Sherral Devine

Background: Remote, self-administered digital assessments provide scalable approaches for monitoring cognition in older adults. Practice effects (PEs), once viewed as measurement noise, are increasingly recognized as potential markers of cognitive decline. As disease-modifying therapies for neurodegenerative conditions emerge, longitudinal digital assessments may capture subtle cognitive changes. Although the majority of PE studies have focused on memory tasks, it is critical to determine whether PEs occur in other domains. Clarifying the potential presence of PEs across multiple time points on the Food and Drug Administration–cleared Defense Automated Neurocognitive Assessment (DANA) platform remains critical for determining whether PEs can serve as a measure of cognitive decline longitudinally. Objective: This study aimed to examine whether PEs occur and whether demographics moderate them across quarterly self-administrations of digital cognitive tasks spanning different cognitive domains in cognitively intact older adults. Methods: Cognitively intact older adults from the Framingham Heart Study self-administered DANA tasks at baseline and every 3 months for 1 year (4 assessments: T1-T4). Tasks included Simple Reaction Time (SRT), Go/No-Go (GNG), and Code Substitution (CS). These tasks varied in complexity, from simple motor responding (SRT) to inhibitory control (GNG) and associative learning with working memory demands (CS). The primary outcome was Cognitive Efficiency, a derived measure of speed and accuracy, which are also recorded as secondary measures. PEs were quantified using linear mixed-effects models, and the change in mean score across time points was expressed in pooled SD units as Cohen d . Results: A total of 717 participants completed all 4 time points (mean age 64.1, SD 9.5 years; n=469, 65.4% female participants; n=481, 67.1% college-educated). PEs were observed on all tasks, with small effects on CS ( d =0.43) and GNG ( d =0.32) and a below-threshold effect on SRT ( d =0.16). Demographics influenced baseline performance: older age predicted lower baseline scores on all tasks, females scored lower on SRT and GNG, and higher education predicted better performance on all tasks. Age was the only demographic moderator of PEs, with smaller improvements on CS among participants aged 65 years or older (B=−0.39; P <.001). Conclusions: Quarterly self-administered digital cognitive assessments elicited measurable PEs across processing speed, inhibitory control, and visual scanning, with stronger effects on more complex tasks. Demographics influenced baseline performance, while age moderated PEs on the most demanding task. These findings provide baseline data against which PEs in clinical populations can be compared in future work, complementing prior evidence on other cognitive instruments.

BACKGROUND Remote, self-administered digital assessments provide scalable approaches for monitoring cognition in older adults. Practice effects (PE), once viewed as measurement noise, are increasingly recognized as potential markers of cognitive decline. As disease-modifying therapies for neurodegenerative conditions emerge, longitudinal digital assessments may capture subtle cognitive changes. Although the majority of PE studies have focused on memory tasks, it is critical to determine whether PE occur in other domains. Clarifying the potential presence of PE across multiple timepoints on the FDA-cleared Defense Automated Neurocognitive Assessment (DANA) platform remains critical for determining whether PE can serve as a measure of cognitive decline longitudinally. OBJECTIVE To examine whether PE occur across quarterly self-administrations of digital cognitive tasks across different cognitive domains. Demographic characteristics were examined as potential moderators. METHODS Cognitively intact older adults from the Framingham Heart Study self-administered DANA tasks at baseline and every three months for one year (four assessments, T1-T4). Tasks included Simple Reaction Time (SRT), Go/No-Go (GNG), and Code Substitution (CS). These tasks varied in complexity, from simple motor responding (SRT) to inhibitory control (GNG) and associative learning with working memory demands (CS). The primary outcome was Cognitive Efficiency, a derived measure of speed and accuracy, which are also recorded as secondary measures. PE were quantified using linear mixed-effects models and change in mean score across timepoints expressed in pooled SD units expressed as Cohen d. RESULTS A total of 717 participants completed all four timepoints (mean age: 64.1±9.5 years; 65.4% females; 67.1% college educated). PE were observed on all tasks with small effect (d: SRT=0.16, GNG=0.32, CS=0.43). Demographics influenced baseline performance: older age predicted lower baseline scores on all tasks, females scored lower on SRT and GNG, and higher education predicted better performance on all tasks. Age was the only demographic moderator of PE, with smaller improvements on CS among participants >65 years (β=-0.39, P<.001). CONCLUSIONS Quarterly self-administered digital cognitive assessments elicited measurable PE across processing speed, inhibitory control, and visual scanning, with stronger effects on more complex tasks. Demographics influenced baseline performance, while age moderated PE on the most demanding task. These findings underscore the importance of characterizing PE in digital assessments and highlight their potential clinical utility in detecting early cognitive impairment.

Background: Widely used neuropsychological test instruments are notoriously biased across the demographics of age, sex/gender, education, language and culture. This includes verbal memory tests that elicit speech such as the paragraph recall or list‐learning memory tests. Language tests are similarly biased, including the Boston Diagnostic Aphasia Examination Cookie Theft Test (CTT) that has been used to elicit both written and spoken responses for decades. Biased tests generate biased results, and this problem is exacerbated by the widespread practice of repurposing tests generated for those in largely high income, Westernized countries for use in other countries that are demographically and culturally distinct. We tested the use of large language model (LLM) tools to automatically generate verbal test stimuli that were more locally relevant.Methods: Starting with the CTT, that involves describing a picture that contains 14 core content units, we used ChatGPT to aid in generating customized pictures with the same number of core content units that were more universally relevant, (e.g., open park area, beach, etc.) by suggesting parameters and integrating resulting features into new pictures. We also tested capabilities to generate pictorial scenes that are more representative of regional settings, such as the Asia Pacific or Middle East. We are testing LLM capacity to generate similarly regionally customized list learning stimuli that maintain the original list learning test word frequency and difficulty across languages.Results: Using an iterative approach, we were able to use automated artificial intelligence (AI)‐driven tools to rapidly generate regionally relevant test stimuli that retained the original test characteristics to preserve generation of comparable results. Descriptive comparisons found that aided by automated AI tool, we could generate new neuropsychological test stimuli in a matter of minutes compared to months long manual methods.Conclusion: The rapid emergence of AI has generated significant concerns about its appropriate use for health and health care applications. Given that AI is here to stay, we are testing the capacity of this technological advance to address long‐standing, persistent problems in AD research and usher in a new era of precision brain health that has global relevance in a rapidly aging world population.

Background Emerging studies indicate that cardiovascular health (CVH) is a modifiable risk factor for AD. However, studies on how mid-life Life’s Essential 8 (LE8) scores affect the AD risk and digital cognitive performance are limited. Objective To examine the associations between CVH in middle age, as defined by LE8 scores, and subsequent digital cognitive performance and incident AD. Methods Linear regression and Cox proportional-hazard models were used to examine the associations of mid-age CVH with dCDT performance and incident AD, respectively. Prediction performance of mid-age LE8 scores for incident AD were assessed using ROC curve analysis. Results Every one-SD increase in mid-age LE8 total score was associated with a 0.16-SD increase in dCDT total score (p < 0.001). Ideal CVH in middle age was associated with higher dCDT scores compared to intermediate CVH. Furthermore, higher mid-age LE8 scores were associated to a decreased risk of AD, with ideal CVH significantly lowering AD risk compared to intermediate CVH. The combination of dCDT performance, mid-age LE8 scores, APOE ε4 status, and other covariates provided the best prediction performance for incident AD, with an AUC of 0.84. Notably, mid-life LE8 scores improved its predictive accuracy by 5.7%. Conclusion Our findings emphasize the critical role of CVH in middle age as a predictor of both digital cognitive performance and the risk of developing AD, highlighting the importance of early intervention on CVH to delay the progression of cognitive decline.

Background: Smartphone‐based digital technology is increasingly being recognized as a cost‐effective, scalable, and noninvasive method of collecting longitudinal cognitive and behavioral data. Accordingly, a state‐of‐the‐art 3‐year longitudinal project focused on collecting multimodal digital data for early detection of cognitive impairment was developed. Methods and Results: A smartphone application collected 2 modalities of cognitive data, digital voice and screen‐based behaviors, from the FHS (Framingham Heart Study) multigenerational Generation 2 (Gen 2) and Generation 3 (Gen 3) cohorts. To understand the feasibility of conducting a smartphone‐based study, participants completed a series of questions about their smartphone and app use, as well as sensory and environmental factors that they encountered while completing the tasks on the app. Baseline data collected to date were from 537 participants (mean age=66.6 years, SD=7.0; 58.47% female). Across the younger participants from the Gen 3 cohort (n=455; mean age=60.8 years, SD=8.2; 59.12% female) and older participants from the Gen 2 cohort (n=82; mean age=74.2 years, SD=5.8; 54.88% female), an average of 76% participants agreed or strongly agreed that they felt confident about using the app, 77% on average agreed or strongly agreed that they were able to use the app on their own, and 81% on average rated the app as easy to use. Conclusions: Based on participant ratings, the study findings are promising. At baseline, the majority of participants are able to complete the app‐related tasks, follow the instructions, and encounter minimal barriers to completing the tasks independently. These data provide evidence that designing and collecting smartphone application data in an unsupervised, remote, and naturalistic setting in a large, community‐based population is feasible.

Background: Smartphone‐based cognitive assessments have emerged as promising tools, bridging gaps in accessibility and reducing bias in Alzheimer disease and related dementia research. However, their congruence with traditional neuropsychological tests and usefulness in diverse cohorts remain underexplored. Methods and Results: A total of 406 FHS (Framingham Heart Study) and 59 BHS (Bogalusa Heart Study) participants with traditional neuropsychological tests and digital assessments using the Defense Automated Neurocognitive Assessment (DANA) smartphone protocol were included. Regression models investigated associations between DANA task digital measures and a neuropsychological global cognitive Z score (Global Cognitive Score [GCS]), and neuropsychological domain‐specific Z scores. FHS participants’ mean age was 57 (SD, 9.75) years, and 44% (179) were men. BHS participants' mean age was 49 (4.4) years, and 28% (16) were men. Participants in both cohorts with the lowest neuropsychological performance (lowest quartile, GCS1) demonstrated lower DANA digital scores. In the FHS, GCS1 participants had slower average response times and decreased cognitive efficiency scores in all DANA tasks ( P P P <0.05) in the DANA Code Substitution task. Conclusions: Our findings demonstrate that smartphone‐based cognitive assessments exhibit concurrent validity with a composite measure of traditional neuropsychological tests. This supports the potential of using smartphone‐based assessments in cognitive screening across diverse populations and the scalability of digital assessments to community‐dwelling individuals.

Background: Early detection of brain‐related disorders is key to prevention of Alzheimer’s disease (AD), and research on Hispanic populations is severely lacking. Obstacles to recruitment include the lack of trust in research and researchers and requirements for English proficiency, which excludes many potential Hispanic participants.Methods: Researchers from El Faro Health and Therapeutics in Starr County, Texas and the Framingham Heart Study are teaming up on a digital phenotyping study and incorporating “trust” as a central theme for recruitment/retention of Hispanic participants. Study materials are in both English and Spanish. The smartphone tasks are presented in two formats: “in‐clinic” using an iPad and “remote” utilizing participants’ smartphones. Activities include voice tasks, gait and balance, and game‐like cognitive tasks. Longitudinal data will be collected every three months for one year.Participants: Recruitment is for anyone aged 18+ in Starr County, which is approximately 98% Hispanic. To maximize inclusivity, there are no other inclusion/exclusion criteria. Study procedures are customized to accommodate individuals. Recruitment goal is 500 participants within one year, with planned expansion to 5,000+ over 5 years.Results: Recruitment approach includes several components: (1) El Faro investigators have been providing clinical services to community members for decades, and are respected and trusted members of the community. (2) All participant‐facing study materials are available in both English and Spanish, with most translated into the specific local dialect. (3) Community fliers include text in both English and Spanish. (4) Social media outreach has sparked a great deal of interest for participating in research, and the team attends community events to further spread the word. (5) Multi‐generational families are encouraged given family decision‐making culture; (6) Word‐of‐mouth within the Starr County community is expected to be an important contributor to recruitment.Conclusions: Although data collection will begin in February 2023, already more than 150 participants have expressed interest in being involved. Leveraging resources to enable highly respected and trusted local researchers to conduct the research, and using participant‐centric design approaches will result in greater success in recruitment and retention of this underrepresented population.

Background: More than 75 common variant loci account for only a portion of the heritability for Alzheimer’s disease (AD). A more complete understanding of the genetic basis of AD can be deduced by exploring associations with AD-related endophenotypes.Methods: We conducted genome-wide scans for cognitive domain performance using harmonized and co-calibrated scores derived by confirmatory factor analyses for executive function, language, and memory. We analyzed 103,796 longitudinal observations from 23,066 members of community-based (FHS, ACT, and ROSMAP) and clinic-based (ADRCs and ADNI) cohorts using generalized linear mixed models including terms for SNP, age, SNP × age interaction, sex, education, and five ancestry principal components. Significance was determined based on a joint test of the SNP’s main effect and interaction with age. Results across datasets were combined using inverse-variance meta-analysis. Genome-wide tests of pleiotropy for each domain pair as the outcome were performed using PLACO software.Results: Individual domain and pleiotropy analyses revealed genome-wide significant (GWS) associations with five established loci for AD and AD-related disorders (BIN1,CR1,GRN,MS4A6A, andAPOE) and eight novel loci.ULK2was associated with executive function in the community-based cohorts (rs157405,P = 2.19 × 10^–9). GWS associations for language were identified withCDK14in the clinic-based cohorts (rs705353,P = 1.73 × 10^–8) andLINC02712in the total sample (rs145012974,P = 3.66 × 10^–8).GRN(rs5848,P = 4.21 × 10^–8) andPURG(rs117523305,P = 1.73 × 10^–8) were associated with memory in the total and community-based cohorts, respectively. GWS pleiotropy was observed for language and memory withLOC107984373(rs73005629,P = 3.12 × 10^–8) in the clinic-based cohorts, and withNCALD(rs56162098,P = 1.23 × 10^–9) andPTPRD(rs145989094,P = 8.34 × 10^–9) in the community-based cohorts. GWS pleiotropy was also found for executive function and memory withOSGIN1(rs12447050,P = 4.09 × 10^–8) andPTPRD(rs145989094,P = 3.85 × 10^–8) in the community-based cohorts. Functional studies have previously linked AD toULK2,NCALD, andPTPRD.Conclusion: Our results provide some insight into biological pathways underlying processes leading to domain-specific cognitive impairment and AD, as well as a conduit toward a syndrome-specific precision medicine approach to AD. Increasing the number of participants with harmonized cognitive domain scores will enhance the discovery of additional genetic factors of cognitive decline leading to AD and related dementias.

Background: The digital Clock Drawing Test (dCDT) has been recently used as a more objective tool to assess cognition. However, the association between digitally obtained clock drawing features and structural neuroimaging measures has not been assessed in large population-based studies. Objective: We aimed to investigate the association between dCDT features and brain volume. Methods: This study included participants from the Framingham Heart Study who had both a dCDT and magnetic resonance imaging (MRI) scan, and were free of dementia or stroke. Linear regression models were used to assess the association between 18 dCDT composite scores (derived from 105 dCDT raw features) and brain MRI measures, including total cerebral brain volume (TCBV), cerebral white matter volume, cerebral gray matter volume, hippocampal volume, and white matter hyperintensity (WMH) volume. Classification models were also built from clinical risk factors, dCDT composite scores, and MRI measures to distinguish people with mild cognitive impairment (MCI) from those whose cognition was intact. Results: A total of 1656 participants were included in this study (mean age 61 years, SD 13 years; 50.9% women), with 23 participants diagnosed with MCI. All dCDT composite scores were associated with TCBV after adjusting for multiple testing (P value <.05/18). Eleven dCDT composite scores were associated with cerebral white matter volume, but only 1 dCDT composite score was associated with cerebral gray matter volume. None of the dCDT composite scores was associated with hippocampal volume or WMH volume. The classification model for differentiating MCI and normal cognition participants, which incorporated age, sex, education, MRI measures, and dCDT composite scores, showed an area under the curve of 0.897. Conclusions: dCDT composite scores were significantly associated with multiple brain MRI measures in a large community-based cohort. The dCDT has the potential to be used as a cognitive assessment tool in the clinical diagnosis of MCI.

Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: The digital Clock Drawing Test (dCDT) has been recently used as a more objective tool to assess cognition. However, the association between digitally obtained clock drawing features and structural neuroimaging measures has not been assessed in large population-based studies. Objective: We aimed to investigate the association between dCDT features and brain volume. Methods: This study included participants from the Framingham Heart Study who had both a dCDT and magnetic resonance imaging (MRI) scan, and were free of dementia or stroke. Linear regression models were used to assess the association between 18 dCDT composite scores (derived from 105 dCDT raw features) and brain MRI measures, including total cerebral brain volume (TCBV), cerebral white matter volume, cerebral gray matter volume, hippocampal volume, and white matter hyperintensity (WMH) volume. Classification models were also built from clinical risk factors, dCDT composite scores, and MRI measures to distinguish people with mild cognitive impairment (MCI) from those whose cognition was intact. Results: A total of 1656 participants were included in this study (mean age 61 years, SD 13 years; 50.9% women), with 23 participants diagnosed with MCI. All dCDT composite scores were associated with TCBV after adjusting for multiple testing (P value <.05/18). Eleven dCDT composite scores were associated with cerebral white matter volume, but only 1 dCDT composite score was associated with cerebral gray matter volume. None of the dCDT composite scores was associated with hippocampal volume or WMH volume. The classification model for differentiating MCI and normal cognition participants, which incorporated age, sex, education, MRI measures, and dCDT composite scores, showed an area under the curve of 0.897. Conclusions: dCDT composite scores were significantly associated with multiple brain MRI measures in a large community-based cohort. The dCDT has the potential to be used as a cognitive assessment tool in the clinical diagnosis of MCI.