Despite the large amount of variation found in the night (scotopic) vision capabilities of healthy volunteers, little effort has been made to characterize this variation and factors, genetic and non-genetic, that influence it. In the largest population of healthy observers measured for scotopic visual acuity (VA) and contrast sensitivity (CS) to date, we quantified the effect of a range of variables on visual performance. We found that young volunteers with excellent photopic vision exhibit great variation in their scotopic VA and CS, and this variation is reliable from one testing session to the next. We additionally identified that factors such as Circadian preference, iris color, astigmatism, depression, sex and education have no significant impact on scotopic visual function. We confirmed previous work showing that the amount of time spent on the vision test influences performance and that laser eye surgery results in worse scotopic vision. We also showed a significant effect of intelligence and photopic visual performance on scotopic VA and CS, but all of these variables collectively explain <30% of the variation in scotopic vision. The wide variation seen in young healthy volunteers with excellent photopic vision, the high test-retest agreement, and the vast majority of the variation in scotopic vision remaining unexplained by obvious non-genetic factors suggests a strong genetic component. Our preliminary genome-wide association study (GWAS) of 106 participants ruled out any common genetic variants of very large effect and paves the way for future, larger genetic studies of scotopic vision.
Citation: Bartholomew AJ, Lad EM, Cao D, Bach M, Cirulli ET (2016) Individual Differences in Scotopic Visual Acuity and Contrast Sensitivity: Genetic and Non-Genetic Influences. PLoS ONE 11(2): e0148192. https://doi.org/10.1371/journal.pone.0148192
Editor: Olaf Strauß, Eye Hospital, Charité, GERMANY
Received: August 28, 2015; Accepted: January 14, 2016; Published: February 17, 2016
Copyright: © 2016 Bartholomew et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All of the data used in this study are included in the supplement, except for the raw genetic data from the participants, which the authors do not have IRB approval to make publicly available. The genetic data will be made available to researchers who contact email@example.com and meet the criteria for access to confidential data. Additionally, the results of the genetic study are included in the supplement.
Funding: Research reported in this publication was supported by the National Institute of Mental Health of the National Institutes of Health under award number K01MH098126. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The visual system operates over a remarkable range of lighting conditions through transduction by two classes of photoreceptor cells, rods and cones . In photopic conditions, three types of cone photoreceptors with overlapping spectral sensitivities produce spatial acuity and color vision [2, 3]. In scotopic conditions, visual performance relies on rod photoreceptors, resulting in a reduction of spatial resolution of approximately 1200:1 in exchange for increased light detection sensitivity . During mesopic vision, an intermediate illumination level between photopic and scotopic conditions, rod and cone pathways operate simultaneously to contribute to vision .
Visual acuity (VA) and contrast sensitivity (CS) are important measures of visual function, though the latter—while being a better predictor of traffic incidents –is often neglected in clinical testing. VA describes spatial resolution under high contrast conditions, while CS describes the ability to distinguish small differences in luminance [7–9]. Visual performance can be differentially affected in differing luminance conditions, with performance in lower luminance conditions frequently being more sensitive to ocular dysfunction [8, 9]. Strong visual performance in mesopic or scotopic conditions tends to predict strong photopic vision, but the reverse does not necessarily hold true .
The majority of scotopic vision research to date has centered on diseased populations and treatment efficacy from refractive procedures such as laser eye surgery and intraocular lens transplantation . In addition, it is known that scotopic vision can be adversely affected by a lack of essential nutrients such as Vitamin A and zinc deficiency [11, 12]. In healthy individuals, studies have preliminarily addressed the effects of age, pupil size, and astigmatism on differences in scotopic visual abilities [13–16]. However, no effort has yet been made to describe the individual differences in dark adaptation or scotopic visual function of healthy observers or to characterize the factors that influence these differences [9, 14, 17]. Differences between normal observers may result from rod density, differential convergence of rod signals, extent of activation of distinct pathways, functional differences in proteins such as rhodopsin, or other post-receptoral mechanisms [18, 19].
The contribution of a genetic component to healthy scotopic vision has yet to be explored, even though obvious heritable differences in photopic visual abilities exist [20, 21]. A host of genetic disorders, such as congenital stationary night blindness (CSNB) and retinitis pigmentosa (RP), are characterized by a compromise in rod function, resulting in night blindness [22–26]. The existence of Mendelian diseases affecting scotopic vision provides further evidence for a genetic influence on scotopic visual abilities [27, 28].
In this study, we first characterize normal scotopic VA and CS performance in a large population of healthy observers following 20 minutes of dark adaptation. We then define the contributions of a wide range of variables to scotopic performance. After controlling for confounding factors, we perform a preliminary genetic analysis using scotopic VA and CS as our outcomes. Our overall goal is to identify genetic variants with an influence on scotopic vision in healthy observers, leading to a better understanding of the complex molecular mechanisms involved in scotopic vision.
Materials and Methods
Participants and Ethics
The Duke University Institutional Review Board approved all procedures, and participants provided written, informed consent (IRB #: Pro00006828).
A total of 734 participants attempted the visual tasks as part of a larger battery in the Duke Genetics of Cognition and Other Normal Variation study [29–31]. Due to the complex protocol for night vision testing, complete and useable visual data was available for 664 individuals due to participant, examiner, or device error. Further exclusionary criteria required to define a healthy population of normal observers resulted in a final sample size of 501 (criteria described below). A comprehensive description of these 501 participants can be seen in Table 1.
Prior to psychometric and visual testing, participants completed an extensive survey that queried demographics, medical history, and several standardized scales. To identify new variables associated with scotopic vision, we investigated multiple measures for association with this phenotype. Many of these variables, such as years of education and ethnicity, have never before been compared to the scotopic visual performance of a person.
A total of 451 participants completed the Beck Depression Inventory-II (BDI) . Participants scoring 14 or higher were categorized as depressed.
The Circadian Energy Scale (CIRENS) is a two-question chronotype measure based on self-report energy levels throughout the day: once at night and once in the morning. Energy levels are described on a Likert scale: [very low (1), low (2), moderate (3), high (4), or very high (5)]. The difference between the evening score and morning score determines the overall chronotype score, ranging from -4 (most marked morning preference) to +4 (most marked evening preference) . Previous studies have shown changes in night vision phenotypes during different times of day and parts of the Circadian cycle [34, 35].
Prior to visual tasks, participants indicated their current level of tiredness on a Likert scale: [energetic (0) to very tired (4)].
Participants self-reported eye health and characteristics. For iris color, participants were divided into one of three groups: a baseline group of participants indicating brown or black irises, a second group with green or hazel irises, and a third group with blue or gray irises. Previous studies have investigated the impact of iris color on certain aspects of ocular function but have not investigated whether it is associated with scotopic vision [36–38]. Participants also self-reported whether they were previously diagnosed with astigmatism in the eye tested or had a color vision deficiency, and provided a history of ocular disease or interventional procedures. No further assessment of these traits was performed beyond the self-report.
Participants completed a cognitive battery assessing diverse areas of cognition as previously described . Principal component analysis was performed on the eleven individual test scores to determine an overall measure of performance . The first principal component explained 37.3% of the total variance in test scores and received approximately equal loadings from all tests. It was therefore taken as a measure of overall cognitive performance on the battery and can be considered a proxy for general intelligence.
Freiburg Visual Acuity and Contrast Test (FrACT)
VA and CS thresholds were assessed using the Freiburg Visual Acuity and Contrast Test (FrACT; Version 3.7.4.c) . FrACT employs a best parameter estimation by sequential testing (best PEST) algorithm to adapt optotypes in real time to the user. FrACT displays Landolt Cs over a large range of VA and CS in an eight alternative forced choice task . FrACT has been utilized in over 400 publications and is available free of charge at michaelbach.de/fract.html. To assess VA, the size of a high-contrast Landolt-C is varied. The size of the gap at threshold, measured in minutes of arc, is taken as the minimum angle of resolution (MAR); its logarithm (“logMAR”) is a standard measure of visual acuity . To assess CS, the contrast of a large (3.3° diameter) Landolt-C is varied, while its size remains constant. The contrast at threshold is initially expressed as fractional Weber contrast CW (CW = luminance difference divided by the surround luminance); its value is typically around 0.01 for photopic vision and 20 times higher for scotopic vision. [For isolated optotypes on a large background the Weber contrast is more relevant than the Michelson contrast.] This contrast threshold is converted to log contrast sensitivity (logCSWeber = log(1/CW)).
Participants first completed both the FrACT VA and CS tasks twice in photopic conditions. The tests were given twice to allow assessment of test-retest agreement and to improve our ability to identify outliers. Participants then dark adapted for 20 minutes and then completed the FrACT VA and CS an additional two times. Participants were required to finish both repeats in five minutes to allow for a standard start time for the next visual task and to standardize the dark adaptation length for all participants; if they did not complete both, then only one measure was used.
The same computer was used for both photopic and scotopic tests. The results were obtained with a 17” monitor at a distance of 1.54 m. The diameter of the Landolt C was set to 200 minutes of arc (arcmin) at a constant optotype contrast of 100%. For high reliability, a total of 24 trials for each task were used . The monitor luminance in photopic conditions was 90 cd/m², and the ambient lighting was 300 lux. Luminance was measured using the i1Display 2 (X-Rite, Grand Rapids, MI). For scotopic conditions, the ambient lights were extinguished, and great care was taken to eliminate all potential sources of light from the dark room. Neutral density filters (Paul C. Buff, Inc., Nashville, US) were placed over the computer monitor, reducing the monitor luminance to 0.00092 cd/m².
Results for all visual tasks are monocular and reflect performance in the participant’s right eye (n = 490) unless a participant self-reported worse vision compared to the left eye. The amount of time spent per task was recorded for FrACT in ms and log-transformed for analysis; referred to as ‘task duration’. We did not use a fixation task but attempted to minimize variation in testing strategies by informing participants to not continuously fixate on the stimulus. Auditory feedback was provided to indicate correct or incorrect.
Absolute detection threshold.
Following FrACT completion that accounted for 25 minutes of dark adaptation, 248 participants completed the absolute detection threshold measurement using a LED-based dark adaptometer built in Dr. Cao’s laboratory. The test size was 4° in visual angle and included a 10° temporal eccentric fixation. The stimuli were 100ms pulse square waves with a dark background (i. e. 0 cd/m2); output was digitally controlled by an M-Audio PCI soundcard . Two independent thresholds were generated using a two-yes/one-no staircase procedure that varied the peak luminance of the square wave until the threshold was determined. The smaller of the two thresholds served as the dark adaptation (DA) phenotype.
In addition to the exclusionary criteria (described below) to define a healthy population, 14 participants were also excluded from DA analyses because their threshold was more than 3 SD above the group mean, indicating greatly impaired scotopic vision. Our final DA sample size was 234 participants.
To evaluate the reliability of our tasks, 40 participants completed both the FrACT and DA tasks at two separate testing sessions, a mean of 85 days apart (SD = 18). We were also able to evaluate test-retest agreement within the same testing session as most participants completed two photopic and scotopic scores during the initial testing session (Fig 1).
Four data sets are depicted: Visual acuity (left panel) and contrast sensitivity (right panel) at photopic luminance (green triangles, near top left and at scotopic luminance (blue discs, near bottom left). Result of the first test on the abscissa, second test on the ordinate. Grey 45°-line is the identity line, next to it the ± limits of agreement (photopic, dashed; scotopic, dotted). Visual acuity in logMAR units have an inverted scale, and contrast sensitivity is in logCSWeber units, meaning that better performance corresponds to the top right for both graphs. As expected, photopic measures of VA or CS are markedly better than scotopic ones. The 95% limits of agreement are remarkably similar. All in all, there is no marked deviation from a normal distribution, and the reliability is good for the range measured.
The primary outcomes were FrACT phenotypes, which were generated based on the average of two independent thresholds for photopic conditions and for scotopic conditions after twenty minutes of dark adaptation. When only one photopic or scotopic score was recorded (n = 102), this measure was used as the phenotype.
For the test-retest agreement, we calculated the 95% limits of agreement .
Useable data were generated for 664 participants who were judged by our psychometricians to fully comprehend the task and have no technological or methodological barriers to performance. To define a healthy population of observers, those with photopic VA worse than 20/25 (corresponding to ≥ 0.1 logMAR), who were older than age 40, who did not recognize any optotypes after 20 minutes of dark adaption, or who were more than 3 SD beyond the mean scotopic VA performance were excluded from all analyses.
In addition, those with previous laser eye surgery were excluded from genetic analyses because this procedure had significant impact on scotopic performance in our task, concordant with the pertinent literature and potentially due to dry eyes [44–46].
All non-genetic statistical analyses were performed using STATA 13.1 . Stepwise forward linear regression analyses with a cutoff for inclusion of p < 0.01 were performed for the VA and CS scotopic phenotypes, with the phenotype as the outcome and variables listed in Table 2 as covariates. Scotopic VA population passed Shapiro-Wilks (p > 0.001) for normal distribution. The scotopic CS population did not pass Shapiro-Wilks (p < 0.001) due to a tail of low performers. Scores were normally distributed when restricting to those who performed no more than 2 SD below the mean (n = 25). This restriction had no significant impact on the predicted stepwise model, so we retained these 25 individuals in the analysis.
Two variables, BDI and iris color, were unavailable for some participants and were subsequently excluded from the stepwise analyses to maintain the largest sample size possible. A total of 50 individuals were missing BDI score, and 146 individuals did not have iris color responses. These variables were not found to be significantly associated with the phenotypes in univariate analyses.
Variables that were significant in the stepwise model were used as covariates in subsequent genetic analyses and in the multivariate analyses results presented in Table 2.
A genome-wide association study (GWAS) was performed on 106 participants who had Illumina Humanexome chip data available. Of these, most were also genotyped with the Infinium HumanCore GWAS chip (n = 93), and others were genotyped with either the Human610-Quad BeadChip (n = 6) or HumanHap550 (n = 5). Two of these 106 samples did not have GWAS genotypes. Variants from any of these chips were included in the analysis provided they passed QC and met the below inclusion criteria.
Our single variant analysis restricted to variants genotyped in at least 50% of these participants with a minor allele frequency (MAF) of at least 0.01 A linear regression was used in PLINK  for scotopic VA and CS performance (n = 106). Two EIGENSTRAT axes  and significant variables from the stepwise models were used as covariates. A total of 273,230 variants were analyzed in this GWAS . Correction for multiple tests therefore required a p-value of 1.8x10-7 to reach significance. We also performed a focused analysis on 94 candidate genes annotated as being involved in the phototransduction and retinol metabolism pathways [50, 51] as well as 55 candidate genes implicated in Mendelian diseases causing night vision defects (http://omim.org/).
To assess the effects of the low frequency variants genotyped with the exome chip, we used a gene-based collapsing analysis as previously described . Briefly, we summarized for each participant whether there existed a ‘qualifying’ variant in each gene, where qualifying was defined as an exonic variant with MAF < 0.01. Linear regression analysis was then performed with two EIGENSTRAT axes, the task duration, and the photopic score, and the first principal component of the cognitive battery as covariates. This allows the identification of genes where qualifying variants are enriched in individuals toward one extreme or the other of each trait. Of the candidate genes described above, 32 had low frequency coding variants included in this gene-based collapsing analysis.
FrACT test-retest agreement
Test-retest agreement of FrACT photopic and scotopic VA and CS were evaluated from 399 participants who completed the tasks twice at the initial testing session (Fig 1) and an additional 40 participants who repeated the tasks at a later session (S1 Fig). As expected, photopic measures of VA or CS were superior to scotopic ones. The 95% limits of agreement for all conditions at the first testing session are remarkably similar. For VA, photopic: 0.082 logMAR, scotopic: 0.115 logMAR, thus corresponding to ±1 line on an acuity chart. For CS, photopic: 0.127 logCSWeber, scotopic: 0.147 logCSWeber. Performance between the two testing sessions was also highly consistent: the 95% limits of agreement are 0.089 for photopic, 0.124 logMAR for scotopic VA, 0.165 for photopic, and 0.242 logCS for scotopic CS.
Photopic FrACT VA and CS performance
Photopic VA scores exhibited a ceiling effect at -0.182 logMAR (mean = -0.105, SD = 0.080) with 30.7% of the final population arriving at this threshold. This ceiling effect was due to the pixel resolution of the computer monitor at the testing distance used. VA performance in photopic conditions was significantly correlated with VA performance in scotopic conditions (p < 0.001) but explained only 4.1% of the variance (Fig 2). Photopic CS scores ranged from 1.66 to 2.60 logCSWeber (mean = 2.13, SD = 0.110); performance was significantly correlated with scotopic CS (p < 0.001) but explained only 2.5% of the variance (Fig 3).
Visual acuity in logMAR units have an inverted scale, meaning that better performance is shown here with a higher score. Photopic VA explained 4.1% of the variance in scotopic VA.
Contrast sensitivity is in logCSWeber units, meaning that better performance is a higher score. Photopic CS explained 2.5% of the variance in scotopic CS.
Individual performance on photopic VA and CS was significantly correlated (p < 0.001), and scores on one explain 9.8% of the variance in the other.
Scotopic FrACT performance and absolute detection threshold
Scotopic VA performance was significantly correlated with scotopic CS performance (p < 0.001; Fig 4) and explained 67.1% of the variance. Scotopic VA (mean = 1.21, SD = 0.159) ranged from 0.79 to 1.76 logMAR and scotopic CS (mean = 0.711, SD = 0.197) ranged from 0.08 to 1.22 logCSWeber. We additionally found that scotopic VA and CS performance was significantly correlated with the absolute detection threshold on the dark adaptometer (VA: r = 0.31, p < 0.001; CS: r = -0.30, p < 0.001).
Multivariate stepwise regressions: scotopic FrACT VA and CS
Stepwise regression analyses assessed the contribution of a number of demographic variables, eye characteristics, photopic visual performance, task duration, time of day, and intelligence to scotopic VA and CS (Table 2). Variables meeting inclusion criteria for the VA and CS models explained a total of 32.7% and 32.9% of the variance, respectively. For both models, the strongest predictor of scotopic performance was task duration (p < 0.0001), accounting for a total of 27.0% of the variance in VA and 26.5% in CS performance. Three additional variables passed inclusion criteria for both models (p VA; p CS): photopic VA or CS (<0.0001; <0.0001), intelligence (0.0014; <0.0001), and previous laser eye surgery (0.0009; 0.0033).
After correcting for multiple tests, we identified no variants or genes with statistically significant associations with these traits (p < 1.8x10-7). Our genome-wide association study had 80% power to identify a common variant explaining at least 25% of the variation in this trait, and our gene-based collapsing analysis of low-frequency coding variants had 80% power to identify associations explaining at least 22% of the variation. When restricting to 139 candidate genes annotated as being involved in the phototransduction and retinol metabolism pathways or implicated in Mendelian diseases causing night vision defects, we still found no significantly associated variants. However, we did identify a trend for participants with rare coding variants in RP1 to exhibit worse scotopic CS (corrected p = 0.018; VA corrected p not significant). This association was driven by three nonsynonymous variants—rs137887415, rs16920621, and rs142318038—each of which had one heterozygous carrier in our dataset.
This study characterized the performance of 504 healthy observers on scotopic VA and CS and analyzed numerous factors with the potential to affect performance, most of which had not previously been assessed with regard to this phenotype. Focusing on young individuals with excellent photopic vision, we found a wide variation in performance (Fig 1) that is largely unexplained by a range of factors such as Circadian preference, photopic visual performance, intelligence, or eye characteristics. Combined with high test-retest agreement and the existence of diseases uniquely targeting rod systems, our findings argue for a strong genetic component of healthy variation in night vision that demands exploration.
It is clear that the amount of time spent observing the optotype plays a significant role in scotopic conditions in both the VA and CS test. The amount of time that the participant chose to spend observing the optotype before deciding how to respond is essentially their reaction time, and here we represent this concept as overall task duration. Note that the number of trials in the VA and CS tests used here was fixed, as was the dark adaptation time for each participant, so the relationship between task duration and performance is not a product of differences in the number of trials or the dark adaptation length. While the duration of optotype presentation has been previously documented to impact photopic performance, ours is the first study to show a strong effect under scotopic conditions . In fact, our study shows a much stronger effect of task duration on scotopic performance than on photopic performance (Fig 5 and S2 Fig). In addition, we find intelligence to be a predictor of scotopic performance, which may reflect an increased ability to navigate this testing paradigm or reflect different approaches or level of interest in the testing according to intelligence. Intelligence was also positively correlated with task duration, indicating that higher intelligence may have led to improved testing strategies. Despite the significance of these variables to performance, they collectively only explain < 30% of the variation in scotopic VA and CS, leaving the vast majority of the variation in scotopic visual performance unexplained.
Visual acuity in logMAR units have an inverted scale, and contrast sensitivity is in logCSWeber units, meaning that better performance corresponds to higher scores. Task duration is shown on a log scale and explained 27% of the variance in scotopic VA and CS.
Our results confirm that individuals who have undergone laser eye surgery perform significantly worse on both scotopic VA and CS in addition to previously reported night vision issues in such patients [44, 45, 55]. In addition, photopic performance was significantly correlated with scotopic performance, but only explained < 5% of the variance. In a previous study of 47 normal observers, CS performance under mesopic and photopic conditions was also significantly correlated, with 27% of the variance in common . This work documented a notable range of mesopic CS performance that was independent of photopic CS performance. We confirm and extend this finding into scotopic conditions for CS, and we demonstrate the same marked dissociation between scotopic and photopic VA.
In addition, the results demonstrate no effect of astigmatism, iris color, Circadian preference, depression status, time of day, fatigue or color vision deficiency on scotopic performance. We were underpowered to assess the effects of color vision deficiency as our study only contained 7 individuals; the mean performance of these participants was in line with that of the remainder of the population for both VA and CS tasks. This finding is concordant with recent work demonstrating no expected effect of color vision deficiency , but stands in opposition to older research demonstrating lower light perception thresholds in these individuals . We do find worse scotopic VA and CS performance in those of Asian ancestry when performing univariate regression analyses, but these associations were no longer statistically significant after accounting for laser eye surgery, task duration, intelligence, and photopic performance.
The preliminary genetic analyses carried out utilized low frequency coding variants as well as common GWAS variants. We performed both single variant analyses and gene-based collapsing analyses that combined the effects of all rare coding variants in each gene. Our focused analysis of candidate genes identified a trend for three participants with rare coding variants in RP1 to exhibit worse scotopic CS. Loss of function mutations in this gene are known to cause autosomal dominant retinitis pigmentosa, a retinal disease with night blindness and progressive retinal degeneration [58, 59]. The variants in our study are not loss of function mutations: they are rare nonsynonymous variants, most of which are predicted by Polyphen2 to be benign . This preliminary observation warrants follow-up in additional samples to determine whether variation in this gene impacts the normal range of human scotopic vision. Our genome-wide analysis was powered to identify associations explaining at least 25% of the variation in scotopic vision, and we did not identify any variants or genes with such a large effect. This result is not unexpected given the complexity of night vision, but it does lay the groundwork for the types of sample sizes and genetic data that will be required for future studies of this trait.
While our study was large in size, it may have been limited by self-report responses given by participants. We do not have access to the medical records of our participants, and they did not undergo full ophthalmologic exams to determine the presence of any organic eye disease beyond those that are self-reported. In future studies, a full ophthalmic assessment of the participants with indications of organic eye disease or poor performance would be helpful for confirming and extending our findings.
Our study demonstrates the wide variation that is seen between healthy volunteers in their scotopic visual performance, even after controlling for age and photopic vision. We confirm that this inter-individual variation is stable and largely uninfluenced by known factors, which strongly supports there being a genetic component to these scotopic visual traits. Future studies may expand on these findings through a comprehensive, large-scale characterization of scotopic vision in healthy observers to continue assessing the impact of genetics on visual acuity and scotopic visual performance.
S1 Fig. Test-retest assessment over different days.
The tests presented here were taken a mean of 87 days apart (SD = 21). Four data sets are depicted: Visual acuity (left panel) and contrast sensitivity (right panel) at photopic luminance (green triangles, near top left and at scotopic luminance (blue discs, near bottom left). Result of the first test on the abscissa, second test on the ordinate. Grey 45°-line is the identity line, next to it the ± limits of agreement (photopic, dashed; scotopic, dotted). Visual acuity in logMAR units have an inverted scale, and contrast sensitivity is in logCSWeber units, meaning that better performance corresponds to the top right for both graphs.
S2 Fig. Correlation between photopic VA, photopic CS, and task duration.
Visual acuity in logMAR units have an inverted scale, and contrast sensitivity is in logCSWeber units, meaning that better performance corresponds to higher scores. Task duration is shown on a log scale and explained <0.1% of the variance in scotopic VA and 3.3% of the variance in scotopic CS.
S1 File. Phenotypes and Covariates.
Phenotypic and covariate information for the samples included in this study.
S2 File. Collapsing Analyses.
Results from the gene-based collapsing analyses included in this study.
We acknowledge D. Goldstein for study support, Q. Wang, M. Halvorsen and A. Gussow for assistance with programming, D. Attix for design of the cognitive battery, J. Quinlan for aid with the dark adaptometer, and K. Linney, J.M. McEvoy, P. Hunt, V. Dixon, T. Pennuto, K. Cornett, D. Swilling, L. Phillips, O. Chiba-Falek, M. Silver, J. Covington, and R. Wilson for recruitment and testing of participants.
Conceived and designed the experiments: ETC AJB. Performed the experiments: AJB. Analyzed the data: ETC AJB MB EML DC. Contributed reagents/materials/analysis tools: MB DC. Wrote the paper: ETC AJB MB EML DC.
- 1. Barbur J, Stockman A. Photopic, Mesopic, and Scotopic Vision and Changes in Visual Performance. 2010.
- 2. Zele AJ, Cao D. Vision under mesopic and scotopic illumination. Frontiers in Psychology. 2014;5:1594. pmid:25657632
- 3. Smith VC, Pokorny J. Spectral sensitivity of the foveal cone photopigments between 400 and 500 nm. Vision research. 1975;15(2):161–71. pmid:1129973
- 4. Hecht S, Mintz EU. The visibility of single lines at various illuminations and the retinal basis of visual resolution. The Journal of general physiology. 1939;22(5):593–612. pmid:19873122
- 5. Zele AJ, Cao D. Vision under mesopic and scotopic illumination. Frontiers in psychology. 2014;5.
- 6. Owsley C, McGwin G. Vision and driving. Vision research. 2010;50(23):2348–61.
- 7. Hertenstein H, Bach M, Gross NJ, Beisse F. Marked dissociation of photopic and mesopic contrast sensitivity even in normal observers. Graefe's Archive for Clinical and Experimental Ophthalmology. 2015:1–12.
- 8. Kniestedt C, Stamper RL. Visual acuity and its measurement. Ophthalmology Clinics of North America. 2003;16(2):155–70, v. pmid:12809155
- 9. Hohberger B, Laemmer R, Adler W, Juenemann AG, Horn FK. Measuring contrast sensitivity in normal subjects with OPTEC® 6500: influence of age and glare. Graefe's Archive for Clinical and Experimental Ophthalmology. 2007;245(12):1805–14.
- 10. Lee Y-C, Hu F-R, Wang I-J. Quality of vision after laser in situ keratomileusis: influence of dioptric correction and pupil size on visual function. Journal of Cataract & Refractive Surgery. 2003;29(4):769–77.
- 11. Sherwin JC, Reacher MH, Dean WH, Ngondi J. Epidemiology of vitamin A deficiency and xerophthalmia in at-risk populations. Transactions of the Royal Society of Tropical Medicine and Hygiene. 2012;106(4):205–14. pmid:22326691
- 12. Christian P, West K. Interactions between zinc and vitamin A: an update. The American journal of clinical nutrition. 1998;68(2):435S–41S. pmid:9701158
- 13. Schefrin BE, Tregear SJ, Harvey LO, Werner JS. Senescent changes in scotopic contrast sensitivity. Vision research. 1999;39(22):3728–36. pmid:10746143
- 14. Muth ER, Laurent JM, Jasper P. The effect of bilberry nutritional supplementation on night visual acuity and contrast sensitivity. Alternative Medicine Review. 2000;5(2):164–73. pmid:10767671
- 15. Jackson GR, Owsley C. Scotopic sensitivity during adulthood. Vision research. 2000;40(18):2467–73. pmid:10915886
- 16. Curcio C, Millican C, Allen K, Kalina R. Aging of the human photoreceptor mosaic: evidence for selective vulnerability of rods in central retina. Investigative ophthalmology & visual science. 1993;34(12):3278–96.
- 17. Kasper T, Bühren J, Kohnen T. Visual performance of aspherical and spherical intraocular lenses: intraindividual comparison of visual acuity, contrast sensitivity, and higher-order aberrations. Journal of Cataract & Refractive Surgery. 2006;32(12):2022–9.
- 18. Wikler KC, Williams RW, Rakic P. Photoreceptor mosaic: number and distribution of rods and cones in the rhesus monkey retina. Journal of Comparative Neurology. 1990;297(4):499–508. pmid:2384610
- 19. Pulvermüller A, Maretzki D, Rudnicka-Nawrot M, Smith WC, Palczewski K, Hofmann KP. Functional differences in the interaction of arrestin and its splice variant, p44, with rhodopsin. Biochemistry. 1997;36(30):9253–60. pmid:9230059
- 20. Dirani M, Chamberlain M, Garoufalis P, Chen C, Guymer RH, Baird PN. Refractive errors in twin studies. Twin Research and Human Genetics. 2006;9(04):566–72.
- 21. Jaakko K, Markku K. Heritability of defects of far vision in young adults—a twin study. Scandinavian Journal of Public Health. 1992;20(2):73–8.
- 22. Sung C-H, Schneider BG, Agarwal N, Papermaster DS, Nathans J. Functional heterogeneity of mutant rhodopsins responsible for autosomal dominant retinitis pigmentosa. Proceedings of the National Academy of Sciences. 1991;88(19):8840–4.
- 23. Bech-Hansen NT, Naylor MJ, Maybaum TA, Pearce WG, Koop B, Fishman GA, et al. Loss-of-function mutations in a calcium-channel α1-subunit gene in Xp11. 23 cause incomplete X-linked congenital stationary night blindness. Nature genetics. 1998;19(3):264–7.
- 24. Cremers FP, van de Pol DJ, van Driel M, den Hollander AI, van Haren FJ, Knoers NV, et al. Autosomal recessive retinitis pigmentosa and cone-rod dystrophy caused by splice site mutations in the Stargardt's disease gene ABCR. Human molecular genetics. 1998;7(3):355–62. pmid:9466990
- 25. Rosenfeld PJ, Cowley GS, McGee TL, Sandberg MA, Berson EL, Dryja TP. A null mutation in the rhodopsin gene causes rod photoreceptor dysfunction and autosomal recessive retinitis pigmentosa. Nature genetics. 1992;1(3):209–13. pmid:1303237
- 26. Miyake Y, Yagasaki K, Horiguchi M, Kawase Y, Kanda T. Congenital stationary night blindness with negative electroretinogram: a new classification. Archives of ophthalmology. 1986;104(7):1013–20. pmid:3488053
- 27. Kabanarou S, Holder G, Fitzke F, Bird A, Webster A. Congenital stationary night blindness and a “Schubert-Bornschein” type electrophysiology in a family with dominant inheritance. British journal of ophthalmology. 2004;88(8):1018–22. pmid:15258017
- 28. Roosing S, Rohrschneider K, Beryozkin A, Sharon D, Weisschuh N, Staller J, et al. Mutations in RAB28, encoding a farnesylated small GTPase, are associated with autosomal-recessive cone-rod dystrophy. The American Journal of Human Genetics. 2013;93(1):110–7. pmid:23746546
- 29. Cirulli ET, Kasperavičiūtė D, Attix DK, Need AC, Ge D, Gibson G, et al. Common genetic variation and performance on standardized cognitive tests. European Journal of Human Genetics. 2010;18(7):815–20. pmid:20125193
- 30. Cirulli ET, Attix DK, Smith PJ, Chiba-Falek O, Pennuto TOC, Linney KN, et al. Contribution of pastimes and testing strategies to the performance of healthy volunteers on cognitive tests. The Clinical Neuropsychologist. 2011;25(5):778–98. pmid:21722049
- 31. Bartholomew AJ, Cirulli ET. Individual variation in vontagious yawning susceptibility is highly stable and largely unexplained by empathy or other known factors. PLoS One. 2014;9(3):e91773.
- 32. Beck AT, Steer RA, Brown GK. Manual for the beck depression inventory-II. San Antonio, TX: Psychological Corporation; 1996.
- 33. Ottoni GL, Antoniolli E, Lara DR. The Circadian Energy Scale (CIRENS): two simple questions for a reliable chronotype measurement based on energy. Chronobiology International. 2011;28(3):229–37. pmid:21452918
- 34. Bassi CJ, Powers MK. Daily fluctuations in the detectability of dim lights by humans. Physiology & behavior. 1986;38(6):871–7.
- 35. Ribelayga C, Cao Y, Mangel SC. The circadian clock in the retina controls rod-cone coupling. Neuron. 2008;59(5):790–801. pmid:18786362
- 36. Hammond BR, Nanez JE, Fair C, Snodderly DM. Iris color and age‐related changes in lens optical density⋆. Ophthalmic and Physiological Optics. 2000;20(5):381–6.
- 37. Bradley JC, Bentley KC, Mughal AI, Bodhireddy H, Young RS, Brown SM. The effect of gender and iris color on the dark-adapted pupil diameter. Journal of ocular pharmacology and therapeutics. 2010;26(4):335–40. pmid:20698797
- 38. Terman JS, Terman M. Photopic and scotopic light detection in patients with seasonal affective disorder and control subjects. Biol Psychiatry. 1999;46(12):1642–8. pmid:10624545.
- 39. Bach M. The Freiburg Visual Acuity Test-automatic measurement of visual acuity. Optometry & Vision Science. 1996;73(1):49–53.
- 40. Bach M. The Freiburg Visual Acuity Test-variability unchanged by post-hoc re-analysis. Graefe's Archive for Clinical and Experimental Ophthalmology. 2006;245(7):965–71.
- 41. Ferris FL, Kassoff A, Bresnick GH, Bailey I. New visual acuity charts for clinical research. American journal of ophthalmology. 1982;94(1):91–6. pmid:7091289
- 42. Puts MJ, Pokorny J, Quinlan J, Glennie L. Audiophile hardware in vision science; the soundcard as a digital to analog converter. Journal of Neuroscience Methods. 2005;142(1):77–81. pmid:15652619
- 43. Bland JM, Altman D. Statistical methods for assessing agreement between two methods of clinical measurement. The lancet. 1986;327(8476):307–10.
- 44. Kohnen T, Bühren J, Kasper T, Terzi E. Quality of vision after refractive surgery. Cataract and Refractive Surgery: Springer; 2005. p. 303–14.
- 45. Seiler T, Kaemmerer M, Mierdel P, Krinke H-E. Ocular optical aberrations after photorefractive keratectomy for myopia and myopic astigmatism. Archives of ophthalmology. 2000;118(1):17–21. pmid:10636408
- 46. Koh S, Maeda N, Kuroda T, Hori Y, Watanabe H, Fujikado T, et al. Effect of tear film break-up on higher-order aberrations measured with wavefront sensor. American journal of ophthalmology. 2002;134(1):115–7. pmid:12095817
- 47. StataCorp. Stata Statistical Software. Stata Statistical Software. College Station, TX: StataCorp LP; 2013.
- 48. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MAR, Bender D, et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. The American Journal of Human Genetics. 2007;81(3):559–75. pmid:17701901
- 49. Price AL, Patterson NJ, Plenge RM, Weinblatt ME, Shadick NA, Reich D. Principal components analysis corrects for stratification in genome-wide association studies. Nature Genetics. 2006;38(8):904–9. pmid:16862161
- 50. Kanehisa M, Goto S, Sato Y, Kawashima M, Furumichi M, Tanabe M. Data, information, knowledge and principle: back to metabolism in KEGG. Nucleic Acids Res. 2014;42(Database issue):D199–205. pmid:24214961; PubMed Central PMCID: PMC3965122.
- 51. Kanehisa M, Goto S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000;28(1):27–30. pmid:10592173; PubMed Central PMCID: PMC102409.
- 52. Cirulli ET, Lasseigne BN, Petrovski S, Sapp PC, Dion PA, Leblond CS, et al. Exome sequencing in amyotrophic lateral sclerosis identifies risk genes and pathways. Science. 2015:aaa3650.
- 53. Feng S, Wang S, Chen C-C, Lan L. GWAPower: a statistical power calculation software for genome-wide association studies with quantitative traits. BMC genetics. 2011;12(1):12.
- 54. Heinrich SP, Krüger K, Bach M. The effect of optotype presentation duration on acuity estimates revisited. Graefe's Archive for Clinical and Experimental Ophthalmology. 2010;248(3):389–94. pmid:20076965
- 55. Villa C, Gutiérrez R, Jiménez JR, Gonzalez-Méijome JM. Night vision disturbances after successful LASIK surgery. British journal of ophthalmology. 2007;91(8):1031–7. pmid:17314153
- 56. Jägle H, de Luca E, Serey L, Bach M, Sharpe LT. Visual acuity and X-linked color blindness. Graefe's Archive for Clinical and Experimental Ophthalmology. 2006;244(4):447–53. pmid:16133025
- 57. Verhulst S, Maes F. Scotopic vision in colour-blinds. Vision research. 1998;38(21):3387–90. pmid:9893853
- 58. Pierce EA, Quinn T, Meehan T, McGee TL, Berson EL, Dryja TP. Mutations in a gene encoding a new oxygen-regulated photoreceptor protein cause dominant retinitis pigmentosa. Nat Genet. 1999;22(3):248–54. pmid:10391211.
- 59. Sullivan LS, Heckenlively JR, Bowne SJ, Zuo J, Hide WA, Gal A, et al. Mutations in a novel retina-specific gene cause autosomal dominant retinitis pigmentosa. Nat Genet. 1999;22(3):255–9. pmid:10391212; PubMed Central PMCID: PMC2582380.
- 60. Flanagan SE, Patch AM, Ellard S. Using SIFT and PolyPhen to predict loss-of-function and gain-of-function mutations. Genet Test Mol Biomarkers. 2010;14(4):533–7. pmid:20642364.