RESUMEN
Risky choice has been widely studied in experimental settings, but there is a paucity of research examining the effects of self-selected sleep schedules on risky choices. The current study examined incentivised risky choices of 100 young, healthy adults whose self-selected (at-home) sleep schedules were tracked via actigraphy for 1 week prior to decision making. Average nightly sleep was 6.43 h/night. On each trial of the decision task, individuals chose between two monetary gambles, with separate blocks of trials presenting amounts to gain versus amounts to lose for each paired gamble choice. In general, participants were risk-averse when trying to maximise gains (GAINS) and risk-seeking when trying to minimise losses (LOSSES). These tendencies were amplified in trials where gambles differed more (vs less) in their riskiness. Response times were longer for real choices (vs. dummy trials of random choice), LOSS versus GAINS trials, and when gambles were more similar versus different in risk. Gamble choices were not impacted by actigraphy measured average sleep levels, which suggests self-selected moderate sleep deprivation does not affect risky monetary choices, as has been found in studies of experimentally induced sleep deprivation. However, our data showed that sleep variability increased risk-taking behaviour in the LOSS condition. Thus, risky decision-making may relate more to variability in sleep efficiency than to overall sleep duration or quality in naturalistic settings. The current study gives insight into how decision making in experimental sleep settings may or may not translate to more ecologically valid settings of self-directed sleep.