Short sleep recovery partially restores decision making alterations induced by total sleep deprivation

ABSTRACT The dynamics of recovery from impaired decision-making after total sleep deprivation (TSD) are not well understood. We investigated the impact of TSD and subsequent 3-h sleep recovery night period on decision-making. After one baseline night (8-h Time in Bed, TIB, 23:00-07:00), 40 subjects (50% women) followed a sleep deprivation protocol including one night of TSD (40-h continuous awakening), followed by 3-h TIB sleep recovery period (23:00-02:00) and 8-h TIB sleep recovery period. Indices derived from reaction time (RT), Go/No-Go, and complex Go/No-Go tasks (involving perceptual components, motor responses, reaction time, and accuracy) were assessed daily during dual-choice decision-making tasks (Minpulse Digital Battery). Composite indices to describe executive speed, reaction to difficulty, speed/accuracy balance and parameters from Decision Diffusion Model (DDM) analysis were recorded. Errors and RT increased after TSD and remained elevated after a 3-h sleep recovery night, particularly for Go/No-Go tasks. Anticipation and inhibition errors as well as speed/accuracy balance are not restored by 3-h sleep recovery night whereas RT was restored. Lower DDM Drift values (i.e. slower information accumulation) observed in the higher difficulty (complex Go/No-Go) after TSD, persisted after the 3-h recovery night. All parameters were restored after an 8-h TIB recovery night. No effect of sleep loss on executive speed and reaction to difficulty was observed. In conclusion, short sleep recovery partially restored decision-making alterations induced by TSD. Slower perceptual and motor processes that persist after a short recovery night may favor errors, with possible operational consequences for shift or on-call workers.