Genetic influence on contrast sensitivity in young adults

Editor,

Contrast sensitivity, the ability to detect small luminance differences, is an important independent aspect of visual function that can vary more than fourfold across normal individuals (Baker 2013). Little is known about the degree to which this variation is determined by genetic and environmental factors. The only study to date was based on a sample of male military veterans with an age range of 52–60, and estimated the heritability of contrast sensitivity (i.e., the portion of phenotypic variance accounted for by genetic factors) between 14% and 38%; much lower than might be expected of a core, biologically‐based visual function (Cronin‐Golomb et al. 2007). However, due to the relatively homogeneous sample of middle‐aged men, it is unclear whether these estimates reflect environmental influences on development or the rate of age‐related decline, which normally begins at age 40–50 for higher spatial frequencies (Owsley et al. 1983). Here, therefore, I estimated the genetic influence on contrast sensitivity in a sample of healthy young adults of both sexes between 22 and 36 years of age, who can be considered to represent the population at large with respect to ethnic and socio‐economic diversity, and whose visual contrast sensitivity should be fully developed but not yet aged.

The sample contained 149 monozygotic (MZ) and 94 dizygotic (DZ) twin pairs of the WU‐Minn Human Connectome Project (Van Essen et al. 2013) whose twin‐status was confirmed by genetic testing. The mean (SD) age was 29.3 (3.4) and the sample included 295 females (174 MZ, 121 DZ). Inclusion and exclusion criteria are detailed elsewhere (Van Essen et al. 2013). Contrast sensitivity was assessed binocularly using the Mars Letter Contrast Sensitivity test (Arditi 2005) and visual acuity using the Electronic Visual Acuity (EVA) system running the eETDR protocol (Beck et al. 2003). Both tests involved corrected vision, if applicable. Contrast sensitivity was calculated as the log contrast sensitivity value at the final correct letter minus the number of errors prior to the final correct letter times 0.04. The EVA scores were converted to logMAR using the formula –log₁₀(A), where A is the ratio between the EVA nominator and denominator scores.

The heritability of contrast sensitivity, $h^2={\sigma }_g^2/{\sigma }_p^2$, was estimated using SOLAR (Almasy & Blangero 1998) with covariates visual acuity, age and sex. SOLAR estimates the genetic and environmental variances, ${\sigma }_g^2$ and ${\sigma }_e^2$, by comparing the observed phenotypic covariance matrix with the covariance matrix predicted by genetic relatedness (i.e., $\mathrm{\Omega }=2\mathrm{\Phi }{\sigma }_g^2+I{\sigma }_e^2$, where Φ encodes the pair‐wise genetic relatedness among all individuals) and determines the statistical significance of the heritability estimates by a chi‐squared test comparing the log‐likelihood of the model in which ${\sigma }_g^2$ is constrained to be zero to that in which ${\sigma }_g^2$ is estimated. Continuous traits were normalized using the inverse normal transformation.

The mean (SD) log contrast sensitivity across all twin pairs was 1.80 (0.06) and the mean (SD) visual acuity was −0.14 (0.12) logMAR. There were no significant differences between the monozygotic and dizygotic groups in contrast sensitivity, visual acuity, age (all p > 0.58), or the proportion of males/females (χ ² = 1.72, p = 0.19). Contrast sensitivity was moderately heritable, with additive genetic effects explaining 27% of the phenotypic variance (h ² = 0.27, SE = 0.07, p =1.2 × 10⁻⁴), which is consistent with the prior estimates for peak contrast sensitivity in middle‐aged men.

The comparably moderate heritability of contrast sensitivity in early and middle adulthood suggests a strong influence of nongenetic, nonageing‐related factors. While these influences may partly reflect measurement error, variations in cognitive ability and/or task engagement, a large proportion likely involves individual‐specific environmental experiences during childhood and adolescence (Cronin‐Golomb et al. 2007; Baker 2013; Bartholomew et al. 2016). Identifying these experiences is an important direction of future research, as they may be altered to improve visual function in adulthood.