Source
Center for Human Genome Variation, Duke University School of Medicine, Durham, NC, USA.
Abstract
Clinicians routinely query factors known to impact cognitive test scores, including age and education. However, without data delineating the impact of less-frequently tracked variables, clinicians are limited to educated inferences about their effect. We explored the relationship of demographics, pastimes, and strategies with cognitive scores in a sample of 499 healthy young volunteers. As expected, age, education, ethnicity, and native language were strongly associated with most tests, while gender and dysphoria were associated with only some. Interestingly, pastimes such as playing number games and word games, and doing activities similar to the tests, were strongly associated with many measures, and testing strategies with almost all. Importantly, at least an additional 50% of the variation in Digit Span Backward and Animals scores was explained by adding covariates about pastimes and strategies to demographic covariates. These results support the utility of querying these elements.