Abstract 028: Biomarkers Of Kidney Function And Cognitive Ability: A Mendelian Randomization Study

Background: Estimated glomerular filtration rate (eGFR), albuminuria and serum uric acid (SUA) are markers of kidney function that have been associated with cognitive ability. However, whether these associations are causal is unclear. Methods: We performed one-sample Mendelian randomization (MR) to estimate the effects of kidney function markers on cognitive performance using data from 357,590 participants from the UK Biobank. Polygenic scores for serum uric acid (SUA), urine albumin to creatinine ratio (ACR), estimated glomerular filtration rate based on serum creatinine (eGFRcre) and serum cystatin-c (eGFRcys) were used as instruments, and cognitive function outcomes included a test of verbal-numeric reasoning and reaction time. Sensitivity analyses were carried out to address potential pleiotropy using MR-Egger and weighted median regression. Results: We found no evidence of a causal effect of genetically determined SUA, eGFRcre or eGFRcys on either cognitive function outcomes. There was no association between a polygenic score for ACR and verbal-numeric reasoning. However, there was suggestive evidence of a relationship between genetically increased ACR and slower reaction time (β (95% confidence interval [CI])) for 1 standard deviation log ACR = 4.93 (1.60 to 8.26), p=0.004). Pleiotropy adjusted estimates were directionally consistent with those of the principal analysis but overlapped with the null. Conclusions: This MR study does not support causal effects of SUA, eGFRcre or eGFRcys on cognitive performance. Genetically-increased ACR was associated with lower processing speed, but results need confirmation in independent samples.

Pre-pandemic cognitive function and COVID-19 mortality: prospective cohort study

Poorer performance on standard tests of pre-morbid cognitive function is related to an elevated risk of death from lower respiratory tract infections but the link with coronavirus (COVID‑19) mortality is untested. Participants in UK Biobank, aged 40 to 69 years at study induction (2006–10), were administered a reaction time test, an indicator of information processing speed, and also had their verbal-numeric reasoning assessed. Between April 1st and September 23rd 2020 there were 388 registry-confirmed deaths (138 women) ascribed to COVID-19 in 494,932 individuals (269,602 women) with a reaction time test result, and 125 such deaths (38 women) in the subgroup of 180,198 people (97,794 women) with data on verbal-numeric reasoning. In analyses adjusted for age, sex, and ethnicity, a one standard deviation slower reaction time was related to a higher rate of death from COVID-19 (hazard ratio; 95% confidence interval: 1.18; 1.09, 1.28), as was a one standard deviation disadvantage on the verbal-numeric reasoning test (1.32; 1.09, 1.59). While there was some attenuation in these relationships after adjustment for additional covariates which included socio-economic status and lifestyle factors, the two pre-pandemic indicators of cognitive function continued to be related to COVID-19 mortality.

Pre-pandemic cognitive function and COVID-19 mortality: prospective cohort study

BackgroundPoorer performance on standard tests of cognitive function is related to an elevated risk of death from lower respiratory tract infections. Whether pre-pandemic measures of cognition are related to COVID-19 mortality is untested.MethodsUK Biobank, a prospective cohort study, comprises around half a million people who were aged 40 to 69 years at study induction between 2006 and 2010 when a reaction time test was administered to the full sample, and verbal-numeric reasoning assessed in a subgroup. Death from COVID-19 was ascertained from participant linkage to a UK-wide national registry.ResultsBetween April 1st and September 23rd 2020, there were 388 deaths (138 women) ascribed to COVID-19 in the 494,932 individuals (269,602 women) with a reaction time test result, and 125 such deaths (38 women) in the 180,198 (97,794 women) for whom there were data on verbal-numeric reasoning. In analyses adjusted for age, sex, and ethnicity, a one standard deviation (118.2 msec) slower reaction time was related to a higher rate of death from COVID-19 (hazard ratio; 95% confidence interval: 1.18; 1.09, 1.28). A one standard deviation disadvantage (2.16 point) on the verbal-numeric reasoning test was also associated with an elevated risk of death (1.32; 1.09, 1.59). Attenuation after adjustment for additional covariates followed a similar pattern for both measures of cognition. For verbal-numeric reasoning, for instance, the hazard ratios were 1.22 (0.98, 1.51) after control for socioeconomic status, 1.16 (0.96, 1.41) after lifestyle factors, 1.25 (1.04, 1.52) after co-morbidity, and 1.29 (1.01, 1.64) after physiological indices.ConclusionsIn the present study, poorer performance on two pre-pandemic indicators of cognitive function, including reaction time, a knowledge-reduced measure, was related to death ascribed to COVID-19.

Optimal Planning Modulo Theories

We consider the problem of planning with arithmetic theories, and focus on generating optimal plans for numeric domains with constant and state-dependent action costs. Solving these problems efficiently requires a seamless integration between propositional and numeric reasoning. We propose a novel approach that leverages Optimization Modulo Theories (OMT) solvers to implement a domain-independent optimal theory-planner. We present a new encoding for optimal planning in this setting and we evaluate our approach using well-known, as well as new, numeric benchmarks.

Issues and Opportunities in Objective Numeracy Research

This chapter, “Issues and Opportunities in Objective Numeracy Research,” discusses three cross-cutting questions in objective numeracy research. The first two issues concern the correlational nature of most objective numeracy research. Alternative explanations exist for the effects of numeracy on decisions and life outcomes. In particular, this chapter questions whether general intelligence can explain objective numeracy effects and whether the reverse causal path may offer a better explanation. Objective numeracy generally emerges as the better explanation, but alternative explanations remain of some effects (e.g., worse health sometimes may produce lower numeracy). The third issue concerns researchers’ experimental design decisions and what these results might teach us about how to improve numeric reasoning in decisions.

Five-factor model personality traits and cognitive function in five domains in older adulthood

Background Five-factor model (FFM) personality traits have been associated consistently with risk of Alzheimer’s disease and related dementias (ADRD). Less is known about how these traits are associated with functioning in specific domains of cognitive function in older adulthood. Methods Participants (N = 2865) were drawn from the 2016 Harmonized Cognitive Assessment Protocol sub-study of the Health and Retirement Study (HRS). Participants completed a battery of cognitive tasks that measured performance in five domains: Memory (eight tasks), speed-attention-executive (five tasks), visuospatial ability (three tasks), fluency (one task), and numeric reasoning (one task). Participants completed an FFM personality measure as part of the regular HRS assessment in either 2014 or 2016. Linear regression was used to examine the association between the traits and each cognitive task and composite scores for the five domains, controlling for age, sex, race, ethnicity, and education. We also tested whether the associations were moderated by these sociodemographic factors or mental status. Results Neuroticism was associated with worse performance on all of the cognitive tasks. Conscientiousness was associated with better performance across all five cognitive domains, although not necessarily with every task. Openness and Agreeableness were associated with better performance in all domains, except for numeric reasoning. Extraversion was associated with better speed-attention-executive and fluency. There was no robust evidence that the association between personality and cognition was moderated by sociodemographic characteristics or global cognitive function. Conclusions Personality traits have pervasive associations with functioning across five cognitive domains. Consistent with the literature on personality and risk of ADRD, Neuroticism and Conscientiousness were associated with cognitive performance in the expected direction in all domains. Extraversion was the only trait that showed domain-specific associations. The present research supports models of personality and health in the context of cognition and suggests that personality is associated with intermediate markers of cognitive health.

The RANTANPLAN planner: system description

RANTANPLAN is a numeric planning solver that takes advantage of recent advances in satisfiability modulo theories. It extends reduction to SAT approaches with an easy and efficient handling of numeric fluents using background theories. In this paper, we describe the design choices and features of RANTANPLAN, especially, how numeric reasoning is integrated in the system. We also provide experimental results showing that RANTANPLAN is competitive with existing exact numeric planners.

A Hybrid LP-RPG Heuristic for Modelling Numeric Resource Flows in Planning

Although the use of metric fluents is fundamental to many practical planning problems, the study of heuristics to support fully automated planners working with these fluents remains relatively unexplored. The most widely used heuristic is the relaxation of metric fluents into interval-valued variables --- an idea first proposed a decade ago. Other heuristics depend on domain encodings that supply additional information about fluents, such as capacity constraints or other resource-related annotations. A particular challenge to these approaches is in handling interactions between metric fluents that represent exchange, such as the transformation of quantities of raw materials into quantities of processed goods, or trading of money for materials. The usual relaxation of metric fluents is often very poor in these situations, since it does not recognise that resources, once spent, are no longer available to be spent again. We present a heuristic for numeric planning problems building on the propositional relaxed planning graph, but using a mathematical program for numeric reasoning. We define a class of producer--consumer planning problems and demonstrate how the numeric constraints in these can be modelled in a mixed integer program (MIP). This MIP is then combined with a metric Relaxed Planning Graph (RPG) heuristic to produce an integrated hybrid heuristic. The MIP tracks resource use more accurately than the usual relaxation, but relaxes the ordering of actions, while the RPG captures the causal propositional aspects of the problem. We discuss how these two components interact to produce a single unified heuristic and go on to explore how further numeric features of planning problems can be integrated into the MIP. We show that encoding a limited subset of the propositional problem to augment the MIP can yield more accurate guidance, partly by exploiting structure such as propositional landmarks and propositional resources. Our results show that the use of this heuristic enhances scalability on problems where numeric resource interaction is key in finding a solution.

Validity Improvement in Two Reasoning Measures Following the Elimination of the Position Effect

We report on an investigation of the consequences of preventing the position effect from contributing to the correlation between two reasoning scales considered an indicator of convergent validity. Confirmatory factor models served to separate this effect from the ability-specific components of measurement in Raven’s Advanced Progressive Matrices and Horn’s Numeric Reasoning Scale. These models additionally enabled investigation of the convergent validity of these measures. In a sample of 230 participants the ability-specific components of the two reasoning measures showed a very high correlation (.88), whereas the correlation of the position-related components was not significant. Since the measures represented different types of reasoning, we conclude that the ability-specific components represent fluid intelligence and the position-related components reflect the specificities of the different types of reasoning.