Work by day and sleep by night, do not sleep too little or too much: Effects of sleep duration, time of day and circadian synchrony on flanker-task performance in internet brain-game users from teens to advanced age.

Abstract

Research elucidating the effects of sleep and circadian rhythm on cognitive performance is advancing, yet many important questions remain. Using flanker-task performance scores from a large internet sample (N=48,881) with repeated measures of cognitive performance and linked prior-night self-reported sleep duration, we analysed the relationship between sleep duration, time of day of task performance, and chronotype synchrony with performance in participants aged 15-80 years. Results indicate a performance peak at 7hr habitual sleep duration, and point to a variable effect of deviation from habitual sleep duration depending on users' habitual sleep duration and age. Time-of-day effects were notable for a steady decline in performance up until 01:00 hours-02:00 hours for the group as a whole, which was accounted for by nighttime deterioration on trials requiring inhibitory executive functioning, particularly in older subjects. Analyses did not demonstrate an advantage for playing in synchrony with self-identified chronotype. Results strengthen findings indicating an inverted U-shaped relationship between sleep duration and cognitive performance across a broad spectrum of age groups. These findings underscore the importance of daytime task performance for tasks requiring inhibitory function, especially in elderly people. Findings highlight the utility of large-scale internet data in contributing to sleep and circadian science.