ObjectivesTo investigate whether different protocols of sleep deprivation modify sleep perception. MethodsThe effects of total sleep deprivation (TD) and selective rapid eye movement (REM) sleep deprivation (RD) on sleep perception were analyzed in normal volunteers. Thirty-one healthy males with normal sleep were randomized to one of three conditions: (i) normal uninterrupted sleep; (ii) four nights of RD; or (iii) two nights of TD. Morning perception of total sleep time was evaluated for each condition. Sleep perception was estimated using total sleep time (in hours) as perceived by the volunteer divided by the total sleep time (in hours) measured by polysomnography (PSG). The final value of this calculation was defined as the perception index (PI). ResultsThere were no significant differences among the three groups of volunteers in the total sleep time measured by PSG or in the perception of total sleep time at baseline condition. Volunteers submitted to RD exhibited lower sleep PI scores as compared with controls during the sleep deprivation period (P < 0.05). Both RD and TD groups showed PI similar to controls during the recovery period. ConclusionSelective REM sleep deprivation reduced the ability of healthy young volunteers to perceive their total sleep time when compared with time measured by PSG. The data reinforce the influence of sleep deprivation on sleep perception.