Frontiers in Aging Neuroscience

Introduction: Cognitive intraindividual variability (IIV), a measure of fluctuation in performance across time or tasks, is an emerging marker of cognitive decline. However, little is known about within-domain IIV in Latino adults. To address this gap, we examined memory IIV and hippocampal integrity in older Latino individuals. Methods: Sixty-seven participants from the Boston Latino Aging Study underwent cognitive testing and MRI. Linear regressions assessed whether memory IIV was associated with hippocampal volume and hippocampus-cingulum tract fractional anisotropy. Additional analyses evaluated associations using mean memory performance as the outcome. Results: Higher mean performance was associated with greater hippocampal volume (=0.24, p=.03), while memory IIV was not (=0.028, p=.84). Higher memory IIV was associated with lower fractional anisotropy in the hippocampus-cingulum tract (=-0.34, p<.01), while mean performance was not (=-0.07, p=.52). Discussion: Memory IIV may be more closely linked to hippocampal white matter, while mean memory performance may better reflect hippocampal volume.

INTRODUCTION: Alzheimer's disease (AD) involves early sensory and perceptual changes that can precede memory decline. We developed a two-layer rate-based neural model of early and higher visual areas with feedforward, feedback, and recurrent connections to examine how excitatory/inhibitory (E/I) balance shapes visual motion perception in AD. METHODS: Using Motion-Induced Position Shift and Altered Motion Repulsion stimuli, we factorially manipulated excitatory (E) and inhibitory (I) subfield power and connectivity to examine their effects on illusion representation across visual layers. RESULTS: Increased E, decreased I, and increased E/I reduced illusion representation, consistent with reduced illusion susceptibility in AD, whereas the converse increased illusion representation, reflecting neurotypical processing. Higher-level layer 2 representations were more resistant to E/I perturbations than lower-level layer 1. Illusion representation was less affected by E/I increases in intra-areal (recurrent) connectivity than in inter-areal (feedforward/feedback) connectivity. DISCUSSION: This mechanistic framework links E/I balance and connectivity to altered visual motion illusion representations in AD and suggests that E/I and connectivity measures may provide probes of early AD neurobiology. The framework may also generalize beyond vision.

Apolipoprotein E (APOE) genotype is theorised to influence dementia risk partly through effects on blood-brain barrier (BBB) function, but clinical studies examining APOE effects on BBB biomarkers report inconsistent results. Therefore, this systematic review and meta-analysis quantified associations between APOE genotype and two commonly used fluid biomarkers of CNS barrier function: cerebrospinal fluid (CSF)/serum albumin quotient (Q-Alb), and soluble platelet-derived growth factor receptor beta (sPDGFRB). Ovid MEDLINE, Embase, and Web of Science were searched on 30th June, 2026. Eligible peer-reviewed human studies reported Q-Alb or blood or CSF sPDGFRB; by APOE genotype. Risk of bias was assessed using the Joanna Briggs Institute checklist. Twenty-eight studies involving 5,375 participants were included: 17 assessed Q-Alb and 11 assessed sPDGFRB. Random-effects meta-analyses determined that APOE e2 carriage was associated with higher Q-Alb (g = 0.29, 95% CI 0.11 to 0.46). APOE e4 carriage was not associated with Q-Alb overall (g = .13, 95% CI -.17 to .43) or sPDGFRB overall (g = .10, 95% CI -.21 to .40). Genotype-specific Q-Alb analyses suggested medium positive but non-significant effects among e4/e4 and e2/e4 carrier subgroups. For sPDGFRB, exploratory subgroup analysis revealed effect estimates near zero in cognitively unimpaired participants, and a small positive non-significant effect in cognitively impaired participants. Larger effects were also observed in samples with an older mean age, although this was not statistically significant. Overall, evidence for APOE e4-associated alterations in Q-Alb and sPDGFRB remained inconclusive, highlighting the need for further study across the spectrum of cognitive ageing and dementia. In contrast, APOE e2 carriage was associated with greater barrier dysfunction as indexed by Q-Alb, broadening the predominant focus on APOE e4 and identifying e2 as a potentially important but understudied contributor to altered neurovascular function.

The discrepancy between chronological age and predicted brain age, derived from neuroimaging data, serves as a biomarker for neurological health. However, the exact electroencephalography (EEG) mechanisms contributing to the variation of brain age are yet to be understood. This paper comprehensively investigated the relationship between brain age estimated from EEG-based functional connectivity and cognitive or mental manifestations of healthy adults. A mean absolute error (MAE) of 5.97 and R-squared (R2) of 0.85 were reported for training data, while an MAE of 10.73 and R2 of 0.56 were reported for the prediction data. Increased brain age was found to be negatively related to the scores of memory and attention tasks. Most young adults (19 of 26) suffering from mental symptoms such as addictions, depression, phobia, and anorexia nervosa demonstrated higher brain age than their chronological age. Alcohol consumption was also significantly correlated to the difference between brain and chronological ages. The functional connectivity of frontal and central motor regions was found to be most distinguishing between individuals with high and low brain ages. These brain regions were significantly correlated with brain age but not chronological age, suggesting a relatively stronger relationship with biological age. Ablation studies further confirmed the importance of the frontal brain and all bands EEG in the prediction of brain age. This work advances the understanding of EEG mechanisms contributing to brain age and could provide an interpretation for physiological or psychological conditions associated with brain age.

Reduced visibility is a common contributor to falls in older adults, but the neural processes that support safe locomotion under low lighting are not well established. In this study, 22 young (mean age = 22.5) and 20 older (mean age = 70.9) adults walked along a pathway under bright, ambient and dark lighting conditions while stepping over expected and unexpected obstacles projected on the floor. Mobile electroencephalography (EEG), combined with motion capture, was used to quantify spectral power changes during obstacle crossing preparation and the post-crossing phases, relative to a standing baseline. We identified that reduced lighting conditions elicited a widespread increase in cortical engagement, predominantly in the alpha band, during both preparation and reset phases. Obstacle negotiation recruited a distributed sensorimotor network, with reduced theta-alpha-beta desynchronisation during preparation and a robust post-movement beta rebound during the post-crossing reset phase. Expectation effects were modest, with unexpected obstacles inducing greater transient alpha-beta desynchronisation relative to expected obstacles, suggesting additional cortical resources are recruited under uncertainty. Age effects were minimal and limited to reduced frontal alpha desynchronisation in older adults compared to young adults during the preparation phase, possibly reflecting differences in walking speed rather than differences in neural control. In conclusion, environmental features such as low lighting and obstacle presence exert a stronger influence on cortical dynamics during walking than age and expectation effects alone. This provides a foundation for future work examining how these processes may differ in older adults at increased risk of falls.

Sex-steroid hormones influence brain structure across aging, but their relationship to the cerebellum, a region implicated in cognition and vulnerable to age-related decline, remains poorly characterized, particularly longitudinally and across both sexes. We examined whether 17{beta}-estradiol, progesterone, and testosterone relate to cerebellar volume across two timepoints, and whether these relationships differ by sex. In a mixed-sex sample of aging adults (Baseline N = 137; Year 1 N = 93), free salivary hormones and cerebellar lobular volumes were quantified. Linear regression tested cross-sectional, predictive, and change-score hormone-volume associations, and formal hormone-by-sex interactions, in a priori primary and secondary cerebellar lobules of interest. Hormone-cerebellum relationships were consistently sex-differentiated. Estradiol change related to Crus I volume change in opposite directions in males and females, a dissociation extending across bilateral Crus I and Lobule VI, and Right Crus II. In females, greater increases in testosterone were associated with greater decline in Left Crus I and Right Lobule V volumes. In males, testosterone showed no significant associations in primary analyses, however, increases in estradiol were associated with larger volumes. Progesterone showed a male-specific cross-sectional association with Crus II. Several relationships ran opposite to the direction predicted by estradiols established neurotrophic role. Sex-steroid influences on the aging cerebellum are sex-differentiated, regionally specific, and may not be predicted exclusively by cortical or female-focused models. These findings establish the cerebellum as a potential site of sex-specific hormonal influence on brain aging and also further highlight the complexity of the impact of sex-steroid hormones on the aging brain.

Introduction The basal forebrain is an early site of pathology in Alzheimer's disease (AD). Within this region, the nucleus basalis of Meynert (NbM) provides the primary cholinergic input to the neocortex and supports cortical activation during rapid eye movement (REM) sleep. REM electroencephalography (EEG) slowing is observed in older adults with AD and amnestic mild cognitive impairment (MCI). This study examined the relationship between REM EEG slowing and NbM volume, and whether it differed by cognitive status, sex, or sleep apnea severity. Methods Participants aged [&le;] 60 years with cognitive concerns underwent neuropsychological and medical assessments, polysomnography, and magnetic resonance imaging (MRI). REM EEG slowing was quantified as the ratio of delta and theta (<8Hz) to alpha, sigma, and beta (8-32Hz) power. Bilateral NbM and Ch1-3 volume were derived from T1-weighted MRI images. Multiple linear regression and moderation analyses adjusted for age, sex, and sleep apnea severity. Exploratory analyses examined associations between NbM volume, sleep macroarchitecture, and memory. Results The sample comprised 116 participants (59.5% female; mean age = 70.5 years, 64.7% MCI). Greater REM EEG slowing was associated with smaller NbM volume (all p < 0.05). Although moderation effects were not significant after FDR correction, within-group analyses suggested a stronger relationship in MCI. Smaller NbM volume was associated with longer sleep latency and poorer episodic memory (all p < 0.05). Discussion REM EEG slowing may reflect early NbM degeneration and provide a non-invasive biomarker of basal forebrain integrity in older adults with cognitive concerns. Longitudinal studies are required to establish its prognostic value.

INTRODUCTION: This study examined the extent to which educational attainment, Alzheimer's disease (AD) knowledge, and research attitudes are associated with willingness to participate in AD research, and whether AD knowledge and research attitudes mediate associations between educational attainment and willingness to participate. METHODS: 201 older adults completed self-report measures of AD knowledge, research attitudes, prior research participation, and willingness to participate in diagnostic procedures (neuropsychological assessment, blood draw, brain MRI) and non-pharmacological intervention research (tracking physical activity, aerobic exercise). RESULTS: Favorable research attitudes were consistently associated with greater willingness to participate across all proposed research activities, and research attitudes mediated the relationship between educational attainment and willingness to participate in research. In contrast, AD knowledge did not mediate the association between educational attainment and willingness to participate. DISCUSSION: Research attitudes may be a key, modifiable determinant of willingness to participate in AD research and a promising focus for broadening AD research participation.

Objectives: The rising prevalence of financial exploitation among older adults has prompted research on the cognitive mechanisms underlying financial decision making in later life. Prior work has demonstrated a link between cognitive impairment and financial vulnerability. The present study examined the unique contributions of composite measures of memory and executive functioning to scam susceptibility in middle-aged to older adults without dementia. Methods: One hundred and fifty-one participants (age= 68.2 {+/-} 8.0 years, 72.2% female, years of education= 16.4 {+/-} 2.2) completed standard neuropsychological measures from the Uniform Data Set (UDS), Version 3, and the California Verbal Learning Test, Second Edition, and a self-report scam susceptibility measure. Psychometrically robust composite scores, the UDS-M+ and UDS3-EF, were computed to represent memory and executive functioning domains. Linear regression models examined independent associations between UDS-M+ and UDS3-EF performance and scam susceptibility. Nested model comparisons evaluated unique contributions of the UDS-M+ and UDS3-EF when included in a joint model. Results: Lower UDS-M+ (b= -0.25, 95% CI [-0.45, -0.04], p= .018) and lower UDS3-EF (b= -0.33, 95% CI [-0.54, -0.12], p= .002) scores were independently associated with greater scam susceptibility after adjusting for age, sex, years of education, and income. However, only UDS3-EF remained significant in a joint model. Post hoc analyses revealed that only memory was associated with scam susceptibility in middle-aged adults whereas only executive functioning was significant in older adults. Conclusions: These findings highlight the importance of monitoring memory and executive functioning across middle to older adulthood when considering susceptibility to scams.

Introduction: Cerebral small vessel disease (CSVD) lesions, including white matter hyperintensities (WMHs), lacunar infarcts, cerebral microbleeds (CMBs), and enlarged perivascular spaces (PVSs) are associated with increased risk of dementia. In this study, we aimed to examine their associations with risk factors, neurodegeneration, and cognition over a follow-up period of ten years. Methods: 289 participants (183 females, mean age: 73.7 +/- 5.7) with multi-modal magnetic resonance imaging (MRI) and clinical assessments from the Consortium for the Early Identification of Alzheimer's Disease (CIMA-Q) were included. CSVD lesions and neurodegeneration were quantified using automated tools. Linear regression and mixed effects models examined the cross-sectional and longitudinal associations between lesions, risk factors, neurodegeneration, and cognitive decline. Results: Risk factors showed the strongest cross-sectional associations with PVS number and CMB presence, and the strongest longitudinal associations with WMH and PVS volume progression (p < 0.05). WMH and PVS volumes significantly increased over time (p < 0.01), with faster WMH progression among participants who experienced cognitive decline during study follow-ups. Higher WMH volume was associated with poorer global cognition and executive function, and presence of lacunae was associated with faster decline in executive function. Finally, both WMH and PVS were associated with longitudinal atrophy across several cortical regions. Conclusions: Vascular risk factors were associated with CSVD burden and progression, and WMH and PVS progression were further associated with cortical atrophy and cognitive decline, underscoring the interconnected contributions of vascular risk, CSVD, and neurodegeneration to cognitive aging and dementia risk.

INTRODUCTION: Whether APOE e4 modifies relationships between subclinical cardiac phenotypes and brain structure is unknown. METHODS: In Framingham Heart Study Offspring participants, we related antecedent echocardiographic remodeling (Examinations 4-6) to repeated brain MRI, and cardiac magnetic resonance phenotypes to regional cortical morphometry, testing cardiac x APOE e4 interactions with false-discovery-rate (FDR) correction and permutation testing of the regional screen. RESULTS: APOE e4 modified the association between antecedent left atrial (LA) remodeling and subsequent lateral ventricular expansion (beta=-0.00858, P=1.45 x 10^-4, q=0.0023; N=1,305; 3,787 MRI observations), the only FDR-significant result of 16 longitudinal tests, including with participant-specific MRI-time slopes (q=0.034); it weakened when intracranial volume x MRI time was modeled. Four regional ventricular cardiac x APOE e4 interactions met within-family FDR criteria, but their number was consistent with chance (permutation P=0.70). Maximum LA volume index showed an APOE e4-dependent association with left middle-occipital cortical-thickness variability (whole-brain q=0.017; permutation P=0.041) that persisted after ventricular adjustment. Cardiac phenotypes and trajectories showed no FDR-significant APOE e4-dependent associations with plasma Abeta42/40, p-tau181, GFAP, or NfL, and adjustment for these biomarkers changed the regional interactions by at most 3.2%. Allen Human Brain Atlas data showed cholesterol-biosynthesis enrichment in the right superior frontal cortex, implicated by the LV end-systolic volume interaction, but without convergent pathway evidence. DISCUSSION: APOE e4 modified atrial-brain relationships, with the strongest evidence linking antecedent atrial remodeling to later ventricular expansion. Replication with prespecified models is needed.

Background Attentional control declines with normal aging, yet the role of modifiable lifestyle factors in accelerating this decline remains incompletely characterised in large, diverse populations. We examined associations between four modifiable risk factors (smoking intensity, alcohol binge drinking, social satisfaction and blood pressure) and sustained attention performance in 13,866 adults drawn from the All of Us Research Program. Methods Cognitive performance was assessed using the Gradual-Onset Continuous Performance Task (GradCPT). A composite cognitive efficiency score was derived via principal component analysis of four GradCPT metrics: d' (perceptual sensitivity), response criterion, median reaction time and reaction time variability. Associations between cognitive efficiency and each risk factor were examined using multiple linear regression adjusted for age, sex and race/ethnicity. Effect sizes were expressed as age-equivalent years of cognitive aging (age-equivalent years = regression coefficient for the risk factor divided by the regression coefficient for age) to aid clinical interpretation. Results GradCPT performance declined progressively with age (p < 0.001). Social satisfaction emerged as the strongest lifestyle predictor: low social satisfaction was associated with approximately 5.8 additional age-equivalent years of cognitive aging relative to high social satisfaction. Smoking showed a dose-response pattern, with heavy smokers (> 20 cigarettes/day) exhibiting an approximately 5-year age-equivalent deficit versus never-smokers. Stage 2 hypertension showed a modest, non-significant negative association. Binge drinking showed limited significant associations, likely complicated by survivor and abstainer biases. Bootstrapped analyses of combined exposures showed that combinations of risk factors, particularly those involving low social satisfaction, produced larger cognitive deficits than expected under an additive model. Conclusions These findings support an interactive rather than a purely additive model of lifestyle-related cognitive vulnerability and highlight the GradCPT as a subclinical measure of attention-based cognitive risk in non-clinical populations. Keywords Cognitive aging, Sustained attention, GradCPT, Modifiable risk factors, Social isolation, Smoking, Hypertension, All of Us Research Program

Purpose: Age is the leading factor for age-related macular degeneration (AMD), in which retinal pigmented epithelium (RPE) atrophy is a hallmark. The nuclear factor-erythroid 2-related factor 2 (Nrf2), a major anti-stress transcription factor, decreases in AMD, suggesting bioenergetic stress interacts with age during AMD pathogenesis. We profiled visual function and gene expression of the retina and RPE-choroid with Nrf2 deletion and aging. Methods: We tested 3mo and 14mo Nrf2 null (Nrf2 KO) and Nrf2 WT mice for optokinetic tracking (OKT) and dark-adapted electroretinography (ERG) responses. Eyes were harvested for retina outer nuclear layer (ONL) thickness analysis and retina and RPE-choroid cup RNA isolation for RNAseq profiling. Results: Visual acuity was reduced with aging with lower values in the Nrf2 KO vs Nrf2 WT comparison at 14mo. We observed aged-related reduction of a-, b-, and c-wave ERG amplitudes in both WT and KO mice. Age-related c-wave decline was exacerbated by Nrf2 loss. By RNAseq, two-way ANOVA of RPE-choroid samples revealed differentially expressed genes (DEGs) with age and genotype main effects, as well as DEGs with age and genotype interaction effects. Age DEGs were over-represented in pathways related to inflammation and phagocytosis. Genotype DEGs were associated with alterations in antioxidant and inflammatory processes. Retina RNAseq showed age and genotype DEGs. Conclusions: Ablation of Nrf2 exacerbates age-related declines suggesting that decrease in Nrf2 with aging has functional consequences for inflammation/resiliency. We provide retina and RPE-choroid bulk-RNAseq datasets from the same cohort of mice that can be interrogated to better understand degeneration of retina/RPE-choroid.

Social isolation and loneliness are major determinants of well-being in older adults. Shrinking social networks and age-related shifts in social behavior and decision-making may increase vulnerability to loneliness. These changes in social decision-making strategies can be conceptualized within an exploration-exploitation tradeoff framework, in which social options with known outcomes are prioritized over uncertain ones. Yet, no study has characterized how complementary decision-making strategies, undirected novelty seeking and goal-directed exploration, relate to social network constructs or are supported by distinct neural systems. These strategies capture distinct forms of decision-making: undirected novelty seeking is driven by the intrinsic salience of novel options independent of expected informational gain, whereas goal-directed exploration involves deliberately sampling uncertain options to reduce uncertainty and optimize future choices. Here, we used fMRI and a novel social bandit reinforcement task to identify dissociable social constructs and brain systems linked to undirected novelty seeking and goal-directed exploration in older adults. Undirected novelty seeking was associated with social network size and with activity of temporoparietal and mid-cingulate brain areas. Goal-directed exploration predicted subjective feelings of social connectedness and was associated with activity of anterior cingulate and prefrontal brain areas. Crucially, choice probabilities from a non-social version of the three-arm bandit task were not associated with social decision-making nor with social constructs. Our findings suggest that, in later life, dissociable social decision-making strategies, when faced with unknown options, rely on distinct neural systems supporting social network size as well as subjective feelings of social connectedness.

Speech-based cognitive monitoring for Alzheimer's employs various tasks, but whether timing measures transfer across tasks is unclear. Twelve senior living residents completed four 60-second tasks: category fluency, two picture descriptions, and a self-generated narrative. Speech timing measures were calculated identically across tasks. Speech-rate correspondence among picture descriptions and narrative was substantial (mean Spearman rho; = 0.72). Correspondence with category fluency increased from Cookie Theft to contemporary picture to narrative in terms of speech rate (Spearman rho; = 0.36, 0.51, 0.64) and pause time ratio (Spearman rho; = 0.14, 0.45, 0.71). Measures were not interchangeable; correspondence followed an ordering consistent with shared retrieval demand.

Modest positive associations between episodic memory performance and whole hippocampal and hippocampal subfield volumes have been reported in numerous prior studies. A smaller number of studies have reported associations between hippocampal volume and performance on tests of non-mnemonic cognition. The present study examined whether these associations were evident in a lifespan sample of cognitively healthy adults. Of particular interest was whether any identified associations were sensitive to age, and whether associations between subfield volumes and mnemonic and non-mnemonic performance were subfield dependent. We acquired high-resolution T1- and T2-weighted structural images from 163 adults (18-87 years of age). Participants also undertook a comprehensive neuropsychological test battery and an in-scanner test of source memory. Principal components analysis was employed to reduce the neuropsychological test scores to 5 cognitive components. Two components reflected memory performance while the other three reflected different aspects of non-mnemonic cognition. Hippocampal subfields (Cornu Ammonis (CA)1, CA2-3, dentate gyrus (DG) and subiculum) were segmented and measured with the Automated Segmentation of Hippocampus Subfields (ASHS) package. Source memory performance was selectively associated across participants with CA2-3 volume. By contrast, both mnemonic and non-mnemonic component scores derived from the test battery were associated exclusively with the volume of the DG. All associations were age-invariant. The findings indicate that different cognitive domains can be dissociated by virtue of their associations with different hippocampal subfields. Of importance, these associations appear to be life-long and hence are unlikely to reflect individual differences in age-related decline in structural integrity.

Sleep disturbances compromise cognition, and overall physical and mental health; however, animal models of these effects remain limited, and typically impose more extreme disturbances than those experienced by humans. Here, we use behavioural interventions in mice to model two common types of sleep disruption: (1) sleep displacement associated with shiftwork and (2) sleep fragmentation associated with aging and disease. C57BL6/J mice were subjected to either a sleep deprivation protocol (n = 8, females = 4) or a fragmentation protocol (n = 8, females = 4), in which they were completely deprived of sleep or awoken every 2 min respectively, for 12 hours during their typical rest phase (lights-on) across 7 consecutive days. Age-matched control mice (n = 14, females = 7) remained undisturbed in their home cages. Activity during the 12-hour active phase was recorded via infrared motion sensors. After protocol completion, mice were injected with bromodeoxyuridine (BrdU, 200mg/kg) 2 hours prior to brain and blood collection; hippocampal neurogenesis and plasma corticosterone were then assessed. Results show that the sleep deprivation and fragmentation protocols had different, dynamic effects on behaviour; sleep deprivation was associated with an accumulating sleep need, operationalized as progressively reduced locomotor activity across the week, whereas fragmentation produced only an acute reduction of locomotor activity. Cell proliferation rates and corticosterone did not differ from controls. These protocols provide a foundation for future research into the mechanistic links between shiftwork and sleep fragmentation on health and cognition.

INTRODUCTION: Adult hippocampal neurogenesis (AHN) plays an important role in memory and is regulated by astrocytes. Since astrocytic function deteriorates with pathological ageing and Alzheimers disease (AD), we hypothesize that the astrocytic secretome may rescue AHN and cognition. METHODS: We tested the effect of astrocyte-conditioned solution (ACS) on AHN and memory performances in young and aged mice and in two models of AD. RESULTS: ACS increased AHN and improved performance in the object location test in young and aged wild-type mice. In THY-Tau22 mice, a model of tau pathology, ACS restored AHN and cognitive performance. In APP/PS1 mice, a model of amyloid pathology, ACS reduced gliogenesis, and rescued deficits in AHN and spatial memory. In addition, ACS attenuated circulating pro-inflammatory cytokines and the reaction of plaque-associated microglia. DISCUSSION: The astrocytic secretome is a promising, niche-targeted strategy to enhance hippocampal plasticity and memory across ageing and AD.

Identifying progressive dementia during its prodromal stage is a challenging yet critical task for healthcare providers. Traditional clinical assessments are often not sensitive enough to detect warning signs of cognitive decline. Machine learning (ML) methods have shown promise in modeling clinical trajectories associated with neurodegenerative disorders such as Alzheimer's disease (AD). Validation of these models' forecasting abilities, however, is widely lacking in the literature. To address this gap, we established a longitudinal cohort from the National Alzheimer's Coordinating Center (n = 15,727) comprising individuals who were cognitively normal (CN) or had mild cognitive impairment (MCI) at baseline. Cognitive status and clinical observations collected over up to 10 visits were used to train several ML classifiers. We employed the gradient-boosted tree ensemble method XGBoost with a custom feature-engineering pipeline. On non-imputed data, our models achieved an AUC-ROC of 0.9447 [0.9417-0.9476] and PR-AUC of 0.8554 [0.8497-0.8611] for the CN to MCI task, and an AUC-ROC of 0.9059 [0.8983-0.9134] and PR-AUC of 0.9214 [0.9156-0.9272] for the MCI to AD task. In a lead-time forecasting analysis spanning mean follow-up periods of 56.1 months (range, 9.1-214.1) for CN to MCI and 42.7 months (range, 9.1-126.1) for MCI to AD, the non-imputed models detected future progression before diagnosis in 29.58% [28.29%-30.87%] of CN to MCI progressors and 22.41% [21.36%-23.46%] of MCI to AD progressors. Among these individuals, the mean lead time was 33.04 months [32.52-33.56] and 17.95 months [17.71-18.19], respectively. False alarm rates among non-progressors were 12.6% [11.5%-13.7%] and 20.0% [18.4%-21.6%], respectively. These findings suggest that longitudinal machine learning models can identify future cognitive decline before diagnosis, indicating clinical utility for early dementia risk stratification.

Background: Associations of sleep duration and insomnia symptoms with cognitive decline are inconsistent, and whether they represent modifiable risk factors or early manifestations of neurodegeneration may depend on life stage. We examined how sleep-cognition associations vary with age (44 to >90 years). Methods: We studied 16,884 participants from three cohorts with harmonised measures. In the English Longitudinal Study of Ageing (ELSA: n=9,473, median age 64, range 50 to >=90), linear mixed-effects models examined age-specific differences in cognitive change (recall and fluency) associated with sleep duration, difficulty falling asleep, nocturnal awakenings, incorporating sleep-by-age-by-time interactions, with initial adjustment for sociodemographic factors and further adjustment for health and lifestyle factors. Sleep characteristics showing consistent age-dependent associations in ELSA were further evaluated in Whitehall II (n=5,580, median age 55, range 44 to 68) and the Rush Memory and Aging Project (MAP: n=1,831, median age 80, range 53 to 102). Findings: In ELSA, difficulty falling asleep >=1 time/week (vs <1 time/week) was associated with less favourable 5-year change in recall at age 50 (difference -0.12 SD, 95% CI -0.20 to -0.04), with the association attenuating and reversing towards more favourable change at older ages (p-interaction=0.009 [linear], 0.016 [quadratic]). These persisted after further adjustment for health and lifestyle factors. Short sleep (<=6 h) showed similar patterns that attenuated after further adjustment. Similar age gradients were observed for difficulty falling asleep in Whitehall II and MAP. There was no evidence of age modification on long sleep and nocturnal awakenings. Interpretation: The cognitive implications of sleep disturbances vary by life stage. In midlife, difficulty falling asleep was associated with faster cognitive decline, supporting its role as a modifiable risk factor. In advanced age, the reversal is consistent with diminished arousal capacity accompanying prodromal dementia. Funding: US National Institute on Aging