Original Article
Symmetric faces are a sign of successful cognitive aging

https://doi.org/10.1016/j.evolhumbehav.2009.06.001Get rights and content

Abstract

It has been proposed that a common cause underlies individual differences in bodily and cognitive decline in old age. No good marker for this common cause has been identified to date. Here, fluctuating asymmetry (FA), an indicator of developmental stability that relates to intelligence differences in young adults, was measured from facial photographs of 216 surviving members of the Lothian Birth Cohort 1921 at age 83 and related to their intelligence at ages 11, 79 and 83 years. FA at age 83 was unrelated to intelligence at ages 11 and 79 and to cognitive change between 11 and 79 years. It was, however, associated with intelligence and information processing efficiency at age 83 and with cognitive change between 79 and 83 years. Significant results were limited to men, a result predicted by sex differences in life history tradeoffs and life expectancy. Results were stronger when directional asymmetries were corrected in facial FA measures. Thus, FA is a candidate marker for the common cause of differential senescence.

Introduction

The threat of cognitive decline is the most feared aspect of human aging for many people, but the reasons why some people show more cognitive decline than others remain largely unknown. The observation that cognitive functions often decline in concert both with each other (Wilson et al., 2002) and with sensory, organic, and muscular functionality has made the “common cause” hypothesis an influential account for individual differences in cognitive aging (Christensen et al., 2001, Lindenberger and Baltes, 1994). This hypothesis suggests that cognitive decline is in part an aspect of body-wide senescence in what we shall call system integrity. It is assumed that no single one mechanism underlies declining system integrity. Instead, the proximate causes of this decline may be accumulated environmental stresses, recurrent failures to adapt to stressors (i.e., allostatic load), diminishing cellular repair efficiency, and genetic and epigenetic chance events (Batty, Deary, & Gottfredson, 2007; Finch and Kirkwood, 2000, McEwen, 2006, Seeman et al., 2001). The loss of system integrity with senescence can be considered to be the result of a life-history trade-off in which reduced investment in somatic maintenance and repair (or “own embodied capital”) frees more energy for reproductive effort (Kaplan et al., 2003, Kirkwood, 2008).

A ready marker summarizing all these small, diverse, detrimental processes that impede system integrity in old age would be invaluable, and since the formulation of the common cause hypothesis, several such markers have been proposed. For example, the length of telomeres (nucleo-protein complexes that protect the ends of chromosomes and that shrink with every cell division), was thought to be a promising candidate, but supportive empirical evidence is limited so far (Harris et al., 2006). Alternative markers such as mitochondrial DNA integrity remain active research targets (Krishnan, Greaves, Reeve, & Turnbull, 2007).

One marker of high system integrity, including successful cognitive aging, might be facial symmetry. This is suggested by a very similar common cause hypothesis that has been proposed to explain individual differences in general cognitive ability (i.e., intelligence) at any age (Furlow et al., 1997, Penke et al., 2007). According to this hypothesis, developmental instability (DI; Polak, 2003) is the latent causal factor that links neurophysiological and bodily condition. DI results from imprecision in the expression of developmental design due to imperfect buffering against genetic and environmental perturbations. The most established indicator for DI is fluctuating asymmetry (FA), the degree to which the size of bilateral body parts deviates from the population mean, usually aggregated across several traits (Palmer & Strobeck, 2003). Research on predominantly young people has revealed significant associations between higher general intelligence and lower FA in nine samples (Bates, 2007, Furlow et al., 1997Luxen and Buunk, 2006, Thoma et al., 2006, Rahman et al., 2004Thoma et al., 2005, Thoma et al., 2006), while the one study reporting a null finding used middle-aged adults (Johnson, Segal, & Bouchard, 2007) (weighted average r=−.20 across all published samples; range .00 to −.51). Development, however, is a lifelong process (Baltes, Lindenberger, & Staudinger, 2006), and developmental stability needs to be maintained not only during early growth but also during adulthood and late in life (though the mechanisms that support developmental stability might or might not be the same in different life phases, Klingenberg, 2003, Finch and Kirkwood, 2000). Thus, DI might provide a theoretical framework for the post hoc and merely descriptive common cause hypothesis of cognitive aging, since the underlying factor might be one and the same. If this is true, FA should be a good candidate marker for this putative “common cause.”

Applied to cognitive aging, the DI model predicts that long-term accumulation of stresses and decreasing efficiency of maintenance and repair processes in the body (Finch & Kirkwood, 2000) will result in increased DI and FA in the elderly. This has been demonstrated by Kobyliansky and Livshits (1989) in cross-sectional comparisons within a sample of 2213 Israelis. They found significantly higher levels of FA in individuals beyond age 80 years than in any other age group from birth to age 45.

Miller (2000) argued that human intellectual abilities evolved as a fitness indicator. Sexual selection predicts that males, who show higher variance in reproductive success than females, receive greater average fitness benefits from investing in fitness indicators (Andersson, 1994). Since fitness indicators are assumed to be dependent on phenotypic condition (Tomkins, Radwan, Kotiaho, & Tregenza, 2004), the link between DI and cognitive functioning should be more pronounced in men than in women. Guided by this logic, three studies that tested this relationship (Prokosch et al., 2005, Thoma et al., 2005, Thoma et al., 2006) chose to limit their samples to male subjects. Empirical evidence for this sex difference is, however, mixed: whereas stronger links between FA and intelligence in men than in women were found in three samples (Bates, 2007, Study 1; Raham et al., 2004; marginal in Luxen & Buunk, 2006), Johnson et al. (2007) and the two studies in Furlow et al. (1997) found no difference, and Bates (2007, study 2) found the opposite sex difference.

However, greater male investment in reproductive effort (including fitness indicators) is assumed to come at the cost of lower investment in somatic maintenance, and these costs should become especially obvious when they have accumulated in old age. Indeed, this life-history trade-off has been used to explain the shorter life expectancy of males compared to females (Finch, 2007), which is still about 4 years in Western countries, including the current study populations (Trovato & Heyen, 2006; www.mortality.org). Thus, it can be expected that the relationship between cognitive ability and DI will be especially strong in elderly men. Note that this sex-specific prediction can be derived from a combination of life history theory and parental investment theory regardless of whether Miller's (2000) hypothesis of intelligence as a fitness indicator is correct or not. However, if intelligence was indeed a condition-dependent fitness indicator, the predicted sex difference in the FA-intelligence link in old age should be especially pronounced.

Here, for the first time, we examine FA in relation to cognitive functioning and cognitive decline in old age. The sample used is a narrow age cohort tested on cognitive ability at ages 11, 79 and 83 years. Assessments included both standard cognitive ability tests and reaction time (RT) measures. RT tasks are considered to be indicators of the efficiency of information processing and are especially prone to cognitive aging (Der and Deary, 2006, Jensen, 2006). They are also candidate markers for the common cause of cognitive aging (Deary & Der, 2005). We predicted that increased FA (and thus increased DI) would be associated with cognitive decline and RT deterioration from age 79 to 83 years, especially in men.

Section snippets

Subjects

The sample consisted of surviving participants of the Scottish Mental Survey 1932 (SMS1932) (Scottish Council for Research in Education, 1933), all born in 1921. They took part, as members of the Lothian Birth Cohort 1921 (LBC1921), in a follow-up study between 1999 and 2001 at a mean age of 79 years (Deary, Whiteman, Starr, Whalley, & Fox, 2004). At that time, all were generally healthy and living independently in the community. Further cognitive testing took place at a mean age of 83 years (

Results

After excluding participants with unsuitable photos or signs of dementia, 216 subjects (95 men, 121 women) remained in the main LBC1921 sample for the following analyses.

The raw frequency distributions of the CFA facial asymmetry scores of both the elderly main sample and the younger comparison sample are shown in Fig. 2. As expected, CFA showed both a significantly higher mean [t(364)=9.68, p<.001, d=1.01] and increased variance [F(1, 364)=12.39, p<.001] in the older sample. The mean

Discussion

In this first-ever study on developmental instability in relation to intelligence and cognitive aging in the elderly, we found that, in men, greater facial FA was significantly associated with cognitive decline between 79 and 83 years, as well as with slower and more variable reaction times that were concurrently assessed at age 83, and marginally with concurrently assessed general intelligence. There were no significant relationships between FA and earlier-age intelligence or midlife cognitive

Acknowledgments

This study was funded by a British Academy small research grant awarded to IJD and DIP. The young comparison sample was part of a project supported by grant As 59/15 of the German Research Foundation (DFG). LP is supported by a grant from the UK Medical Research Council Grant (No. 82800), which is part of the Help The Aged/Research into Ageing-funded Disconnected Mind project. This work was undertaken within The University of Edinburgh Centre for Cognitive Ageing and Cognitive Epidemiology,

References (57)

  • ThornhillR. et al.

    Facial sexual dimorphism, developmental stability, and susceptibility to disease in men and women

    Evolution and Human Behavior

    (2006)
  • TomkinsJ.L. et al.

    Genic capture and resolving the lek paradox

    Trends in Ecology and Evolution

    (2004)
  • WhalleyL.J. et al.

    Cognitive reserve and the neurobiology of cognitive aging

    Ageing Research Reviews

    (2004)
  • ZehJ.A. et al.

    Maternal inheritance, sexual conflict and the maladapted male

    Trends in Genetics

    (2005)
  • AnderssonM.

    Sexual selection

    (1994)
  • BaltesP.B. et al.

    Life span theory in developmental psychology

  • BrownW.M. et al.

    Fluctuating asymmetry and preferences for sex-typical bodily characteristics

    Proceedings of the National Academy of Sciences USA

    (2008)
  • ChristensenH. et al.

    The common cause hypothesis of cognitive aging

    Psychology and Aging

    (2001)
  • DearyI.J. et al.

    Reaction time explains IQ's association with death

    Psychological Science

    (2005)
  • DearyI.J. et al.

    Physical fitness and lifetime cognitive change

    Neurology

    (2006)
  • DearyI.J. et al.

    The impact of childhood intelligence on later life: following up the Scottish Mental Surveys of 1932 and 1947

    Journal of Personality and Social Psychology

    (2004)
  • DerG. et al.

    Reaction time age changes and sex differences in adulthood. Results from a large, population based study: the UK Health and Lifestyle Survey

    Psychology and Aging

    (2006)
  • FinchC.E.

    The biology of human longevity

    (2007)
  • FinchC.E. et al.

    Chance, development, and aging

    (2000)
  • FurlowF.B. et al.

    Fluctuating asymmetry and psychometric intelligence

    Proceedings of the Royal Society London B

    (1997)
  • GangestadS.W. et al.

    A latent variable model of developmental instability in relation to men's sexual behaviour

    Proceedings of the Royal Society of London B

    (2001)
  • GowA.J. et al.

    Mental ability in childhood and cognitive aging

    Gerontology

    (2008)
  • GrammerK. et al.

    Human (homo sapiens) facial attractiveness and sexual selection: The role of symmetry and averageness

    Journal of Comparative Psychology

    (1994)
  • Cited by (43)

    • Emergence of a left-right symmetric body plan in vertebrate embryos

      2024, Current Topics in Developmental Biology
    • The association of three indicators of developmental instability with mating success in humans

      2017, Evolution and Human Behavior
      Citation Excerpt :

      Alternatively, one could argue that an FAbody aggregate of twelve traits is still not sufficient and that the number of traits needs to be increased. For example, Penke et al. (2009) used an aggregate facial FA consisting of 35 traits and found associations with elderly men's cognitive decline. Even higher aggregated, spatially dense FA indicators have been assessed from 3D face scans using geometric morphometrics (Claes, Walters, Vandermeulen, & Clement, 2011; Hill et al., 2017).

    • General cognitive ability and fluctuating asymmetry of brain surface area

      2016, Intelligence
      Citation Excerpt :

      In our recent work, for example, we have emphasized the possible importance of global network features such as efficiency, connectivity, and characteristic path length (Yeo et al., 2016), and also neurophysiological responses to novel stimuli (Euler, Weisend, Jung, Thoma, & Yeo, 2015). At a more cognitive level of analysis, these global brain features may confer greater processing speed and complex reaction time, which in turn contributes greatly to GCA (Hope et al., 2015; Penke et al., 2009; Thoma et al., 2006). However, the current results most clearly draw attention to frontal and parietal functioning, consistent with growing evidence from structural and functional neuroimaging (e.g., Basten et al., 2015) for the P-FIT model of intelligence (Jung & Haier, 2007).

    • Bodily symmetry increases across human childhood

      2013, Early Human Development
      Citation Excerpt :

      The study of symmetry in childhood may, therefore, shed light on the origins of symmetry, and of the traits with which it is correlated such as intelligence [5–7]. It may also aid understanding of late-life processes of physical and cognitive decline [8]. Animal research has shown that in swordfish the point at which an organism begins to shift resources from growth to maintaining symmetry varies according to genotype [9].

    View all citing articles on Scopus
    View full text