Our study included 8,511 European-ancestry individuals with DNA genotype data, behavioral questionnaire data and cognitive assessment data available, including 7 different cognitive ability tasks which were administered repetitively with an average 6.5-year interval (SD = 1.25 year). The average age of the study participants was 48.6 years at the time of the first round of cognitive assessment (WLS survey round 4, 1992–1994, SD = 15.4 years), 64.2 years at the second assessment (WLS survey round 5, 2003–2005, SD = 4.1 years), and 70.7 years at the time of the last assessment (WLS survey round 6, 2011, SD = 4.2 years). The sample was 51.8% female, 47.8% completed high school education or less than one year of college (number of years of education), 78.2% were born in Wisconsin, USA.
PheWAS of Cognitive GPSs in the Cognitive/Behavioral Phenome
1) Cognitive Phenotypes
Across all of the PheWAS results, IQ score showed the strongest associations with the four cognitive GPSs in terms of p-value and the increase proportion of variance explained (strongest with CP GPS, p-value = 7.74E-179, β = 1.42, 95% CI=(1.32, 1.51)) (Fig. 1, Table 1, Supplementary Table 1). The variance of IQ scores explained by CP GPS was 10.4% (Adjusted R2) whereas the baseline covariate model without GPS variable explained 0.8% of IQ score variance.
The years of educational attainment measure (strongest with EA GPS, p-value = 1.62E-129, β = 1.73, 95% CI=(1.59, 1.87)) and high school class rank (strongest with EA GPS, p-value = 3.07E-101, β = 1.86, 95% CI=(1.69, 2.02)) were also significantly associated with all of four cognitive GPSs, following IQ score. The variance of high school class rank explained by EA GPS was 17.1% (Adjusted R2) whereas the baseline covariate model without GPS variable explained 9.2% of high school class rank variance.
Among the cognitive tasks, the Similarities task presented the strongest p-value significance and positive effect sizes with cognitive GPSs in all the three rounds (strongest between timepoint3 Similarities and EA GPS, p-value = 3.59E-74, β = 1.36, 95% CI=(1.22, 1.51)). Cognitive GPSs also showed robust associations with Number Series (strongest with HM GPS, p-value = 2.55E-78, β = 0.94, 95% CI=(0.85, 1.04)) and Digit Ordering tasks (strongest with CP GPS, p-value = 8.63E-41, β = 0.78, 95% CI=(0.67, 0.89)) across different types of cognitive GPS. Several cognitive tasks were also consistently and significantly associated across the cognitive GPSs with positive effect sizes, including Letter Fluency (strongest association between timpoint3 Letter Fluency and CP GPS, p-value = 5.01E-30, β = 0.61, 95% CI=(0.50, 0.71)), Category Fluency (strongest association between timpoint3 Category Fluency with EA GPS, p-value = 2.03E-18, β = 0.95, 95% CI=(0.74, 1.16)), Immediate Recall (strongest association between timpoint3 Immediate Recall with EA GPS, p-value = 5.29E-22, β = 0.80, 95% CI=(0.64, 0.96)), Delayed Recall (strongest association between timpoint3 Delayed Recall with CP GPS, p-value = 5.00E-15, β = 0.45, 95% CI=(0.34, 0.56)), NVS Health Literacy assessments (strongest with CP GPS, p-value = 2.28E-29, β = 0.92, 95% CI=(0.76, 1.07)), which indicates genetic contribution to cognitive abilities are positively correlated to higher cognitive scores of several assessments (Table 1).
2) Behavioral Phenotypes
Among the Big 5 Personality traits, all the four GPS associations of Openness (strongest with EA GPS, p-value = 2.19E-14, β = 0.57, 95% CI=(0.42, 0.71)) met phenome-wide significance with positive effect sizes. In addition to Openness, HM and MA GPSs presented significant associations with Neuroticism (strongest with MA GPS, p-value = 1.92E-06, β=-0.32, 95% CI=(-0.45, -0.19)), showing negative effect sizes.
Social participation of reading books, magazines, newspapers or other reading material (strongest with EA GPS, p-value = 2.03E-21, β = 0.50, 95% CI=(0.40, 0.60)) and attending cultural events (strongest with EA GPS, p-value = 2.06E-23, β = 0.60, 95% CI=(0.49, 0.72)) presented phenome-wide significant associations across the cognitive GPSs with positive directions. Notably, social participation of watching TV watching TV (strongest with EA GPS, p-value = 4.16E-18, β=-0.48, 95% CI=(-0.59, -0.37)) and fishing/hunting (strongest with EA GPS, p-value = 1.72E-11, β=-0.59, 95% CI=(-0.77, -0.42)) showed significant negative associations with all the cognitive GPSs. Other phenome-wide significant activities included writing letters (strongest with EA GPS, p-value = 2.28E-11, β = 0.32, 95% CI=(0.23, 0.41)), working on crosswords or word games (strongest with CP GPS, p-value = 1.27E-10, β = 0.39, 95% CI=(0.27, 0.51)), and vigorous physical activities (alone) (strongest with EA GPS, p-value = 7.39E-10, β = 0.61, 95% CI=(0.42, 0.80)) (Table 1, Fig. 1).
Occupational education scores (strongest with EA GPS, p-value = 4.77E-61, β = 1.22, 95% CI=(1.08, 0.36)) and occupational income scores (strongest with EA GPS, p-value = 3.50E-37, β = 0.90, 95% CI=(0.76, 1.03)) presented positive relationships across all the cognitive GPSs. The association of Spouse IQ did not reach the phenome-wide significance with any of the cognitive GPSs. Full PheWAS results of the phenome-wide significant associations are available in Supplementary Table 1.
Cognitive GPSs correlate with the changes of Immediate Recall
Our linear mixed effect model identified the significant age-x-GPS interaction effect in the changes of Immediate Recall task with the participants’ age. All four kinds of cognitive capacity GPS showed significant interaction effects with participants’ age (Age:GPS) for the Immediate Recall test scores (strongest with EA GPS, p-value = 1.79E-03, β = 1.86E-01) (Table 2). Their positive effect sizes suggest that an individual who has higher EA GPS tend to show less changes in cognitive assessments as an individual ages.
Compared to the individuals in the lowest GPS quartile (EA GPS mean=-0.480), individuals in the highest GPS quartile (EA GPS mean=-0.089) showed less decrease in Immediate Recall score changes in later survey rounds. Slope of participants’ age in the lowest GPS quartile (β= -1.97E-01, 95% CI=(-0.231, -0.163), p-value of slope = 8.61E-31) distinctively showed more intense decrease compared to the highest GPS quartile group (β=-1.24E-01, 95% CI=(-0.158, -0.090), p-value of slope = 6.67E-13) (Fig. 2(a)). The pseudo-R2 of our linear mixed models in explaining Immediate Recall by cognitive capacity GPS was up to 0.063 with fixed effects and 0.170 with both fixed and random effects (both with EA GPS).
To visually depict the degree of cognitive changes according to GPS, we divided the cohort into 4 quartiles based on the GPS of each individual and analyzed the average phenotypic changes of each group over time. The average Immediate Recall task scores of the individuals in the highest GPS quartile were 0.044 (z-score) at timepoint 2 (average age of participants 64.5), and increased to 0.073 (z-score) at timepoint 3 (average age 71.5) (1.65 fold increase). In contrast, the average task scores of the individuals in the lowest GPS quartile were − 0.031 (z-score) at timepoint 2 and decreased to -0.077 (z-score) at timepoint 3 (2.48 fold decrease) (Fig. 2(b)).