarities Test CDT Clock Drawing Test CVFT Category Verbal Fluency Test BNT Boston Naming Test SDMT Symbol Digit Modalities Test TMT Trail Making Test SCWT-C Stroop Color-Word Test C FDR False Discovery Rate FWE Family-Wise Error VBM Voxel-Based Morphometry SBM Surface-Based Morphometry ROI Region of Interest AAL Automated Anatomical Labeling GMV Gray Matter Volume TIV Total Intracranial Volume VIFs Variance Inflation Factors SD Standard Deviation ITG Inferior Temporal Gyrus ACC Anterior Cingulate Cortex Declarations Ethics approval and consent to participate The study was conducted in accordance with the Declaration of Helsinki and was approved by the institutional review board (IRB) at the State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University (protocol code was ICBIR_A_0041_002_02). All participants provided written informed consent for our protocol. Consent for publication Not applicable Competing interests The authors declare no competing interests. Funding This work was supported by STI2030-Major Projects (2022ZD0211600), Beijing Natural Science Foundation (5262011), Open Research Fund of the State Key Laboratory of Cognitive Neuroscience and Learning and Tang Scholar. Author Contribution LRNX, JW, XL, and ZJZ contributed to the conception and design of the study; LY, JWL, and TL contributed to the acquisition and analysis of data; LRNX, and LY contributed to drafting the original manuscript and preparing the figures; LXY, KWC, XL, and ZJZ contributed to the reviewing and editing the manuscript. All authors approved the final manuscript. Acknowledgements The authors would like to express their gratitude to the participants and staff involved in data collection and management in the Beijing Aging Brain Rejuvenation Initiative. Data Availability The datasets used and analysed during the current study are available from the corresponding author on reasonable request. References Jia L, Quan M, Fu Y, Zhao T, Li Y, Wei C, Tang Y, Qin Q, Wang F, Qiao Y, Shi S. Dementia in China: epidemiology, clinical management, and research advances. The Lancet Neurology. 2020 Jan 1;19(1):81–92. Chapko D, McCormack R, Black C, Staff R, Murray A. Life-course determinants of cognitive reserve (CR) in cognitive aging and dementia–a systematic literature review. Aging & mental health. 2018 Aug 3;22(8):921 − 32. Lövdén M, Fratiglioni L, Glymour MM, Lindenberger U, Tucker-Drob EM. Education and cognitive functioning across the life span. Psychological science in the public interest. 2020 Aug;21(1):6–41. Hultsch DF, Hertzog C, Small BJ, Dixon RA. Use it or lose it: engaged lifestyle as a buffer of cognitive decline in aging?. Psychology and aging. 1999 Jun;14(2):245. Schooler C. Psychological effects of complex environments during the life span: A review and theory. Intelligence. 1984 Oct 1;8(4):259 − 81. Cain PS, Treiman DJ. The dictionary of occupational titles as a source of occupational data. American Sociological Review. 1981 Jun 1:253 − 78. Andel R, Crowe M, Pedersen NL, Mortimer J, Crimmins E, Johansson B, Gatz M. Complexity of work and risk of Alzheimer's disease: a population-based study of Swedish twins. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences. 2005 Sep 1;60(5):P251-8. Karp A, Andel R, Parker MG, Wang HX, Winblad B, Fratiglioni L. Mentally stimulating activities at work during midlife and dementia risk after age 75: follow-up study from the Kungsholmen Project. The American journal of geriatric psychiatry. 2009 Mar 1;17(3):227 − 36. Smart EL, Gow AJ, Deary IJ. Occupational complexity and lifetime cognitive abilities. Neurology. 2014 Dec 9;83(24):2285-91. Kröger E, Andel R, Lindsay J, Benounissa Z, Verreault R, Laurin D. Is complexity of work associated with risk of dementia? The Canadian Study of Health and Aging. American journal of epidemiology. 2008 Apr 1;167(7):820 − 30. Oltmanns J, Godde B, Winneke AH, Richter G, Niemann C, Voelcker-Rehage C, Schömann K, Staudinger UM. Don’t lose your brain at work–The role of recurrent novelty at work in cognitive and brain aging. Frontiers in Psychology. 2017 Feb 6;8:117. Finkel D, Andel R, Gatz M, Pedersen NL. The role of occupational complexity in trajectories of cognitive aging before and after retirement. Psychology and aging. 2009 Sep;24(3):563. Andel R, Silverstein M, Kåreholt I. The role of midlife occupational complexity and leisure activity in late-life cognition. Journals of Gerontology Series B: Psychological Sciences and Social Sciences. 2015 Mar 1;70(2):314 − 21. Stern Y, Barnes CA, Grady C, Jones RN, Raz N. Brain reserve, cognitive reserve, compensation, and maintenance: operationalization, validity, and mechanisms of cognitive resilience. Neurobiology of aging. 2019 Nov 1;83:124-9. Pettigrew C, Soldan A. Defining cognitive reserve and implications for cognitive aging. Current neurology and neuroscience reports. 2019 Jan;19(1):1. Szcześniak D, Lenart-Bugla M, Misiak B, Zimny A, Sąsiadek M, Połtyn-Zaradna K, Zatońska K, Zatoński T, Szuba A, Smith EE, Yusuf S. Unraveling the protective effects of cognitive reserve on cognition and brain: a cross-sectional study. International Journal of Environmental Research and Public Health. 2022 Sep 27;19(19):12228. Nithianantharajah J, Hannan AJ. The neurobiology of brain and cognitive reserve: mental and physical activity as modulators of brain disorders. Progress in neurobiology. 2009 Dec 10;89(4):369 − 82. Groot C, van Loenhoud AC, Barkhof F, van Berckel BN, Koene T, Teunissen CC, Scheltens P, van der Flier WM, Ossenkoppele R. Differential effects of cognitive reserve and brain reserve on cognition in Alzheimer disease. Neurology. 2018 Jan 9;90(2):e149-56. Suo C, León I, Brodaty H, Trollor J, Wen W, Sachdev P, Valenzuela MJ. Supervisory experience at work is linked to low rate of hippocampal atrophy in late life. Neuroimage. 2012 Nov 15;63(3):1542-51. Kaup AR, Xia F, Launer LJ, Sidney S, Nasrallah I, Erus G, Allen N, Yaffe K. Occupational cognitive complexity in earlier adulthood is associated with brain structure and cognitive health in midlife: The CARDIA study. Neuropsychology. 2018 Nov;32(8):895. Nyberg L, Lövdén M, Riklund K, Lindenberger U, Bäckman L. Memory aging and brain maintenance. Trends in cognitive sciences. 2012 May 1;16(5):292–305. Boots EA, Schultz SA, Almeida RP, Oh JM, Koscik RL, Dowling MN, Gallagher CL, Carlsson CM, Rowley HA, Bendlin BB, Asthana S. Occupational complexity and cognitive reserve in a middle-aged cohort at risk for Alzheimer's disease. Archives of Clinical Neuropsychology. 2015 Nov 1;30(7):634 − 42. Lenhart L, Nagele M, Steiger R, Beliveau V, Skalla E, Zamarian L, Gizewski ER, Benke T, Delazer M, Scherfler C. Occupation-related effects on motor cortex thickness among older, cognitive healthy individuals. Brain Structure and Function. 2021 May;226(4):1023-30. Blix E, Perski A, Berglund H, Savic I. Long-term occupational stress is associated with regional reductions in brain tissue volumes. PloS one. 2013 Jun 11;8(6):e64065. Savic I. Structural changes of the brain in relation to occupational stress. Cerebral Cortex. 2015 Jun 1;25(6):1554-64. Petersen RC, Morris JC. Mild cognitive impairment as a clinical entity and treatment target. Archives of neurology. 2005 Jul 1;62(7):1160-3. 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. Science bulletin. 2022 Nov 30;67(22):2285-90. Dekhtyar S, Marseglia A, Xu W, Darin-Mattsson A, Wang HX, Fratiglioni L. Genetic risk of dementia mitigated by cognitive reserve: a cohort study. Annals of neurology. 2019 Jul;86(1):68–78. Kuh D, Karunananthan S, Bergman H, Cooper R. A life-course approach to healthy ageing: maintaining physical capability. Proceedings of the Nutrition Society. 2014 May;73(2):237 − 48. Boyle R, Knight SP, De Looze C, Carey D, Scarlett S, Stern Y, Robertson IH, Kenny RA, Whelan R. Verbal intelligence is a more robust cross-sectional measure of cognitive reserve than level of education in healthy older adults. Alzheimer's research & therapy. 2021 Jul 12;13(1):128. Chen Y, Lv C, Li X, Zhang J, Chen K, Liu Z, Li H, Fan J, Qin T, Luo L, Zhang Z. The positive impacts of early-life education on cognition, leisure activity, and brain structure in healthy aging. Aging (Albany NY). 2019 Jul 17;11(14):4923. Reed BR, Mungas D, Farias ST, Harvey D, Beckett L, Widaman K, Hinton L, DeCarli C. Measuring cognitive reserve based on the decomposition of episodic memory variance. Brain. 2010 Aug 1;133(8):2196 − 209. Stern Y, Arenaza-Urquijo EM, Bartrés‐Faz D, Belleville S, Cantilon M, Chetelat G, Ewers M, Franzmeier N, Kempermann G, Kremen WS, Okonkwo O. Whitepaper: Defining and investigating cognitive reserve, brain reserve, and brain maintenance. Alzheimer's & Dementia. 2020 Sep;16(9):1305-11. Zahodne LB, Manly JJ, Brickman AM, Narkhede A, Griffith EY, Guzman VA, Schupf N, Stern Y. Is residual memory variance a valid method for quantifying cognitive reserve? A longitudinal application. Neuropsychologia. 2015 Oct 1;77:260-6. van Loenhoud AC, Wink AM, Groot C, Verfaillie SC, Twisk J, Barkhof F, van Berckel B, Scheltens P, van der Flier WM, Ossenkoppele R. A neuroimaging approach to capture cognitive reserve: application to Alzheimer's disease. Human brain mapping. 2017 Sep;38(9):4703-15. Nelson ME, Veal BM, Andel R, Martinkova J, Veverova K, Horakova H, Nedelska Z, Laczó J, Vyhnalek M, Hort J. Moderating effect of cognitive reserve on brain integrity and cognitive performance. Frontiers in aging neuroscience. 2022 Nov 3;14:1018071. Peitz K, Bittner N, Heim S, Caspers S. Bilingualism and “brain reserve” in subregions of the hippocampal formation. GeroScience. 2025 Jun;47(3):4935-54. Raine PJ, Rao H. Volume, density, and thickness brain abnormalities in mild cognitive impairment: an ALE meta-analysis controlling for age and education. Brain imaging and behavior. 2022 Oct;16(5):2335-52. Liu Y, Julkunen V, Paajanen T, Westman E, Wahlund LO, Aitken A, Sobow T, Mecocci P, Tsolaki M, Vellas B, Muehlboeck S. Education increases reserve against Alzheimer’s disease—evidence from structural MRI analysis. Neuroradiology. 2012 Sep;54(9):929 − 38. Wisch JK, Petersen K, Millar PR, Abdelmoity O, Babulal GM, Meeker KL, Braskie MN, Yaffe K, Toga AW, O'Bryant S, Ances BM. Cross-Sectional Comparison of Structural MRI Markers of Impairment in a Diverse Cohort of Older Adults. Human Brain Mapping. 2025 Feb 1;46(2):e70133. Schwarz J, Zistler F, Usheva A, Fix A, Zinn S, Zimmermann J, Knolle F, Schneider G, Nuttall R. Investigating dynamic brain functional redundancy as a mechanism of cognitive reserve. Frontiers in Aging Neuroscience. 2025 Feb 4;17:1535657. Turcotte V, Potvin O, Dadar M, Hudon C, Duchesne S, Alzheimer’s Disease Neuroimaging Initiative. Birth cohorts and cognitive reserve influence cognitive performances in older adults. Journal of Alzheimer’s Disease. 2022 Jan 18;85(2):587–604. Cocchi L, Halford GS, Zalesky A, Harding IH, Ramm BJ, Cutmore T, Shum DH, Mattingley JB. Complexity in relational processing predicts changes in functional brain network dynamics. Cerebral Cortex. 2014 Sep 1;24(9):2283-96. Owen LL, Manning JR. High-level cognition is supported by information-rich but compressible brain activity patterns. Proceedings of the National Academy of Sciences. 2024 Aug 27;121(35):e2400082121. Pallares Di Nunzio M, Martín Tenti J, Arlego M, Rosso OA, Montani F. Exploring the role of synaptic plasticity in the frequency-dependent complexity domain. Chaos: An Interdisciplinary Journal of Nonlinear Science. 2025 Feb 1;35(2). Soh Y, Eng CW, Mayeda ER, Whitmer RA, Lee C, Peterson RL, Mungas DM, Glymour MM, Gilsanz P. Association of primary lifetime occupational cognitive complexity and cognitive decline in a diverse cohort: Results from the KHANDLE study. Alzheimer's & Dementia. 2023 Sep;19(9):3926-35. Coleman ME, Roessler ME, Peng S, Roth AR, Risacher SL, Saykin AJ, Apostolova LG, Perry BL. Social enrichment on the job: Complex work with people improves episodic memory, promotes brain reserve, and reduces the risk of dementia. Alzheimer's & Dementia. 2023 Jun;19(6):2655-65. Zhong T, Li S, Liu P, Wang Y, Chen L. The impact of education and occupation on cognitive impairment: a cross-sectional study in China. Frontiers in Aging Neuroscience. 2024 Jul 11;16:1435626. Vélez-Coto M, Andel R, Pérez-García M, Caracuel A. Complexity of work with people: Associations with cognitive functioning and change after retirement. Psychology and Aging. 2021 Mar;36(2):143. Nappo N. Job stress and interpersonal relationships cross country evidence from the EU15: A correlation analysis. BMC Public Health. 2020 Jul 20;20(1):1143. Suh C, Punnett L. High emotional demands at work and poor mental health in client-facing workers. International Journal of Environmental Research and Public Health. 2022 Jun 20;19(12):7530. Fantozzi IC, Di Luozzo S, Schiraldi MM. On tasks and soft skills in operations and supply chain management: analysis and evidence from the O* NET database. The TQM Journal. 2024 Dec 16;36(9):53–74. Handel MJ. The O* NET content model: strengths and limitations. Journal for Labour Market Research. 2016 Oct;49(2):157 − 76. Hosoda C, Tsujimoto S, Tatekawa M, Honda M, Osu R, Hanakawa T. Plastic frontal pole cortex structure related to individual persistence for goal achievement. Communications Biology. 2020 Apr 28;3(1):194. Reeve J, Lee W. Autonomy recruits neural support for interest and learning. Motivation and Emotion. 2025 Apr 11:1–5. Liu X, He T, Yu S, Duan J, Gao R. The effects of emotional labor on work strain and nonwork strain among dancers: A Person-centered approach. Psychology research and behavior management. 2023 Dec 31:3675-85. Xiong W, Huang M, Okumus B, Leung XY, Cai X, Fan F. How emotional labor affect hotel employees’ mental health: A longitudinal study. Tourism Management. 2023 Feb 1;94:104631. Liu C, Tu YH, Lin LJ, Chen H, Liu TH, Lin HL, Liu R, Chiou WK. Doctor-Patient communication models, patient decision-making participation, and patient emotional expression: a cross-cultural comparison of samples from the UK and China. Patient preference and adherence. 2025 Dec 31:2505-24. Huwaë S, Schaafsma J. Cross-cultural differences in emotion suppression in everyday interactions. International Journal of Psychology. 2018 Jun;53(3):176 − 83. Zheng S, Masuda T, Matsunaga M, Noguchi Y, Ohtsubo Y, Yamasue H, Ishii K. Cultural differences in social support seeking: The mediating role of empathic concern. Plos one. 2021 Dec 30;16(12):e0262001. Miller JG, Goyal N, Wice M. A cultural psychology of agency: Morality, motivation, and reciprocity. Perspectives on Psychological Science. 2017 Sep;12(5):867 − 75. Gollo LL, Roberts JA, Cropley VL, Di Biase MA, Pantelis C, Zalesky A, Breakspear M. Fragility and volatility of structural hubs in the human connectome. Nature neuroscience. 2018 Aug;21(8):1107-16. Metzler-Baddeley C, Caeyenberghs K, Foley S, Jones DK. Task complexity and location specific changes of cortical thickness in executive and salience networks after working memory training. Neuroimage. 2016 Apr 15;130:48–62. Jiao Y, Burzynska AZ, Fisher GG, Bielak AA. Occupational experiences and brain health outcomes in older age. O’Shea A, Cohen RA, Porges EC, Nissim NR, Woods AJ. Cognitive aging and the hippocampus in older adults. Frontiers in aging neuroscience. 2016 Dec 8;8:298. Yang W, Wang J, Guo J, Dove A, Qi X, Bennett DA, Xu W. Association of cognitive reserve indicator with cognitive decline and structural brain differences in middle and older age: findings from the UK Biobank. The Journal of Prevention of Alzheimer's Disease. 2024 May 1;11(3):739 − 48. Elshiekh A, Subramaniapillai S, Rajagopal S, Pasvanis S, Ankudowich E, Rajah MN. The association between cognitive reserve and performance-related brain activity during episodic encoding and retrieval across the adult lifespan. Cortex. 2020 Aug 1;129:296–313. Lichenstein SD, Verstynen T, Forbes EE. Adolescent brain development and depression: a case for the importance of connectivity of the anterior cingulate cortex. Neuroscience & Biobehavioral Reviews. 2016 Nov 1;70:271 − 87. Yu J, Kua EH, Mahendran R, Ng TK. ChatGPT-estimated occupational complexity predicts cognitive outcomes and cortical thickness above and beyond socioeconomic status among older adults. GeroScience. 2025 Aug;47(4):5709-23. Fisher GG, Stachowski A, Infurna FJ, Faul JD, Grosch J, Tetrick LE. Mental work demands, retirement, and longitudinal trajectories of cognitive functioning. Journal of occupational health psychology. 2014 Apr;19(2):231. Xue B, Cadar D, Fleischmann M, Stansfeld S, Carr E, Kivimäki M, McMunn A, Head J. Effect of retirement on cognitive function: the Whitehall II cohort study. European journal of epidemiology. 2018 Oct;33(10):989–1001. Calatayud E, Oliván-Blázquez B, Aguilar-Latorre A, Cuenca-Zaldivar JN, Magallón-Botaya RM, Gómez-Soria I. Analysis of the effectiveness of a computerized cognitive stimulation program designed from Occupational Therapy according to the level of cognitive reserve in older adults in Primary Care: Stratified randomized clinical trial protocol. Experimental Gerontology. 2024 Oct 15;196:112568. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9143855","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":612228716,"identity":"b2c4c5b2-c458-4bd3-ab4c-f0829c1a965b","order_by":0,"name":"Lingruina Xu","email":"","orcid":"","institution":"Beijing Normal University","correspondingAuthor":false,"prefix":"","firstName":"Lingruina","middleName":"","lastName":"Xu","suffix":""},{"id":612228717,"identity":"77ed0dca-5acf-415c-88bd-1b5a9f2a56fc","order_by":1,"name":"Lei Yu","email":"","orcid":"","institution":"Beijing Normal University","correspondingAuthor":false,"prefix":"","firstName":"Lei","middleName":"","lastName":"Yu","suffix":""},{"id":612228718,"identity":"2b9aebbf-31d0-4579-af51-61710ba49b68","order_by":2,"name":"Jiawen Liu","email":"","orcid":"","institution":"Beijing Normal University","correspondingAuthor":false,"prefix":"","firstName":"Jiawen","middleName":"","lastName":"Liu","suffix":""},{"id":612228719,"identity":"1462c3cd-af01-4d81-9716-26e094c7845b","order_by":3,"name":"Ting Li","email":"","orcid":"","institution":"Beijing Normal University","correspondingAuthor":false,"prefix":"","firstName":"Ting","middleName":"","lastName":"Li","suffix":""},{"id":612228720,"identity":"15f29355-4305-4278-972c-9c73ea55433f","order_by":4,"name":"Lixia Yang","email":"","orcid":"","institution":"Toronto Metropolitan University","correspondingAuthor":false,"prefix":"","firstName":"Lixia","middleName":"","lastName":"Yang","suffix":""},{"id":612228721,"identity":"1b0ec6a3-b65d-42d4-8037-88359038ea34","order_by":5,"name":"Kewei Chen","email":"","orcid":"","institution":"Banner Alzheimer’s Institute","correspondingAuthor":false,"prefix":"","firstName":"Kewei","middleName":"","lastName":"Chen","suffix":""},{"id":612228722,"identity":"c674592f-751b-4554-a5b0-5b1662b01f71","order_by":6,"name":"Jun Wang","email":"","orcid":"","institution":"Beijing Normal University","correspondingAuthor":false,"prefix":"","firstName":"Jun","middleName":"","lastName":"Wang","suffix":""},{"id":612228725,"identity":"1f58b0fb-4365-47ff-bc02-44a2b7ff8bab","order_by":7,"name":"Xin Li","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyUlEQVRIiWNgGAWjYBACPmYGhgMMFTZyBmAuGxFa2MBazqQZk6AFRDC2HU7cQLwWdh7Dw4Vth9O3S/cYMHwoO8zAP7uBkMPYEg7POJeeu3POGQPGGecOM0jcOUBIC/OBwzxl1rkbbuQYMPO2HWYwkEggpIWx4TAPG3O6AUjLX+K0gGxpc04Aa2EkTgvQLzxn0gx3zkgrONhzLp1H4gYBLfz8Z4w/81TYyJtLJG988KPMWo5/BgEtKOAAEPOQoH4UjIJRMApGAS4AAGwzPVpkOTysAAAAAElFTkSuQmCC","orcid":"","institution":"Beijing Normal University","correspondingAuthor":true,"prefix":"","firstName":"Xin","middleName":"","lastName":"Li","suffix":""},{"id":612228726,"identity":"7a1d0d84-07ca-407a-93be-f09acea61369","order_by":8,"name":"Zhanjun Zhang","email":"","orcid":"","institution":"Beijing Normal University","correspondingAuthor":false,"prefix":"","firstName":"Zhanjun","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2026-03-17 04:23:32","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9143855/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9143855/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105548232,"identity":"98de1fb9-41b0-48d7-b5af-9485171efccf","added_by":"auto","created_at":"2026-03-27 09:28:35","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":507905,"visible":true,"origin":"","legend":"\u003cp\u003eParticipant flow chart. BABRI: Beijing Aging Brain Rejuvenation Initiative\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-9143855/v1/fb6e3be5dd2651142e7f70fd.png"},{"id":105548231,"identity":"fd16ea5e-cfbc-479d-99bd-2953988df8a1","added_by":"auto","created_at":"2026-03-27 09:28:34","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":516164,"visible":true,"origin":"","legend":"\u003cp\u003eModerating effects of retirement duration on the associations between occupational complexity dimensions and visuospatial processing\u003c/p\u003e\n\u003cp\u003eNote. (a) Johnson–Neyman analysis shows the conditional effect of data complexity across the full range of retirement duration. The y-axis represents the regression coefficient (β) for data complexity. (b) Simple slopes depict the effect of data complexity at low (−1 SD), medium (mean), and high (+1 SD) levels of centered retirement duration. (c) Johnson–Neyman analysis shows the conditional effect of people complexity across the full range of retirement duration. The y-axis represents the regression coefficient (β) for people complexity. (d) Simple slopes depict the effect of people complexity at low (−1 SD), medium (mean), and high (+1 SD) levels of centered retirement duration. \u0026nbsp;*\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05; **\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01; ***\u003cem\u003ep\u003c/em\u003e\u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-9143855/v1/bc65a646fe48e90aa62ef293.png"},{"id":105548212,"identity":"8219f95f-dfcc-4f7a-aec6-546fd258cd82","added_by":"auto","created_at":"2026-03-27 09:28:26","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":546421,"visible":true,"origin":"","legend":"\u003cp\u003eModerated Mediation Models of Occupational Complexity on Cognitive Outcomes\u003c/p\u003e\n\u003cp\u003eNote. Path coefficients are unstandardized. These models tested whether intelligent activity mediated the effects of occupational complexity on cognitive performance, and whether this mediation was moderated by retirement duration. Subplots show different complexity–outcome pairs: (a) data complexity → visuospatial processing, (b) data complexity → attention, and (c) people complexity → visuospatial processing. All models were adjusted for sex, age, education, and cardiometabolic conditions. *\u003cem\u003ep\u003c/em\u003e\u0026lt; 0.05; ***\u003cem\u003ep\u003c/em\u003e\u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-9143855/v1/7f4926b602ae4381a9785340.png"},{"id":105548208,"identity":"a41ca2e0-2cd6-4a75-87f6-2813680a6d50","added_by":"auto","created_at":"2026-03-27 09:28:26","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":350508,"visible":true,"origin":"","legend":"\u003cp\u003eStructural correlates of occupational complexity across the cortex.\u003c/p\u003e\n\u003cp\u003eNote. Colors correspond to occupational complexity dimensions: green = data, yellow = things, blue = people.R-ITG, right inferior temporal gyrus; L-MFG, left middle frontal gyrus; R-VL Thal, right ventrolateral thalamic nucleus; L-ACC, left anterior cingulate cortex.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-9143855/v1/1b88e2160c2c9ef77b496219.png"},{"id":105548227,"identity":"ef7f2284-bd1c-422b-9c54-1d375997a360","added_by":"auto","created_at":"2026-03-27 09:28:28","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":402313,"visible":true,"origin":"","legend":"\u003cp\u003eDistinct neurostructural and functional mediation pathways linking occupational data complexity to cognitive preservation.\u003c/p\u003e\n\u003cp\u003eNote. (a) Schematic representation of the dual-pathway mediation model. The model illustrates how occupational data complexity contributes to cognitive function through two distinct mechanisms: a structural BR pathway (top panel) and a functional CR pathway (bottom panel). The BR pathway is indexed by the structural integrity of specific brain regions, including the left hippocampus and right inferior temporal gyrus. The CR pathway is quantified with a residual-based latent score reflecting functional adaptability.\u003c/p\u003e\n\u003cp\u003e(b) Forest plot displaying the standardized indirect effects (β) of occupational data complexity on eight cognitive domains. Estimates and 95% confidence intervals are shown for three specific pathways, distinguished by color and shape: Light blue circles: Indirect effects via the CR pathway (residual-based score), representing functional reserve. Royal blue squares: Indirect effects via the BR pathway mediated specifically by the left hippocampus. Navy blue triangles: Indirect effects via the BR pathway mediated specifically by the right inferior temporal gyrus. Error bars indicate 95% bias-corrected bootstrap confidence intervals. Intervals that do not cross the vertical dashed line (\u003cem\u003ex\u003c/em\u003e=0) are statistically significant (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05). The right-hand column lists the precise standardized point estimates and [95% CI] for each path. Note that while CR mediates effects across most domains, structural BR mediation is domain-specific, and a negative indirect effect is observed for executive function.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-9143855/v1/2fba29fa7c93f68698326cd0.png"},{"id":109489828,"identity":"480f29cf-807c-4095-8152-0da134b218b4","added_by":"auto","created_at":"2026-05-18 17:24:52","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2531619,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9143855/v1/cd9ce615-1b6f-43cb-8b70-9c08da5b85bc.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Distinct Effects of Data Occupational Complexity On Cognitive Aging: Evidence for Dual Brain–Cognitive Reserve Pathways","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eThe continuously accelerated aging of global population has made healthy cognitive aging a major public health challenge. Cognitive impairment is one of the most devastating health problems associated with aging and poses a particularly severe burden in China, where the absolute number of individuals living with dementia is the highest worldwide, exceeding 15\u0026nbsp;million and accounting for nearly one quarter of the global dementia population\u003csup\u003e[1]\u003c/sup\u003e. It becomes a crucial topic in healthy aging research to identify modifiable protective factors for cognitive aging. In recent years, occupation has attracted a growing attention as a potential protective factor, on its own or as a result of education, for healthy cognitive aging\u003csup\u003e[2, 3]\u003c/sup\u003e. Because work takes a large proportion of time across lifespan, particularly in adulthood, occupational complexity may play a predominant role in cognitive functions, including cognitive trajectories well after retirement.\u003c/p\u003e \u003cp\u003eTheoretically, the \u0026ldquo;use it or lose it\u0026rdquo; hypothesis emphasized the promoting effect of continuous cognitive engagement on neuroplasticity\u003csup\u003e[4]\u003c/sup\u003e. Similarly, the theory of environmental complexity highlighted the cumulative effects of cognitive reserve (CR) as assessed by the characteristics of professional tasks (such as decision variations, conflicting information resolution)\u003csup\u003e[5]\u003c/sup\u003e. Most researchers use metrics of occupational complexity to examine the impact of occupations on cognitive abilities in older adults. Occupational complexity (OC), as a multi-dimensional concept to quantify cognitive stimulation, has advantages in two aspects: (1) it quantifies the complexity through objective characteristics of tasks; (2) it systematically covers a three-dimension concept of \u0026ldquo;data-people-things\u0026rdquo;\u003csup\u003e[6]\u003c/sup\u003e, thus enabling the analysis of the effects of each dimension independently and their interactions on cognition. Briefly, data complexity refers to cognitive demands related to information processing, reasoning, and decision-making; people complexity captures the extent of interpersonal interaction, communication, and coordination at work; whereas things complexity reflects demands associated with handling tools, machines, or physical objects.\u003c/p\u003e \u003cp\u003eMost longitudinal studies have shown that high people complexity (e.g., management or education) is associated with a lower risk of dementia and better cognitive function, and the protective effect remains significant after controlling for education\u003csup\u003e[7, 8]\u003c/sup\u003e. The role of data complexity remains controversial: Some studies show that it can enhance processing speed and memory\u003csup\u003e[9]\u003c/sup\u003e, while others suggest that prolonged high-intensity work may actually increase the risk of Alzheimer's disease due to chronic stress\u003csup\u003e[10]\u003c/sup\u003e. The influence of things complexity is generally weak and unstable, which may depend on the task novelty rather than the complexity itself\u003csup\u003e[11]\u003c/sup\u003e. Furthermore, the cognitive effect of occupational complexity on cognitive function is moderated by the retirement transition and leisure activities in old age, suggesting that the effect is context-dependency\u003csup\u003e[12, 13]\u003c/sup\u003e. Overall, high occupational complexity is generally associated with better cognitive ability in old age, with significant heterogeneity across dimensions.\u003c/p\u003e \u003cp\u003eNeuroimaging studies provide crucial evidence supporting the mechanisms of brain reserve (BR) and CR in relation to occupational complexity. BR (i.e., a neuroanatomic resource reflecting structural properties that afford surplus capacity to tolerate neuropathology) and CR (i.e., the acquired skills and flexible processing strategies that allow more efficient use of remaining neural resources) together can explain why individuals with similar brain pathology often show different clinical outcomes\u003csup\u003e[14]\u003c/sup\u003e. Research on BR and CR clarifies interindividual variability in cognitive aging trajectory, informs prevention strategies for dementia, and guides interventions to promote resilient cognitive aging\u003csup\u003e[15, 16]\u003c/sup\u003e. Experimental and observational studies further indicate that life-course activities \u0026mdash; including mentally stimulating work \u0026mdash; modulate both reserves via neuroplastic mechanisms\u003csup\u003e[17]\u003c/sup\u003e. Furthermore, the relative contributions of BR and CR may differ across pathologies and cognitive domains\u003csup\u003e[18]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eConsistent with with reserve-based theories of cognitive aging rooted in environmental complexity, evidence suggests that occupational complexity contributes to brain reserve accumulation across the life course. Longitudinal and cross-sectional studies in older adults demonstrate that supervisory and managerial experience in midlife is associated with larger hippocampal and medial temporal gray matter volumes and substantially slower rates of late-life hippocampal atrophy, independent of demographic, genetic, and late-life lifestyle factors\u003csup\u003e[19]\u003c/sup\u003e. Extending these findings to earlier stages of adulthood, occupational cognitive complexity during early to midadulthood has also been linked to higher white matter integrity and superior executive functioning in midlife, indicating that work-related cognitive demands may shape neural substrates of reserve decades before old age\u003csup\u003e[20]\u003c/sup\u003e. These findings suggest that complex professional experiences promote BR accumulation, in light of the \"brain maintenance\" theory\u003csup\u003e[21]\u003c/sup\u003e. Furthermore, even when experiencing brain atrophy, individuals with high occupational complexity can maintain relatively preserved cognitive function\u003csup\u003e[22]\u003c/sup\u003e, through CR by which the brain responds to aging and pathological changes by more efficient and flexible resource utilization. Moreover, different occupational complexity dimensions showed distinct effects on different brain regions. For example, high physical demands correlate with cortical thickness in pre- and postcentral gyri\u003csup\u003e[23]\u003c/sup\u003e, whereas high-pressure working environment may be associated with the degeneration of the prefrontal cortex and the limbic structures\u003csup\u003e[24, 25]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eHowever, there remained several key issues understudied. First, it is unclear about the specific mechanisms through which different dimensions of occupational complexity affect cognitive function. In particular, the relative contributions of retirement timing and post-retirement lifestyle require further clarifications. Second, it remains unclear whether the effects might be mediated by BR, CR, or both. Finally, further research is needed to elucidate how occupational complexity contributes to post-retirement cognitive engagement and brain structural maintenance, and whether late-life activities reinforce or compensate for the reserve mechanisms established during one\u0026rsquo;s occupational years. Therefore, this study used a multi-modal data analysis approach to systematically examine the differential effects of different occupational complexity dimensions on cognitive function in old age. We aim to (1) analyze the heterogeneous effects of data, people, and things complexity on specific cognitive domains; (2) reveal the association patterns between different dimensions of occupational complexity and brain structural changes; and (3) construct a mediation model of \u0026ldquo;occupational complexity\u0026ndash;BR/CR\u0026ndash;cognitive function\u0026rdquo; to elucidate the associated cognitive protection mechanism. This study will provide a theoretical and empirical basis for developing cognitive intervention strategies based on occupational characteristics\u003c/p\u003e"},{"header":"2. Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Participants\u003c/h2\u003e \u003cp\u003e A total of 3,754 community-dwelling retirees aged 50 years and older participated in the Beijing Aging Brain Rejuvenation Initiative (BABRI), an ongoing community-based prospective cohort study in China focusing on higher-order cognitive aging and its neural mechanisms (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).The study protocol was approved by the Institutional Review Boards of the Beijing Normal University Imaging Center for Brain Research and Beijing Tiantan Hospital and was conducted in accordance with the Declaration of Helsinki. All participants provided informed consent in written format.\u003c/p\u003e \u003cp\u003eParticipants were screened with the following inclusion criteria: (1) retired individuals (\u0026ge;\u0026thinsp;50 years for women, \u0026ge; 55 years for men); (2) native Chinese speakers; (3) right-handed; and (4) with normal or corrected-to-normal vision and hearing, as well as the following exclusion criteria: (1) a history of major neurological disorders (e.g., stroke, intracranial lesions); (2) a current clinical diagnosis or a history of neuropsychiatric conditions (e.g., depression, schizophrenia, or epilepsy); and (3) a history of substance abuse (e.g., alcohol dependence or sedative abuse).\u003c/p\u003e \u003cp\u003eAmong these participants, a sub-sample of 851 MRI eligible participants voluntarily underwent MRI scanning. All participants met the MRI safety requirements, with no metallic implants (e.g., dental prostheses, cardiac stents, orthopedic plates or screws), and had no history of claustrophobia, M\u0026eacute;ni\u0026egrave;re\u0026rsquo;s disease, or other ear-related disorders that might interfere with MRI procedures. All participants provided written informed consent prior to scanning. No participant withdrew during image acquisition, and all scans were successfully completed with high image quality and minimal head motion artifacts. Participants completed T1-weighted structural MRI and an extensive cognitive battery within a month.\u003c/p\u003e \u003cp\u003eMild cognitive impairment (MCI) was diagnosed according to Petersen\u0026rsquo;s criteria\u003csup\u003e[26]\u003c/sup\u003e. Specifically, participants were classified as having MCI if they met the following conditions: (1) the presence of subjective cognitive complaints reported by the participant and/or an informant; (2) preserved general cognitive functioning, defined as a Mini-Mental State Examination (MMSE) score greater than 23; (3) intact activities of daily living, indicated by a score of 0 (no functional impairment) on both the Activities of Daily Living (ADL) and Instrumental Activities of Daily Living (IADL) scales; and (4) objective cognitive impairment in at least one cognitive domain, operationalized as performance 1.5 standard deviations below the normative mean on at least one neuropsychological test\u003csup\u003e[27]\u003c/sup\u003e. Participants who did not meet the diagnostic criteria for MCI were classified as cognitively normal, with neuropsychological test scores within 1.5 standard deviations.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Measures\u003c/h2\u003e \u003cdiv id=\"Sec5\" class=\"Section3\"\u003e \u003ch2\u003e2.2.1 Occupational Complexity\u003c/h2\u003e \u003cp\u003eOccupational complexity was evaluated based on the matrix of U.S. occupations developed for the 1970 U.S. Census\u003csup\u003e[28, 29]\u003c/sup\u003e and standardized using the O*NET database to ensure cross-cultural comparability. Each participant\u0026rsquo;s primary occupation\u0026mdash;defined as the occupation held for at least five years\u0026mdash;was rated along three dimensions: people complexity(0\u0026ndash;8), data complexity(0\u0026ndash;6), and things complexity(0\u0026ndash;7). In the original coding scheme, higher scores indicated lower occupational demands. For ease of interpretation, we reversed the scores so that higher values reflected greater occupational demands. The three dimensions were then summed to yield a total occupational complexity score ranging from 0 to 21, with higher scores indicating greater overall complexity\u003csup\u003e[30]\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section3\"\u003e \u003ch2\u003e2.2.2 Retirement Duration and Leisure Activities after Retirement\u003c/h2\u003e \u003cp\u003eAs the original dataset did not directly record the retirement duration, it was operationalized through retrospective calculation:\u003c/p\u003e \u003cp\u003eRetirement Duration\u0026thinsp;=\u0026thinsp;Baseline Survey Year\u0026thinsp;\u0026minus;\u0026thinsp;Official Retirement Year.\u003c/p\u003e \u003cp\u003eLeisure activities after retirement were assessed using a personal information questionnaire consisting of 23 items covering cognitive, social, and physical leisure activities. Participants were asked to recall their engagement in leisure activities during the past year, including reading, writing, attending senior university courses, playing chess, poker, or mahjong, and engaging in handcrafts. The frequency of each activity was defined as either \u0026ldquo;frequent\u0026rdquo; if being engaged several times per week or \u0026ldquo;rare\u0026rdquo; if being engaged for less than once per week. A weighted composite score across all 23 items was calculated, with higher scores indicating greater engagement in leisure activities in later life\u003csup\u003e[31]\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e2.2.3 Neuropsychological Test and Personal Information Questionnaire\u003c/h2\u003e \u003cp\u003eAll participants completed a comprehensive neuropsychological battery to assess eight cognitive domains, including: (1) general cognition, assessed by the Mini-Mental State Examination (MMSE) total score; (2) episodic memory, assessed by the Auditory Verbal Learning Test (AVLT)\u0026mdash;covering immediate recall (N1\u0026ndash;N3), short-delay recall (N4), long-delay recall (N5), and total recall (N1\u0026ndash;N5), and the Rey\u0026ndash;Osterrieth Complex Figure Test (ROCF) delayed recall; (3) working memory, measured by the Digit Span Test (DST)\u0026mdash;backward; (4) reasoning, assessed by the Similarities Test (SIM); (5) visuospatial processing, assessed by ROCF copy and the Clock Drawing Test (CDT); (6) language, evaluated by the Category Verbal Fluency Test (CVFT)\u0026mdash;animal, vegetable, and fruit categories, and the Boston Naming Test (BNT); (7) attention, assessed by the Symbol Digit Modalities Test (SDMT) and the Trail Making Test A (TMT-A); and (8) executive function, assessed by the Trail Making Test B (TMT-B) and the Stroop Color-Word Test C (SCWT-C).\u003c/p\u003e \u003cp\u003eDemographic and health information were collected using a standardized personal information questionnaire, including age, sex, and years of education as demographic variables, and chronic conditions such as hypertension, hyperlipidemia, diabetes, coronary heart disease, and cerebral small vessel disease. The number of cardiometabolic conditions, together with age and sex, were used as covariates in subsequent statistical analyses.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Statistic Analysis\u003c/h2\u003e \u003cp\u003eAnalyses were conducted for both behavioral and neuroimaging data. All statistical analyses were performed using SPSS version 22.0 and R version 4.0. MRI morphological analyses were implemented using the Computational Anatomy Toolbox (CAT12; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.neuro.uni-jena.de/cat/\u003c/span\u003e\u003cspan address=\"http://www.neuro.uni-jena.de/cat/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) within the Statistical Parametric Mapping software (SPM12; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.fil.ion.ucl.ac.uk/spm/\u003c/span\u003e\u003cspan address=\"https://www.fil.ion.ucl.ac.uk/spm/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) running in the MATLAB (R2020a) environment. To improve interpretability of effect estimates, key continuous predictors were Z-standardized prior to analysis. Statistical significance was set at two-tailed \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, and corrections for multiple comparisons were applied using either the False Discovery Rate (FDR) or Family-Wise Error (FWE) method, as appropriate.\u003c/p\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003e2.3.1 Behavioral Data\u003c/h2\u003e \u003cp\u003eOutliers (|Z| \u0026gt; 3) were removed, and missing values were imputed using a linear regression method with 20 iterative estimations to ensure data completeness and robustness. Cognitive performance was evaluated across eight domains (general cognition, episodic memory, working memory, reasoning, visuospatial processing, language, attention, and executive function). Domain scores were Z-standardized and averaged to obtain composite indices; reverse-coded measures (e.g., TMT-A) were reciprocally transformed prior to standardization.\u003c/p\u003e \u003cp\u003eDescriptive statistics were used to summarize demographic and cognitive characteristics. Group differences across levels of occupational complexity (people, data, and things) were examined using independent-sample t tests and chi-square tests. To assess the independent effects of occupational complexity on cognitive domains, multivariable linear regression models were constructed, in which people, data, and things complexity were entered simultaneously, with sex, age, education years, retirement duration, and cardiometabolic conditions entered hierarchically as covariates.\u003c/p\u003e \u003cp\u003eFurther, within a path analysis framework, the PROCESS macro (Model 1 and Model 4) was applied to examine the mediating role of late-life leisure activities and the moderating effect of retirement duration. Indirect effects were estimated using bias-corrected bootstrap procedures (5,000 resamples), and significance regions were determined via the Johnson\u0026ndash;Neyman technique, distinguishing between \u0026ldquo;enhancement\u0026rdquo; and \u0026ldquo;compensation\u0026rdquo; patterns of association.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e \u003ch2\u003e2.3.2 Neuroimaging\u003c/h2\u003e \u003cp\u003eTo investigate the neural correlates of occupational complexity and to test hypothesized brain reserve pathways, primary neuroimaging analyses focused on whole-brain morphometric measures using voxel-based morphometry (VBM) and surface-based morphometry (SBM).\u003c/p\u003e \u003cp\u003eVBM analyses involved whole-brain voxel-wise regression models and two-sample t tests to examine associations between occupational complexity dimensions and local gray matter volume, controlling for sex, age, retirement duration, cardiometabolic conditions, and total intracranial volume (TIV). SBM analyses were conducted using cortical thickness measures derived from the Desikan\u0026ndash;Killiany atlas, with regression and group comparison models assessing associations between occupational complexity dimensions and regional cortical thickness. Covariates in SBM analyses were identical to those used in the VBM models, excluding TIV.\u003c/p\u003e \u003cp\u003e As supporting analyses, associations between occupational complexity dimensions and global brain structural indices were examined, and exploratory region-of-interest (ROI) analyses based on the Automated Anatomical Labeling (AAL) atlas were performed to provide complementary regional validation of whole-brain findings.\u003c/p\u003e \u003cp\u003eAll whole-brain analyses were corrected for multiple comparisons using the Family-Wise Error (FWE) or FDR method, as appropriate.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section3\"\u003e \u003ch2\u003e2.3.3 CR Quantification and Mediation Modeling\u003c/h2\u003e \u003cp\u003eCR was quantified using the residual-based approach, which operationalizes cognitive reserve as the portion of cognitive performance not explained by structural brain integrity, an approach that has been well established in prior work\u003csup\u003e[32\u0026ndash;34]\u003c/sup\u003e. Conceptually, this framework defines CR as better-than-expected cognitive functioning given measurable brain structure and relevant covariates.\u003c/p\u003e \u003cp\u003eFirst, overall cognitive performance (observed value, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{Y}_{cog}\\)\u003c/span\u003e\u003c/span\u003e)was treated as the dependent variable, and total gray matter volume (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{X}_{GMV}\\)\u003c/span\u003e\u003c/span\u003e)as the independent variable. GMV was selected because it represents a global macrostructural marker of brain integrity that has been explicitly incorporated into residual-based neuroimaging models of cognitive reserve, particularly in aging and neurodegenerative contexts\u003csup\u003e[35]\u003c/sup\u003e.A multiple linear regression model was constructed, controlling for sex(\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{X}_{sex}\\)\u003c/span\u003e\u003c/span\u003e), age(\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{X}_{age}\\)\u003c/span\u003e\u003c/span\u003e), retirement duration(\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{X}_{retire}\\)\u003c/span\u003e\u003c/span\u003e), cardiometabolic conditions(\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{X}_{disease}\\)\u003c/span\u003e\u003c/span\u003e), and total intracranial volume (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{X}_{TIV}\\)\u003c/span\u003e\u003c/span\u003e):\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:{Y}_{cog}={\\beta\\:}_{0}+{\\beta\\:}_{1}{X}_{GMV}+{\\beta\\:}_{2}{X}_{sex}+{\\beta\\:}_{3}{X}_{age}+{\\beta\\:}_{4}{X}_{retire}+{\\beta\\:}_{5}{X}_{disease}+{\\beta\\:}_{6}{X}_{TIV}+ϵ$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eThe predicted cognitive score was then computed as:\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$\\\\hat :{{Y}_{cog}}=\\hat {{\\beta\\:}}_{0}+\\hat {{\\beta\\:}}_{1}{X}_{GMV}+\\hat {{\\beta\\:}}_{2}{X}_{sex}+\\hat {{\\beta\\:}}_{3}{X}_{age\\:}+\\hat {{\\beta\\:}}_{4}{X}_{retire}+\\hat {{\\beta\\:}}_{5}{X}_{disease}+\\hat {{\\beta\\:}}_{6}{X}_{TIV}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eCR was defined as the residual difference between the observed and predicted cognitive scores:\u003cdiv id=\"Equc\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equc\" name=\"EquationSource\"\u003e\n$$\\:CR\\:=\\:{Y}_{cog}\\hat -{{Y}_{cog}}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eHigher CR values indicate better-than-expected cognitive performance given an individual\u0026rsquo;s level of brain structure. Differences in CR between occupational complexity groups were examined using t tests, and associations between continuous measures of occupational complexity and CR were assessed using partial correlations and multivariable regression models controlling for the same covariates.\u003c/p\u003e \u003cp\u003eTo evaluate whether CR functioned as a pathway linking occupational complexity to cognitive performance, mediation models were constructed using the PROCESS macro with 5,000 bootstrap resamples. These models estimated the indirect effect of occupational complexity on cognition through CR while adjusting for demographic and health variables. Similar mediation analyses were then conducted to examine BR as an alternative pathway, employing the same modeling strategy and covariates.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section3\"\u003e \u003ch2\u003e2.3.4 Model Robustness and Validation\u003c/h2\u003e \u003cp\u003eAll regression models were evaluated for residual normality and multicollinearity among predictors, including the three occupational complexity dimensions and covariates, with variance inflation factors (VIFs) below 5, indicating acceptable model fit. The findings remained consistent across different sampling strategies and covariate adjustments, supporting the robustness of the results.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\n \u003ch2\u003e3.1 Demographic and Cognitive Characteristics\u003c/h2\u003e\n \u003cp\u003eIn the total sample (N\u0026thinsp;=\u0026thinsp;3,754), participants in the high occupational complexity group were older, more often male, and had more years of education. They also engaged more frequently in cognitive, social, and physical activities, and exhibited lower levels of depressive symptoms, body mass index, and a lower prevalence of cardiometabolic conditions (all \u003cem\u003ep\u003c/em\u003e\u0026lt; .05).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\n \u003ch2\u003e3.2 Associations Between Occupational Complexity Dimensions and Cognitive Function\u003c/h2\u003e\n \u003cp\u003eAcross multiple cognitive domains, data complexity demonstrated a robust and consistent positive association with cognitive performance. Before controlling for education, data complexity predicted performance across all domains except for executive function (standardized \u0026beta;\u0026thinsp;=\u0026thinsp;0.057\u0026ndash;0.109, all \u003cem\u003ep\u003c/em\u003e\u0026lt; .001) and was associated with a 14.3% reduction in MCI risk (\u003cem\u003ep\u003c/em\u003e\u0026lt; .001). After controlling for education, associations with global cognition, episodic memory, working memory, reasoning, visuospatial processing, and attention remained significant (\u0026beta;\u0026thinsp;=\u0026thinsp;0.016\u0026ndash;0.058, all \u003cem\u003ep\u003c/em\u003e\u0026lt; .05), corresponding to a 5.7% reduction in MCI risk (\u003cem\u003ep\u003c/em\u003e\u0026lt; .05). The people complexity was only associated with reasoning (\u0026beta;\u0026thinsp;=\u0026thinsp;0.034, \u003cem\u003ep\u003c/em\u003e\u0026lt; .001) and attention (\u0026beta;\u0026thinsp;=\u0026thinsp;0.018, \u003cem\u003ep\u003c/em\u003e\u0026lt; .05), whereas the things complexity showed no significant effects.\u003c/p\u003e\n \u003cp\u003eModeration analyses indicated that retirement duration significantly moderated the association between occupational complexity and cognitive performance. Specifically, the positive effect of the data complexity on visuospatial processing was stronger with increasing retirement duration. (interaction \u0026beta;\u0026thinsp;=\u0026thinsp;0.002, \u003cem\u003ep\u003c/em\u003e = .029) and enhanced the predictive effect of the people complexity on visuospatial processing (interaction \u0026beta;\u0026thinsp;=\u0026thinsp;0.003, \u003cem\u003ep\u003c/em\u003e = .001), suggesting that the benefits of occupational complexity may accumulate over time.The moderating effects of retirement duration on the associations between occupational complexity dimensions and visuospatial processing are visually depicted in Fig. \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e\n \u003cp\u003eMediation analyses further suggested that late-life activities constituted key behavioral pathways linking occupational complexity to cognitive performance. Specifically, data complexity was indirectly associated with attention through engagement in intelligent activities. This indirect pathway showed conditional effects ranging from 0.0071 to 0.0080 across low, medium, and high retirement-duration strata and was consistently stronger than the corresponding pathways through social activities (0.0028\u0026ndash;0.0036) and physical activities (0.0019\u0026ndash;0.0024), indicating that intelligent activity represented the strongest and most stable behavioral mediator linking data complexity to attention..In contrast, people complexity demonstrated a weaker but significant indirect association with visuospatial processing through social activities, with effects varying across retirement duration (0.0001\u0026ndash;0.0018) .\u003c/p\u003e\n \u003cp\u003eModerated mediation models further revealed compensatory effects among individuals with low things complexity. Specifically, these participants showed greater cognitive gains in multiple domains, including episodic memory and attention, through more frequent intelligent (\u0026beta;\u0026thinsp;=\u0026thinsp;0.19\u0026ndash;0.24, \u003cem\u003ep\u003c/em\u003e\u0026lt; .01) and social activities (\u0026beta;\u0026thinsp;=\u0026thinsp;0.15\u0026ndash;0.18, \u003cem\u003ep\u003c/em\u003e\u0026lt; .05). Enhancing effects were observed only for reasoning (\u0026beta;\u0026thinsp;=\u0026thinsp;0.12, \u003cem\u003ep\u003c/em\u003e\u0026lt; .05) and executive function (\u0026beta;\u0026thinsp;=\u0026thinsp;0.09, \u003cem\u003ep\u003c/em\u003e\u0026lt; .05).The moderated mediation models illustrating the pathways from occupational complexity to cognitive outcomes through intelligent activity are presented in Fig. \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\n \u003ch2\u003e3.3 BR as a Pathway Linking Occupational Complexity to Cognition\u003c/h2\u003e\n \u003cp\u003eTo establish the structural basis for mediation analyses, we first examined the associations between occupational complexity and brain morphology in the MRI subsample. After adjusting for demographic and health covariates, regression analyses on global brain measures indicated that data complexity significantly predicted larger intracranial volume (\u0026beta;\u0026thinsp;=\u0026thinsp;4.93, \u003cem\u003ep\u003c/em\u003e \u0026lt; .05), whereas its associations with total gray matter and white matter volumes were not statistically significant. Region-specific analyses revealed robust structural correlates. After FDR correction, several associations remained significant (Fig. \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e), including greater cortical thickness in the right inferior temporal gyrus for data complexity (\u003cem\u003ep\u003c/em\u003e = .011), and significant associations of things complexity with the right ventrolateral thalamic nucleus and the left middle frontal gyrus (both FDR-corrected \u003cem\u003ep\u003c/em\u003e \u0026lt; .05). People complexity showed negative associations with cortical thickness in both the left anterior cingulate cortex (FDR-corrected \u003cem\u003ep\u003c/em\u003e \u0026lt; .05) and the left inferior orbital frontal gyrus (FDR-corrected \u003cem\u003ep\u003c/em\u003e = .015). These findings indicate that data-related occupational demands provide the strongest and most spatially extensive neural correlates relevant for subsequent brain-reserve analyses.\u003c/p\u003e\n \u003cp\u003eBuilding on these structural associations, we then tested whether occupational complexity influenced cognitive outcomes via BR, indexed by hippocampal gray matter and regional cortical thickness. Hippocampal volume was selected as it reflects episodic memory and global cognition, and larger volume has been linked to experiences that increase brain reserve\u003csup\u003e[36\u0026ndash;38]\u003c/sup\u003e. Cortical thickness was included as it captures local neuronal integrity, with greater thickness associated with higher reserve and better tolerance to cortical loss\u003csup\u003e[15, 39, 40]\u003c/sup\u003e. As shown in Fig. \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, mediation models revealed that data complexity exerted significant indirect effects on multiple cognitive domains through structural markers of BR. Specifically, higher data complexity predicted greater left hippocampal gray matter volume and increased cortical thickness in the right inferior temporal gyrus, which in turn mediated performance in general cognition (\u0026beta;\u0026thinsp;=\u0026thinsp;0.005\u0026ndash;0.006, \u003cem\u003ep\u003c/em\u003e \u0026lt; .05), episodic memory (\u0026beta;\u0026thinsp;=\u0026thinsp;0.004, \u003cem\u003ep\u003c/em\u003e \u0026lt; .05), language (\u0026beta;\u0026thinsp;=\u0026thinsp;0.004, \u003cem\u003ep\u003c/em\u003e \u0026lt; .05), and attention (\u0026beta;\u0026thinsp;=\u0026thinsp;0.004, \u003cem\u003ep\u003c/em\u003e \u0026lt; .05). A small negative indirect effect was observed for executive function (\u0026beta; = \u0026minus;\u0026thinsp;0.005, \u003cem\u003ep\u003c/em\u003e \u0026lt; .05), suggesting domain-specific differences. While data complexity\u0026ndash;related structural adaptations generally benefited memory, language, and attention, their influence on executive function was limited, consistent with prior findings that cognitive and brain reserve often show weaker effects on executive domains compared to memory or attention\u003csup\u003e[18, 41, 42]\u003c/sup\u003e.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\n \u003ch2\u003e3.4 CR as a Pathway Linking Occupational Complexity to Cognition\u003c/h2\u003e\n \u003cp\u003eCR was quantified using a residual-based modeling approach (see Methods) and examined as a potential mediator between occupational complexity and cognitive function. After adjusting for sex, age, years of education, and cardiometabolic conditions, data complexity remained an independent predictor (\u0026beta;\u0026thinsp;=\u0026thinsp;0.19, \u003cem\u003ep\u003c/em\u003e = .018).\u003c/p\u003e\n \u003cp\u003eMediation analyses demonstrated that CR played a central role in linking data complexity to cognitive performance. The detailed mediation effects of data complexity on cognitive domains through CR are summarized in Fig. \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e. CR significantly mediated the associations between data complexity and global cognition, reasoning, episodic memory, visuospatial processing, language, and executive function, with indirect effects ranging from \u0026beta;\u0026thinsp;=\u0026thinsp;0.01 to 0.03 for most domains, and all 95% bootstrap confidence intervals excluding zero. A negative indirect effect was observed for executive function. Partial mediation effects were observed for working memory and attention. The largest total effect was observed for attention (\u0026beta;\u0026thinsp;=\u0026thinsp;0.068, \u003cem\u003ep\u003c/em\u003e \u0026lt; .001).\u003c/p\u003e\n \u003cp\u003eExploratory, uncorrected analyses further suggested potential contributions of the right inferior temporal gyrus and left hippocampus; however, these effects did not survive FDR correction. Together, these findings indicate that CR serves as a key mechanism through which data complexity supports cognitive performance in later life.\u003c/p\u003e\n \u003cp\u003e(b) Forest plot displaying the standardized indirect effects (\u0026beta;) of occupational data complexity on eight cognitive domains. Estimates and 95% confidence intervals are shown for three specific pathways, distinguished by color and shape: Light blue circles: Indirect effects via the CR pathway (residual-based score), representing functional reserve. Royal blue squares: Indirect effects via the BR pathway mediated specifically by the left hippocampus. Navy blue triangles: Indirect effects via the BR pathway mediated specifically by the right inferior temporal gyrus. Error bars indicate 95% bias-corrected bootstrap confidence intervals. Intervals that do not cross the vertical dashed line (\u003cem\u003ex\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0) are statistically significant (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). The right-hand column lists the precise standardized point estimates and [95% CI] for each path. Note that while CR mediates effects across most domains, structural BR mediation is domain-specific, and a negative indirect effect is observed for executive function.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"4 Discussion","content":"\u003cp\u003eIn the present sample, occupational complexity demonstrated clear dimension-specific associations with later-life cognition and brain structure. Data complexity showed the most robust and generalized benefits, predicting higher global cognition, reasoning, and attention after adjustment for education, demographic covariates, and cardiometabolic conditions. It was also associated with a lower risk of MCI and exhibited the broadest indirect pathways through CR and structural markers. In contrast, people complexity showed selective associations with reasoning and attention but was negatively related to cortical thickness in the left anterior cingulate cortex (ACC-L), whereas things complexity displayed only modest effects largely restricted to sensorimotor regions. Late-life cognitive and social engagement partially mediated several of these associations, and retirement duration amplified specific pathways.\u003c/p\u003e \u003cp\u003eThe cognitive and neural protective effects of occupational complexity exhibited marked dimensional heterogeneity. Data complexity emerged as the most robust protective factor in this study. Even after controlling for education, it independently contributed to global cognition, reasoning, and attention, and lower risk of MCI, consistent with previous findings that occupations with high cognitive demands delay cognitive decline\u003csup\u003e[7, 9]\u003c/sup\u003e. Mechanistically, data complexity was associated with larger gray matter volumes and greater cortical thickness in frontotemporal\u0026ndash;limbic regions, including the hippocampus and inferior temporal gyrus, and was the only dimension showing a significant positive correlation with CR quantified via the residual-based approach.\u003c/p\u003e \u003cp\u003eOne plausible explanation for these advantages lies in the core cognitive features of data-intensive occupations, which emphasize relational processing, abstract reasoning, and the integration of multiple information streams. Tasks characterized by high data complexity require individuals to simultaneously coordinate multiple variables and rules, a form of relational complexity that has been shown to recruit large-scale cognitive control networks involving frontoparietal and cingulo-opercular systems, with the dorsolateral prefrontal cortex supporting flexible, trial-by-trial control and integration of task sets\u003csup\u003e[43]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn addition, data-intensive work is associated with information-rich yet efficient neural processing. Neuroimaging evidence suggests that higher-order cognitive engagement is supported by brain activity patterns that are both highly informative and highly compressible, reflecting optimized representational efficiency rather than increased neural redundancy\u003csup\u003e[44]\u003c/sup\u003e. Such information-rich yet compressible neural activity patterns may reflect an efficient and flexible mode of processing during high-level cognitive engagement. This characterization is conceptually aligned with descriptions of neural efficiency in the cognitive reserve framework.\u003c/p\u003e \u003cp\u003eFinally, the sustained and cumulative nature of data-related cognitive demands may facilitate long-term neuroplastic adaptation. Continuous engagement in abstract problem solving and decision-making has been linked to synaptic plasticity mechanisms that support adaptive reorganization of neural systems across multiple temporal scales\u003csup\u003e[45]\u003c/sup\u003e. Consistent with a life-course perspective, higher occupational cognitive complexity earlier in adulthood has been associated with better white matter integrity and executive function in midlife, capturing a potential pathway through which prolonged exposure to data complexity contributes to reserve accumulation over time\u003csup\u003e[20]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThese results indicate that data-intensive occupations can simultaneously enhance brain and CRs, optimizing both brain structure and neural efficiency, thereby providing broad protection across multiple cognitive domains\u003csup\u003e[19, 21, 22]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eWe observed that people complexity did not uniformly confer cognitive benefits: although it modestly predicted better reasoning, it was also associated with poorer working memory and language performance, as well as reduced cortical thickness in the left anterior cingulate cortex (ACC-L). This pattern contrasts with several studies reporting protective effects of \u0026ldquo;people-oriented\u0026rdquo; work. For example, higher lifetime people complexity has been linked to better executive and episodic memory performance in the KHANDLE cohort\u003csup\u003e[46]\u003c/sup\u003e, and to lower risks of MCI/dementia and greater MRI-based BR in the SNAD study\u003csup\u003e[47]\u003c/sup\u003e. Yet other community-based or cross-cultural studies have failed to detect consistent cognitive benefits of people complexity after accounting for education, socioeconomic gradients, or heterogeneity in work environments\u003csup\u003e[48]\u003c/sup\u003e, suggesting that the effects of this dimension are not universally positive.\u003c/p\u003e \u003cp\u003eThree sources of heterogeneity may help reconcile the discrepancies between our findings and prior reports suggesting protective effects of people complexity. One important source lies in cultural and occupational-context differences. Studies reporting strong benefits of people complexity typically draw samples with higher education, stable professional careers, and limited involvement in manual or informal labor\u003csup\u003e[9, 47, 49]\u003c/sup\u003e. In these contexts, interpersonal demands often reflect cognitively stimulating managerial or supervisory responsibilities. In contrast, our community-based cohort likely includes a broader range of people-oriented occupations\u0026mdash;particularly service and caregiving roles characterized by low autonomy and high emotional demands. Such roles may transform \u0026ldquo;social engagement\u0026rdquo; into chronic stress rather than cognitive stimulation, echoing recent concerns regarding occupational generalizability\u003csup\u003e[50, 51]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eBeyond contextual variation, methodological heterogeneity in the operationalization of people complexity constitutes a further source of inconsistency across studies. The present study employs the Dictionary of Occupational Titles (DOT), which quantifies interpersonal complexity based on the frequency and intensity of interactions with others. Other studies have also used later occupational coding frameworks such as O*NET; however, both systems share a key limitation: they collapse qualitatively distinct forms of interpersonal demands into a single composite dimension. Methodological reviews have noted that such frameworks often fail to differentiate cognitively stimulating social tasks from emotionally taxing or low-autonomy interactional roles\u003csup\u003e[52, 53]\u003c/sup\u003e. This conflation reduces the ability of occupational complexity measures to capture the critical distinction between cognitively stimulating and emotionally draining forms of interpersonal work, potentially contributing to the mixed pattern of cognitive gains and structural costs observed for the interpersonal dimension in our cohort.\u003c/p\u003e \u003cp\u003eFinally, differences in the motivational and regulatory demands of interpersonal work offer a plausible mechanistic account for the observed trade-offs. High-autonomy interpersonal roles typically support goal-directed cognitive engagement and recruit prefrontal systems associated with reserve accrual, including the frontal pole and dorsolateral prefrontal cortex\u003csup\u003e[54, 55]\u003c/sup\u003e. In contrast, service- and caregiving-oriented roles impose sustained emotional labor, requiring continuous emotion regulation, conflict management, and behavioral compliance, which is associated with elevated allostatic load, reduced cognitive flexibility, and executive-resource competition\u003csup\u003e[56, 57]\u003c/sup\u003e. The magnitude of these trade-offs may further be shaped by cultural norms regarding social roles and relational expectations. In collectivistic contexts that emphasize interdependence and social harmony, interpersonal occupations often carry stricter demands for emotional regulation, role compliance, and collaboration, potentially intensifying cognitive and physiological costs\u003csup\u003e[58\u0026ndash;60]\u003c/sup\u003e. By contrast, in individualistic contexts valuing autonomy and self-expression, similar interpersonal demands may be experienced as less taxing, allowing more open communication and active engagement\u003csup\u003e[58, 61]\u003c/sup\u003e. Collectively, these mechanisms are consistent with the \u0026ldquo;resource competition\u0026rdquo; hypothesis aligns with the profile observed in our cohort, wherein modest improvements in reasoning coexisted with deficits in working memory and language.\u003c/p\u003e \u003cp\u003eThe protective effects of things complexity were the most limited, being associated only with gray matter volume in motor-related regions and failing to contribute to CR, which may be because occupations high in things complexity predominantly involve repetitive, routine tasks with limited innovation or problem-solving, providing minimal stimulation to neural circuits critical for higher-order cognitive processes\u003csup\u003e[62, 63]\u003c/sup\u003e, reflecting the constraints of mechanized tasks in driving higher-order neuroplasticity\u003csup\u003e[64]\u003c/sup\u003e. However, individuals with low things complexity could achieve cross-domain cognitive compensation through engagement in late-life cognitive and social activities, highlighting the value of alternative environmental stimulation\u003csup\u003e[13]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eFurthermore, distinguishing between BR and CR enabled a more precise delineation of the neural pathways through which occupational complexity contributes to cognitive aging. Data complexity was the only dimension simultaneously associated with preserved structural markers and higher CR quantified using the residual-based approach (Methods). Together, these patterns support a dual-pathway model, in which structural preservation provides an anatomical buffer (BR), whereas the residual cognitive advantage reflects more efficient or flexible neural processing (CR). Importantly, the CR index captures variance in cognitive performance unexplained by contemporaneous brain structure, allowing it to be empirically separable from BR. Whereas brain reserve captures the absolute capacity of neural resources, including their structural quantity and integrity, cognitive reserve reflects the efficiency and adaptability of the neural processes that support performance despite age-related structural decline. This methodological distinction permits the dissociation of structural preservation from processing-level compensation, which is essential for isolating the unique contribution of occupational experiences.\u003c/p\u003e \u003cp\u003eAt the regional level, the hippocampus\u0026mdash;central to episodic encoding, retrieval, and attentional allocation\u0026mdash;has been consistently associated with superior cognitive performance in aging cohorts and frequently identified as a structural substrate of occupational or lifestyle reserve proxies\u003csup\u003e[65, 66]\u003c/sup\u003e. The inferior temporal gyrus (ITG), which supports high-level visual\u0026ndash;semantic representation and object/word recognition, may further contribute to cross-domain transfer effects involving language, reasoning, and visuospatial processing. Recent evidence indicates that occupation-related reserve proxies correspond to increased ITG volume and to CR-related moderation of pathology\u0026ndash;performance associations\u003csup\u003e[67]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eBy contrast, people- and things-related complexity showed more restricted and domain-specific neural correlates. People complexity was linked to a localized reduction in cortical thickness within the left anterior cingulate cortex (ACC-L) and did not coincide with higher residual cognitive reserve. Considering the ACC\u0026rsquo;s well-established involvement in conflict monitoring, emotion regulation, and effort allocation\u0026mdash;as well as its documented susceptibility to prolonged emotional and social demands\u0026mdash;this focal reduction in thickness is more plausibly interpreted as reflecting chronic functional load on this region\u003csup\u003e[68]\u003c/sup\u003e. This neural profile is consistent with the selective pattern of cognitive effects observed in the current study, in which slight gains in reasoning ability coexisted with lower performance in working memory and language. In contrast, things complexity was primarily associated with structural variation in sensorimotor cortices, which support motor and perceptual functions, suggesting that object- or skill-focused occupations may enhance localized BR without generating cross-domain CR benefits\u003csup\u003e[23, 69]\u003c/sup\u003e. Together, these dissociable patterns highlight the complementary roles of BR and CR in the context of the present study, in which BR was operationalized through region-specific structural markers, including left hippocampal volume and right inferior temporal gyrus cortical thickness, and CR was quantified using a residual-based modeling approach capturing cognitive performance beyond what could be predicted from brain structure. This methodological distinction likely contributes to the observed dissociation: BR reflects structural reinforcement shaped by data-related occupational demands, whereas CR captures broader, task-independent processing advantages, allowing individuals to maintain cognitive performance despite age-related or localized structural decline.\u003c/p\u003e \u003cp\u003eMoreover, from a life-course perspective, this study confirmed that the protective effects of occupational complexity can persist and be reinforced after retirement through engagement in late-life activities. Path analyses indicated that high occupational complexity, particularly in the data dimension, served as an important predictor of sustained participation in cognitively stimulating activities post-retirement. Notably, retirement duration positively moderated the pathway from \u0026ldquo;occupational complexity \u0026rarr; cognitive activity \u0026rarr; cognitive function.\u0026rdquo; This suggests that cognitive and BR accumulated during the occupational period provide a solid foundation for cognitive health in later life, while leisure activities after retirement play a role in activating and maintaining these reserves. As retirement duration increases, the protective effects, grounded in occupational experience and driven by late-life engagement, become increasingly pronounced, forming a synergistic effect across the lifespan\u003csup\u003e[70]\u003c/sup\u003e. Importantly, the observation that these protective effects become more evident with longer retirement duration may reflect an adaptive process during the post-retirement transition. The initial phase after retirement can involve a sudden loss of structured mental demands, which may temporarily attenuate cognitive benefits, as suggested by longitudinal findings showing accelerated decline in some cognitive domains immediately post-retirement\u003csup\u003e[71]\u003c/sup\u003e. Over time, however, individuals with higher occupational complexity are likely better equipped to navigate this transition, gradually re-establishing cognitively stimulating routines and engaging in leisure activities that reinforce previously accumulated cognitive and brain reserves\u003csup\u003e[72]\u003c/sup\u003e. These findings emphasize the long-term value of cognitive stimulation during the occupational period and integrate it with lifestyle factors in retirement to form a coherent model of healthy cognitive aging.\u003c/p\u003e \u003cp\u003eThe present study makes several key contributions to the literature on occupational complexity and cognitive aging. By disentangling the heterogeneous effects of the three occupational complexity dimensions, our findings refine and extend the theoretical framework of environmental complexity. Building on this distinction, we further propose and empirically validate a dual-pathway model linking occupational complexity to both brain structure and cognitive reserve, thereby elucidating their synergistic yet dissociable mechanisms. Importantly, by situating these pathways within a life-course perspective, the study highlights the dynamic interplay between midlife occupational experience and late-life engagement, offering an integrative account of cognitive aging trajectories.Taken together, these findings not only consolidate previously fragmented evidence but also yield actionable implications for the development of targeted cognitive interventions and policy strategies aimed at promoting healthy aging.\u003c/p\u003e \u003cp\u003eNonetheless, certain limitations should be noted. The cross-sectional and short-term follow-up design restricts causal inference. Although the O*NET indicators offer objective quantification, they cannot fully capture within-occupation heterogeneity. In addition, the cognitive reserve index derived in the present study reflects a model-dependent construct that is contingent on the specific cognitive and neuroanatomical variables included in the estimation procedure. Because CR was inferred from individual differences in cognitive performance relative to structural brain measures, its magnitude may vary with the choice of cognitive domains assessed and the extent to which the selected brain markers capture relevant neurobiological variation. As a result, the residual-based CR measure may incorporate sources of variance that are not uniquely attributable to reserve-related processes, and its comparability across studies employing different variable sets may be limited. Finally, potential biases in self-reported late-life activities warrant caution. Future studies should adopt longitudinal designs with harmonized cognitive and neural assessments to further refine reserve quantification and to validate the dynamic effects of occupational complexity on cognitive aging, while exploring occupation-specific intervention strategies tailored to diverse occupational trajectories.\u003c/p\u003e"},{"header":"5 Conclusion","content":"\u003cp\u003eThis study provides large-scale behavioral and neuroimaging evidence that occupational complexity constitutes an important life-course determinant of late-life cognitive health. Data complexity stands out as the primary driver, conferring broad cognitive benefits through coordinated enhancement of BR and CR. In contrast, people and things complexity exert narrower or context-dependent effects, suggesting substantial heterogeneity in how occupational environments shape cognitive aging. By delineating these domain-specific pathways and identifying reserve-based mechanisms, our findings highlight occupational complexity as a modifiable and quantifiable target for dementia prevention strategies. Interventions that enrich cognitive stimulation during working life or reinforce reserve accumulation after retirement may yield long-term benefits for maintaining cognitive resilience in aging populations.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eBR\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eBrain Reserve\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCR\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eCognitive Reserve\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eMCI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eMild Cognitive Impairment\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eMRI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eMagnetic Resonance Imaging\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eBABRI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eBeijing Aging Brain Rejuvenation Initiative\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eADL\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eActivities of Daily Living\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eIADL\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eInstrumental Activities of Daily Living\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eMMSE\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eMini-Mental State Examination\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAVLT\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eAuditory Verbal Learning Test\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eROCF\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eRey\u0026ndash;Osterrieth Complex Figure Test\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eDST\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eDigit Span Test\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSIM\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eSimilarities Test\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCDT\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eClock Drawing Test\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCVFT\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eCategory Verbal Fluency Test\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eBNT\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eBoston Naming Test\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSDMT\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eSymbol Digit Modalities Test\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eTMT\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eTrail Making Test\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSCWT-C\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eStroop Color-Word Test C\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eFDR\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eFalse Discovery Rate\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eFWE\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eFamily-Wise Error\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eVBM\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eVoxel-Based Morphometry\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSBM\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eSurface-Based Morphometry\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eROI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eRegion of Interest\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAAL\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eAutomated Anatomical Labeling\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eGMV\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eGray Matter Volume\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eTIV\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eTotal Intracranial Volume\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eVIFs\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eVariance Inflation Factors\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSD\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eStandard Deviation\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eITG\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eInferior Temporal Gyrus\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eACC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eAnterior Cingulate Cortex\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":" \u003ch2\u003eEthics approval and consent to participate\u003c/h2\u003e \u003cp\u003e The study was conducted in accordance with the Declaration of Helsinki and was approved by the institutional review board (IRB) at the State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University (protocol code was ICBIR_A_0041_002_02). All participants provided written informed consent for our protocol.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConsent for publication\u003c/strong\u003e \u003cp\u003eNot applicable\u003c/p\u003e \u003ch2\u003eCompeting interests\u003c/h2\u003e \u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e \u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThis work was supported by STI2030-Major Projects (2022ZD0211600), Beijing Natural Science Foundation (5262011), Open Research Fund of the State Key Laboratory of Cognitive Neuroscience and Learning and Tang Scholar.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eLRNX, JW, XL, and ZJZ contributed to the conception and design of the study; LY, JWL, and TL contributed to the acquisition and analysis of data; LRNX, and LY contributed to drafting the original manuscript and preparing the figures; LXY, KWC, XL, and ZJZ contributed to the reviewing and editing the manuscript. All authors approved the final manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eThe authors would like to express their gratitude to the participants and staff involved in data collection and management in the Beijing Aging Brain Rejuvenation Initiative.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe datasets used and analysed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eJia L, Quan M, Fu Y, Zhao T, Li Y, Wei C, Tang Y, Qin Q, Wang F, Qiao Y, Shi S. Dementia in China: epidemiology, clinical management, and research advances. The Lancet Neurology. 2020 Jan 1;19(1):81\u0026ndash;92.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChapko D, McCormack R, Black C, Staff R, Murray A. Life-course determinants of cognitive reserve (CR) in cognitive aging and dementia\u0026ndash;a systematic literature review. Aging \u0026amp; mental health. 2018 Aug 3;22(8):921\u0026thinsp;\u0026minus;\u0026thinsp;32.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eL\u0026ouml;vd\u0026eacute;n M, Fratiglioni L, Glymour MM, Lindenberger U, Tucker-Drob EM. Education and cognitive functioning across the life span. Psychological science in the public interest. 2020 Aug;21(1):6\u0026ndash;41.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHultsch DF, Hertzog C, Small BJ, Dixon RA. Use it or lose it: engaged lifestyle as a buffer of cognitive decline in aging?. Psychology and aging. 1999 Jun;14(2):245.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchooler C. Psychological effects of complex environments during the life span: A review and theory. Intelligence. 1984 Oct 1;8(4):259\u0026thinsp;\u0026minus;\u0026thinsp;81.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCain PS, Treiman DJ. The dictionary of occupational titles as a source of occupational data. American Sociological Review. 1981 Jun 1:253\u0026thinsp;\u0026minus;\u0026thinsp;78.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndel R, Crowe M, Pedersen NL, Mortimer J, Crimmins E, Johansson B, Gatz M. Complexity of work and risk of Alzheimer's disease: a population-based study of Swedish twins. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences. 2005 Sep 1;60(5):P251-8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarp A, Andel R, Parker MG, Wang HX, Winblad B, Fratiglioni L. Mentally stimulating activities at work during midlife and dementia risk after age 75: follow-up study from the Kungsholmen Project. The American journal of geriatric psychiatry. 2009 Mar 1;17(3):227\u0026thinsp;\u0026minus;\u0026thinsp;36.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmart EL, Gow AJ, Deary IJ. Occupational complexity and lifetime cognitive abilities. Neurology. 2014 Dec 9;83(24):2285-91.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKr\u0026ouml;ger E, Andel R, Lindsay J, Benounissa Z, Verreault R, Laurin D. Is complexity of work associated with risk of dementia? The Canadian Study of Health and Aging. American journal of epidemiology. 2008 Apr 1;167(7):820\u0026thinsp;\u0026minus;\u0026thinsp;30.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOltmanns J, Godde B, Winneke AH, Richter G, Niemann C, Voelcker-Rehage C, Sch\u0026ouml;mann K, Staudinger UM. Don\u0026rsquo;t lose your brain at work\u0026ndash;The role of recurrent novelty at work in cognitive and brain aging. Frontiers in Psychology. 2017 Feb 6;8:117.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFinkel D, Andel R, Gatz M, Pedersen NL. The role of occupational complexity in trajectories of cognitive aging before and after retirement. Psychology and aging. 2009 Sep;24(3):563.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndel R, Silverstein M, K\u0026aring;reholt I. The role of midlife occupational complexity and leisure activity in late-life cognition. Journals of Gerontology Series B: Psychological Sciences and Social Sciences. 2015 Mar 1;70(2):314\u0026thinsp;\u0026minus;\u0026thinsp;21.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStern Y, Barnes CA, Grady C, Jones RN, Raz N. Brain reserve, cognitive reserve, compensation, and maintenance: operationalization, validity, and mechanisms of cognitive resilience. Neurobiology of aging. 2019 Nov 1;83:124-9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePettigrew C, Soldan A. Defining cognitive reserve and implications for cognitive aging. Current neurology and neuroscience reports. 2019 Jan;19(1):1.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSzcześniak D, Lenart-Bugla M, Misiak B, Zimny A, Sąsiadek M, Połtyn-Zaradna K, Zatońska K, Zatoński T, Szuba A, Smith EE, Yusuf S. Unraveling the protective effects of cognitive reserve on cognition and brain: a cross-sectional study. International Journal of Environmental Research and Public Health. 2022 Sep 27;19(19):12228.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNithianantharajah J, Hannan AJ. The neurobiology of brain and cognitive reserve: mental and physical activity as modulators of brain disorders. Progress in neurobiology. 2009 Dec 10;89(4):369\u0026thinsp;\u0026minus;\u0026thinsp;82.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGroot C, van Loenhoud AC, Barkhof F, van Berckel BN, Koene T, Teunissen CC, Scheltens P, van der Flier WM, Ossenkoppele R. Differential effects of cognitive reserve and brain reserve on cognition in Alzheimer disease. Neurology. 2018 Jan 9;90(2):e149-56.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuo C, Le\u0026oacute;n I, Brodaty H, Trollor J, Wen W, Sachdev P, Valenzuela MJ. Supervisory experience at work is linked to low rate of hippocampal atrophy in late life. Neuroimage. 2012 Nov 15;63(3):1542-51.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKaup AR, Xia F, Launer LJ, Sidney S, Nasrallah I, Erus G, Allen N, Yaffe K. Occupational cognitive complexity in earlier adulthood is associated with brain structure and cognitive health in midlife: The CARDIA study. Neuropsychology. 2018 Nov;32(8):895.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNyberg L, L\u0026ouml;vd\u0026eacute;n M, Riklund K, Lindenberger U, B\u0026auml;ckman L. Memory aging and brain maintenance. Trends in cognitive sciences. 2012 May 1;16(5):292\u0026ndash;305.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoots EA, Schultz SA, Almeida RP, Oh JM, Koscik RL, Dowling MN, Gallagher CL, Carlsson CM, Rowley HA, Bendlin BB, Asthana S. Occupational complexity and cognitive reserve in a middle-aged cohort at risk for Alzheimer's disease. Archives of Clinical Neuropsychology. 2015 Nov 1;30(7):634\u0026thinsp;\u0026minus;\u0026thinsp;42.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLenhart L, Nagele M, Steiger R, Beliveau V, Skalla E, Zamarian L, Gizewski ER, Benke T, Delazer M, Scherfler C. Occupation-related effects on motor cortex thickness among older, cognitive healthy individuals. Brain Structure and Function. 2021 May;226(4):1023-30.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBlix E, Perski A, Berglund H, Savic I. Long-term occupational stress is associated with regional reductions in brain tissue volumes. PloS one. 2013 Jun 11;8(6):e64065.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSavic I. Structural changes of the brain in relation to occupational stress. Cerebral Cortex. 2015 Jun 1;25(6):1554-64.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePetersen RC, Morris JC. Mild cognitive impairment as a clinical entity and treatment target. Archives of neurology. 2005 Jul 1;62(7):1160-3.\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. Science bulletin. 2022 Nov 30;67(22):2285-90.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDekhtyar S, Marseglia A, Xu W, Darin-Mattsson A, Wang HX, Fratiglioni L. Genetic risk of dementia mitigated by cognitive reserve: a cohort study. Annals of neurology. 2019 Jul;86(1):68\u0026ndash;78.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuh D, Karunananthan S, Bergman H, Cooper R. A life-course approach to healthy ageing: maintaining physical capability. Proceedings of the Nutrition Society. 2014 May;73(2):237\u0026thinsp;\u0026minus;\u0026thinsp;48.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoyle R, Knight SP, De Looze C, Carey D, Scarlett S, Stern Y, Robertson IH, Kenny RA, Whelan R. Verbal intelligence is a more robust cross-sectional measure of cognitive reserve than level of education in healthy older adults. Alzheimer's research \u0026amp; therapy. 2021 Jul 12;13(1):128.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen Y, Lv C, Li X, Zhang J, Chen K, Liu Z, Li H, Fan J, Qin T, Luo L, Zhang Z. The positive impacts of early-life education on cognition, leisure activity, and brain structure in healthy aging. Aging (Albany NY). 2019 Jul 17;11(14):4923.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReed BR, Mungas D, Farias ST, Harvey D, Beckett L, Widaman K, Hinton L, DeCarli C. Measuring cognitive reserve based on the decomposition of episodic memory variance. Brain. 2010 Aug 1;133(8):2196\u0026thinsp;\u0026minus;\u0026thinsp;209.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStern Y, Arenaza-Urquijo EM, Bartr\u0026eacute;s‐Faz D, Belleville S, Cantilon M, Chetelat G, Ewers M, Franzmeier N, Kempermann G, Kremen WS, Okonkwo O. Whitepaper: Defining and investigating cognitive reserve, brain reserve, and brain maintenance. Alzheimer's \u0026amp; Dementia. 2020 Sep;16(9):1305-11.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZahodne LB, Manly JJ, Brickman AM, Narkhede A, Griffith EY, Guzman VA, Schupf N, Stern Y. Is residual memory variance a valid method for quantifying cognitive reserve? A longitudinal application. Neuropsychologia. 2015 Oct 1;77:260-6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Loenhoud AC, Wink AM, Groot C, Verfaillie SC, Twisk J, Barkhof F, van Berckel B, Scheltens P, van der Flier WM, Ossenkoppele R. A neuroimaging approach to capture cognitive reserve: application to Alzheimer's disease. Human brain mapping. 2017 Sep;38(9):4703-15.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNelson ME, Veal BM, Andel R, Martinkova J, Veverova K, Horakova H, Nedelska Z, Lacz\u0026oacute; J, Vyhnalek M, Hort J. Moderating effect of cognitive reserve on brain integrity and cognitive performance. Frontiers in aging neuroscience. 2022 Nov 3;14:1018071.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeitz K, Bittner N, Heim S, Caspers S. Bilingualism and \u0026ldquo;brain reserve\u0026rdquo; in subregions of the hippocampal formation. GeroScience. 2025 Jun;47(3):4935-54.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRaine PJ, Rao H. Volume, density, and thickness brain abnormalities in mild cognitive impairment: an ALE meta-analysis controlling for age and education. Brain imaging and behavior. 2022 Oct;16(5):2335-52.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu Y, Julkunen V, Paajanen T, Westman E, Wahlund LO, Aitken A, Sobow T, Mecocci P, Tsolaki M, Vellas B, Muehlboeck S. Education increases reserve against Alzheimer\u0026rsquo;s disease\u0026mdash;evidence from structural MRI analysis. Neuroradiology. 2012 Sep;54(9):929\u0026thinsp;\u0026minus;\u0026thinsp;38.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWisch JK, Petersen K, Millar PR, Abdelmoity O, Babulal GM, Meeker KL, Braskie MN, Yaffe K, Toga AW, O'Bryant S, Ances BM. Cross-Sectional Comparison of Structural MRI Markers of Impairment in a Diverse Cohort of Older Adults. Human Brain Mapping. 2025 Feb 1;46(2):e70133.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchwarz J, Zistler F, Usheva A, Fix A, Zinn S, Zimmermann J, Knolle F, Schneider G, Nuttall R. Investigating dynamic brain functional redundancy as a mechanism of cognitive reserve. Frontiers in Aging Neuroscience. 2025 Feb 4;17:1535657.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTurcotte V, Potvin O, Dadar M, Hudon C, Duchesne S, Alzheimer\u0026rsquo;s Disease Neuroimaging Initiative. Birth cohorts and cognitive reserve influence cognitive performances in older adults. Journal of Alzheimer\u0026rsquo;s Disease. 2022 Jan 18;85(2):587\u0026ndash;604.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCocchi L, Halford GS, Zalesky A, Harding IH, Ramm BJ, Cutmore T, Shum DH, Mattingley JB. Complexity in relational processing predicts changes in functional brain network dynamics. Cerebral Cortex. 2014 Sep 1;24(9):2283-96.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOwen LL, Manning JR. High-level cognition is supported by information-rich but compressible brain activity patterns. Proceedings of the National Academy of Sciences. 2024 Aug 27;121(35):e2400082121.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePallares Di Nunzio M, Mart\u0026iacute;n Tenti J, Arlego M, Rosso OA, Montani F. Exploring the role of synaptic plasticity in the frequency-dependent complexity domain. Chaos: An Interdisciplinary Journal of Nonlinear Science. 2025 Feb 1;35(2).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSoh Y, Eng CW, Mayeda ER, Whitmer RA, Lee C, Peterson RL, Mungas DM, Glymour MM, Gilsanz P. Association of primary lifetime occupational cognitive complexity and cognitive decline in a diverse cohort: Results from the KHANDLE study. Alzheimer's \u0026amp; Dementia. 2023 Sep;19(9):3926-35.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eColeman ME, Roessler ME, Peng S, Roth AR, Risacher SL, Saykin AJ, Apostolova LG, Perry BL. Social enrichment on the job: Complex work with people improves episodic memory, promotes brain reserve, and reduces the risk of dementia. Alzheimer's \u0026amp; Dementia. 2023 Jun;19(6):2655-65.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhong T, Li S, Liu P, Wang Y, Chen L. The impact of education and occupation on cognitive impairment: a cross-sectional study in China. Frontiers in Aging Neuroscience. 2024 Jul 11;16:1435626.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eV\u0026eacute;lez-Coto M, Andel R, P\u0026eacute;rez-Garc\u0026iacute;a M, Caracuel A. Complexity of work with people: Associations with cognitive functioning and change after retirement. Psychology and Aging. 2021 Mar;36(2):143.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNappo N. Job stress and interpersonal relationships cross country evidence from the EU15: A correlation analysis. BMC Public Health. 2020 Jul 20;20(1):1143.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuh C, Punnett L. High emotional demands at work and poor mental health in client-facing workers. International Journal of Environmental Research and Public Health. 2022 Jun 20;19(12):7530.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFantozzi IC, Di Luozzo S, Schiraldi MM. On tasks and soft skills in operations and supply chain management: analysis and evidence from the O* NET database. The TQM Journal. 2024 Dec 16;36(9):53\u0026ndash;74.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHandel MJ. The O* NET content model: strengths and limitations. Journal for Labour Market Research. 2016 Oct;49(2):157\u0026thinsp;\u0026minus;\u0026thinsp;76.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHosoda C, Tsujimoto S, Tatekawa M, Honda M, Osu R, Hanakawa T. Plastic frontal pole cortex structure related to individual persistence for goal achievement. Communications Biology. 2020 Apr 28;3(1):194.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReeve J, Lee W. Autonomy recruits neural support for interest and learning. Motivation and Emotion. 2025 Apr 11:1\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu X, He T, Yu S, Duan J, Gao R. The effects of emotional labor on work strain and nonwork strain among dancers: A Person-centered approach. Psychology research and behavior management. 2023 Dec 31:3675-85.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXiong W, Huang M, Okumus B, Leung XY, Cai X, Fan F. How emotional labor affect hotel employees\u0026rsquo; mental health: A longitudinal study. Tourism Management. 2023 Feb 1;94:104631.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu C, Tu YH, Lin LJ, Chen H, Liu TH, Lin HL, Liu R, Chiou WK. Doctor-Patient communication models, patient decision-making participation, and patient emotional expression: a cross-cultural comparison of samples from the UK and China. Patient preference and adherence. 2025 Dec 31:2505-24.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuwa\u0026euml; S, Schaafsma J. Cross-cultural differences in emotion suppression in everyday interactions. International Journal of Psychology. 2018 Jun;53(3):176\u0026thinsp;\u0026minus;\u0026thinsp;83.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng S, Masuda T, Matsunaga M, Noguchi Y, Ohtsubo Y, Yamasue H, Ishii K. Cultural differences in social support seeking: The mediating role of empathic concern. Plos one. 2021 Dec 30;16(12):e0262001.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiller JG, Goyal N, Wice M. A cultural psychology of agency: Morality, motivation, and reciprocity. Perspectives on Psychological Science. 2017 Sep;12(5):867\u0026thinsp;\u0026minus;\u0026thinsp;75.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGollo LL, Roberts JA, Cropley VL, Di Biase MA, Pantelis C, Zalesky A, Breakspear M. Fragility and volatility of structural hubs in the human connectome. Nature neuroscience. 2018 Aug;21(8):1107-16.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMetzler-Baddeley C, Caeyenberghs K, Foley S, Jones DK. Task complexity and location specific changes of cortical thickness in executive and salience networks after working memory training. Neuroimage. 2016 Apr 15;130:48\u0026ndash;62.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiao Y, Burzynska AZ, Fisher GG, Bielak AA. Occupational experiences and brain health outcomes in older age.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eO\u0026rsquo;Shea A, Cohen RA, Porges EC, Nissim NR, Woods AJ. Cognitive aging and the hippocampus in older adults. Frontiers in aging neuroscience. 2016 Dec 8;8:298.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang W, Wang J, Guo J, Dove A, Qi X, Bennett DA, Xu W. Association of cognitive reserve indicator with cognitive decline and structural brain differences in middle and older age: findings from the UK Biobank. The Journal of Prevention of Alzheimer's Disease. 2024 May 1;11(3):739\u0026thinsp;\u0026minus;\u0026thinsp;48.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eElshiekh A, Subramaniapillai S, Rajagopal S, Pasvanis S, Ankudowich E, Rajah MN. The association between cognitive reserve and performance-related brain activity during episodic encoding and retrieval across the adult lifespan. Cortex. 2020 Aug 1;129:296\u0026ndash;313.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLichenstein SD, Verstynen T, Forbes EE. Adolescent brain development and depression: a case for the importance of connectivity of the anterior cingulate cortex. Neuroscience \u0026amp; Biobehavioral Reviews. 2016 Nov 1;70:271\u0026thinsp;\u0026minus;\u0026thinsp;87.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu J, Kua EH, Mahendran R, Ng TK. ChatGPT-estimated occupational complexity predicts cognitive outcomes and cortical thickness above and beyond socioeconomic status among older adults. GeroScience. 2025 Aug;47(4):5709-23.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFisher GG, Stachowski A, Infurna FJ, Faul JD, Grosch J, Tetrick LE. Mental work demands, retirement, and longitudinal trajectories of cognitive functioning. Journal of occupational health psychology. 2014 Apr;19(2):231.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXue B, Cadar D, Fleischmann M, Stansfeld S, Carr E, Kivim\u0026auml;ki M, McMunn A, Head J. Effect of retirement on cognitive function: the Whitehall II cohort study. European journal of epidemiology. 2018 Oct;33(10):989\u0026ndash;1001.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCalatayud E, Oliv\u0026aacute;n-Bl\u0026aacute;zquez B, Aguilar-Latorre A, Cuenca-Zaldivar JN, Magall\u0026oacute;n-Botaya RM, G\u0026oacute;mez-Soria I. Analysis of the effectiveness of a computerized cognitive stimulation program designed from Occupational Therapy according to the level of cognitive reserve in older adults in Primary Care: Stratified randomized clinical trial protocol. Experimental Gerontology. 2024 Oct 15;196:112568.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Occupational complexity, Cognitive aging, BR, CR, Structural MRI, Life-course exposures, Retirement, Cognitive resilience","lastPublishedDoi":"10.21203/rs.3.rs-9143855/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9143855/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e \u003cb\u003eIntroduction\u003c/b\u003e: Occupational complexity is a major source of long-term cognitive stimulation across adulthood, yet its multidimensional effects on cognitive aging and their neural mechanisms remain unclear. This study examined how three dimensions of occupational complexity\u0026mdash;data, people, and things\u0026mdash;shape late-life cognitive performance, and whether brain reserve(BR) and cognitive reserve(CR) mediate these associations.\u003c/p\u003e \u003cp\u003e\u003cb\u003eMethods\u003c/b\u003e: A total of 3,754 retirees meeting sex-specific age eligibility criteria were recruited from the Beijing Aging Brain Rejuvenation Initiative completed a battery of multidomain neuropsychological assessments and standardized occupational history assessments coded using O*NET-based ratings.. A subsample of 851 participants also underwent structural MRI. Linear models, structural equation modeling, and voxel-/surface-based morphometry were used to test (1) dimension-specific associations with cognitive domains, (2) links to global and regional brain structure, and (3) dual-reserve mediation pathways.\u003c/p\u003e \u003cp\u003e \u003cb\u003eResults\u003c/b\u003e: Data complexity emerged as the primary protective dimension, independently predicting higher reasoning (β\u0026thinsp;=\u0026thinsp;0.058), attention (β\u0026thinsp;=\u0026thinsp;0.041), and reduced mild cognitive impairment(MCI) risk (\u0026minus;\u0026thinsp;5.7% after adjustment for education, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05; \u0026minus;14.3% unadjusted, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). After accounting for data complexity, higher people complexity was associated with poorer working memory and language performance.. Things complexity, referring to demands related to tools and physical objects, showed no direct associations but demonstrated compensation via late-life leisure activities.\u003c/p\u003e \u003cp\u003eNeuroimaging showed that data complexity was uniquely associated with larger gray matter volume in frontotemporal\u0026ndash;limbic regions and higher CR. Mediation models revealed that data complexity protected cognition via both BR and CR, with mediation effects observed across multiple cognitive domains..\u003c/p\u003e \u003cp\u003e \u003cb\u003eConclusions\u003c/b\u003e: Occupational complexity, particularly data complexity, is associated with enhanced cognitive aging outcomes, including improved reasoning, attention, and reduced mild cognitive impairment (MCI) risk. Neuroimaging revealed that data complexity predicts greater frontotemporal\u0026ndash;limbic gray matter volume and higher cognitive reserve. Mediation analyses suggested dual reserve pathways, with cognitive reserve mediating multiple cognitive domains, while brain reserve influenced hippocampal and temporal regions. These findings underscore the role of occupational environments in promoting cognitive health and mitigating late-life MCI risk.\u003c/p\u003e","manuscriptTitle":"Distinct Effects of Data Occupational Complexity On Cognitive Aging: Evidence for Dual Brain–Cognitive Reserve Pathways","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-27 09:27:23","doi":"10.21203/rs.3.rs-9143855/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"29a84086-ace3-41dc-9c1e-617e736faf1c","owner":[],"postedDate":"March 27th, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Rejected","date":"2026-05-18T17:21:07+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-15T21:06:05+00:00","index":31,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-05-18T17:24:35+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-27 09:27:23","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9143855","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9143855","identity":"rs-9143855","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}