Elsevier

Intelligence

Volume 68, May–June 2018, Pages 70-81
Intelligence

Age-moderation of genetic and environmental contributions to cognitive functioning in mid- and late-life for specific cognitive abilities

https://doi.org/10.1016/j.intell.2017.12.004Get rights and content

Highlights

  • The etiology of cognitive functioning varied across age and cognitive ability.

  • Genetic contributions were higher for older ages, but only for verbal tasks.

  • Processing speed and spatial processing tests had stable heritability across age.

  • The shared environment had little to no effect on any of the tests we examined.

Abstract

Age moderation of genetic and environmental contributions to Digits Forward, Digits Backward, Block Design, Symbol Digit, Vocabulary, and Synonyms was investigated in a sample of 14,534 twins aged 26 to 98 years. The Interplay of Genes and Environment across Multiple Studies (IGEMS) consortium contributed the sample, which represents nine studies from three countries (USA, Denmark, and Sweden). Average test performance was lower in successively older age groups for all tests. Significant age moderation of additive genetic, shared environmental, and non-shared environmental variance components was observed, but the pattern varied by test. The genetic contribution to phenotypic variance across age was smaller for both Digit Span tests, greater for Synonyms, and stable for Block Design and Symbol Digit. The non-shared environmental contribution was greater with age for the Digit Span tests and Block Design, while the shared environmental component was small for all tests, often more so with age. Vocabulary showed similar age-moderation patterns as Synonyms, but these effects were nonsignificant. Findings are discussed in the context of theories of cognitive aging.

Introduction

Cross-sectional and longitudinal research has consistently found that average cognitive test performance declines in late life (Salthouse, 2009). Nonetheless, there are marked individual differences in the timing and rate of cognitive aging, and late-life cognitive function is relatively etiologically distinct from cognitive function at earlier ages (Wilson et al., 2002). Late-life general cognitive ability (GCA) is also moderately to strongly heritable, with minimal shared environmental contributions (Johnson, McGue, & Deary, 2014). An important but largely unaddressed question concerns whether the magnitudes of genetic and environmental contributions to late-life cognitive ability differ from those at earlier life stages.

A prominent finding from the behavioral genetic literature is that heritability of behavioral phenotypes increases with age. In a meta-analysis of relevant twin studies, Bergen, Gardner, and Kendler (2007) reported that heritability of diverse behavioral phenotypes including anxiety, externalizing psychopathology, social attitudes, and GCA all increased with age. Other research has documented age-related declines in the importance of shared environmental influences for GCA (Haworth et al., 2010). There are, however, several important limitations in this literature. First, most of the research has focused on transitions from childhood to early adulthood; much less is known about the magnitudes of genetic and environmental contributions beyond early adulthood (Tucker-Drob & Briley, 2014). Second, research on cognitive transitions from childhood to early adulthood has focused almost exclusively on GCA rather than specific cognitive abilities, despite evidence of domain specific variation in their developmental trajectories. Third, most of the research has focused on standardized, rather than raw, components of variance. Greater heritability, a standardized metric, may be a consequence of less raw environmental contribution to variance, greater genetic variance, or both.

The magnitudes of genetic and environmental contributions to late-life cognitive function might differ from those at earlier ages for several reasons. Reduction in evolutionary pressures in late life as compared to other life stages is posited to lead to amplification of stochastic (i.e., random, Finch & Kirkwood, 2000) and epigenetic processes (Fraga et al., 2005). For example, many individual-level factors (i.e., blood pressure, and physical exercise) are associated with late-life cognitive functioning but not with cognitive status at younger ages (Anstey & Christensen, 2000). The cumulative effect of these factors might be reflected by increased environmental contributions to phenotypic variance with age (c.f., Baltes, Reese, & Lipsitt, 1980). Alternatively, changes in the magnitudes of genetic contributions may reflect amplification of existing genetic factors or mechanisms of gene-environment interplay (Reynolds, Finkel, & Zavala, 2013). For instance, genetic factors that protect against environmental influences leading to cognitive decline (e.g., active developmental processes, Scarr & McCartney, 1983) can lead to greater genetic variance in late life. High educational attainment, occupational complexity, and intellectually-stimulating activities may reflect genetically influenced selections that promote cognitive reserve and prevent decline (Bosma et al., 2002).

Behavioral genetic research on cognitive abilities does not always provide consistent evidence for age differences in relative magnitudes of genetic influences. Finkel and Reynolds (2010) reviewed the behavioral genetic literature on cognitive aging and concluded that heritability of GCA appears to increase through approximately age 60 and declines thereafter. Yet, in a subsequent large cross-sectional study of 2332 Danish twins age 46 to 96 years, McGue and Christensen (2013) reported that the magnitude of genetic influence on a measure of GCA was stable across age. Unlike the differential patterns observed by independent studies, recent meta-analyses of twin studies have better convergence to the patterns observed. In a recent meta-analysis of twin studies, Reynolds and Finkel (2015) reported that the heritabilities of specific cognitive abilities including verbal, spatial and memory, were largely stable or slightly increasing with age. Similarly, a large-scale meta-analysis of all published twin studies by Polderman et al. (2015) also found consistent evidence for stable heritability across age groups across cognitive domains of clustered executive functioning and memory abilities. Although these meta-analyses seem to provide a clearer and more consistent pattern of the genetic and environmental contributions to late life, they may be also obscuring differential trajectories for specific cognitive abilities, and indeed losing important informative differences across time.

Limited sample sizes and study and country differences may contribute to the apparent inconsistency of results concerning age moderation of genetic influences. In many cases, heritability of late-life cognitive ability is estimated in samples with a few hundred twin pairs, making it difficult for a single study to distinguish heritability differences across a wide age range reliably. Moreover, studies do not always report parameter estimates for the same biometric model, making it difficult to compare estimates using meta-analytic methods. For example, the shared environmental contribution is not always reported and some reported heritability estimates are based on models dropping this component.

This study includes 14,534 participants from a twin study consortium to investigate age moderation of genetic and environmental influences on cognitive ability in mid- through late-life. The large sample, broad age range (26 to 98 years), and multiple cognitive abilities included (six tests representing four separate domains of cognitive functioning – short-term/working memory, processing speed, spatial processing, and verbal ability) make this the most comprehensive test to date of the hypothesis that the magnitudes of genetic and environmental influences on cognitive functioning differ in late-life compared to earlier life stages. In addition, the consortium this study is derived from provides a special opportunity to directly assess differential evidence found by independent studies, often from competing independent studies that are included in this consortium group, while simultaneously examining if there are informative differences across time that meta-analytic work may not have been able to observe.

Section snippets

Participants

The sample was drawn from nine studies representing three separate countries (Sweden, Denmark, and the United States) from the Interplay of Genes and Environment across Multiple Studies (IGEMS) consortium (Pedersen et al., 2013). No studies had overlapping participants. To be included in our analysis, participants had to have completed at least one of six cognitive tests (described below), and have a Mini-Mental State Examination (MMSE) score of at least 24, following the typical cutoff for

Descriptive data

Mean test scores on the T-score scale by age group are reported in Table 2 and depicted in Fig. 3. Because of winsorization, the mean and SD in the age 50–59.99 age group deviated slightly from 50 and 10, respectively. Although the age-group effects were statistically significant for all tests, the magnitudes of these effects (as indicated by η2) were large for Block Design and Symbol Digit (i.e., > 27%), moderate for Synonyms (i.e., 7.7%), and small for Digits Forward, Digits Backward, and

Discussion

Cross-sectional analyses of six specific measures of cognitive function in a combined sample of 14,534 twins aged 26 to 98 years revealed age differences in the magnitudes of genetic and environmental contributions to phenotypic variance that varied across test. Before discussing these findings, we acknowledge the limitations of our study. First, pooling cognitive measures across nine twin samples may obscure important between-study differences. Nonetheless, there was limited statistical

Author contributions

All authors helped to conceptualize the project and design the data analysis through conference calls. All authors read, commented, edited multiple drafts of the paper, and approved the final version. SP, NRH and MM took primary responsibility for undertaking the analyses; SP and MM took primary responsibility for writing the initial draft of the paper. Data were contributed by IP & KC (MADT and LSADT); CEF and WSK (VETSA); RFK (MIDUS); JMN (TOSS); NLP, ADA, CAR, DF, and MG (SATSA, OCTO-Twin

Acknowledgement

IGEMS is supported by the National Institutes of Health grant nos. R01 AG037985 and R56 AG037985. SATSA was supported by grants R01 AG04563, R01 AG10175, the MacArthur Foundation Research Network on Successful Aging, the Swedish Council For Working Life and Social Research (FAS) (97:0147:1B, 2009-0795) and Swedish Research Council (825-2007-7460, 825-2009-6141). OCTO-Twin was supported by grant R01 AG08861. Gender was supported by the MacArthur Foundation Research Network on Successful Aging,

References (39)

  • M. McGue et al.

    Growing old but not growing apart: Twin similarity in the latter half of the lifespan

    Behavior Genetics

    (2013)
  • T.A. Salthouse

    When does age-related cognitive decline begin?

    Neurobiology of Aging

    (2009)
  • K. Anstey et al.

    Education, activity, health, blood pressure and apolipoprotein E as predictors of cognitive change in old age: A review

    Gerontology

    (2000)
  • P.B. Baltes et al.

    Life-span developmental psychology

    Annual Review of Psychology

    (1980)
  • S.E. Bergen et al.

    Age-related changes in heritability of behavioral phenotypes over adolescence and young adulthood: A meta-analysis

    Twin Research and Human Genetics

    (2007)
  • K.L. Bopp et al.

    Aging and verbal memory span: A meta-analysis

    The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences

    (2005)
  • H. Bosma et al.

    Engaged lifestyle and cognitive function in middle and old-aged, non-demented persons: A reciprocal association?

    Zeitschrift Fur Gerontologie Und Geriatrie

    (2002)
  • D.A. Briley et al.

    Explaining the increasing heritability of cognitive ability across development a meta-analysis of longitudinal twin and adoption studies

    Psychological Science

    (2013)
  • K. Christensen et al.

    A Danish population-based twin study on general health in the elderly

    Journal of Aging and Health

    (1999)
  • C.E. Finch et al.

    Chance, development, and aging

    (2000)
  • D. Finkel et al.

    The origins of individual differences in memory among the elderly: A behavior genetic

    Psychology and Aging

    (1993)
  • D. Finkel et al.

    Behavioral genetic investigations of cognitive aging

  • M.F. Fraga et al.

    Epigenetic differences arise during the lifetime of monozygotic twins

    Proceedings of the National Academy of Sciences of the United States of America

    (2005)
  • C.H. Gold et al.

    Gender and health: A study of older unlike-sex twins

    The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences

    (2002)
  • C.M.A. Haworth et al.

    The heritability of general cognitive ability increases linearly from childhood to young adulthood

    Molecular Psychiatry

    (2010)
  • W.J. Hoyer et al.

    Adult age and digit symbol substitution performance: A meta-analysis

    Psychology and Aging

    (2004)
  • W. Johnson et al.

    Normative cognitive aging

  • K.S. Kendler et al.

    Sexual orientation in a US national sample of twin and nontwin sibling pairs

    American Journal of Psychiatry

    (2000)
  • W.S. Kremen et al.

    VETSA: The Vietnam era twin study of aging

    Twin Research and Human Genetics

    (2013)
  • Cited by (12)

    View all citing articles on Scopus
    View full text