Experimental Sleep Deprivation Results in Diminished Perceptual Stability Independently of Psychosis Proneness

Abstract

Psychotic disorders as well as psychosis proneness in the general population have been associated with perceptual instability, suggesting weakened predictive processing. Sleep disturbances play a prominent role in psychosis and schizophrenia, but it is unclear whether perceptual stability diminishes with sleep deprivation, and whether the effects of sleep deprivation differ as a function of psychosis proneness. In the current study, we aimed to clarify this matter. In this preregistered study, 146 participants successfully completed an intermittent version of the random dot kinematogram (RDK) task and the 21-item Peters Delusion Inventory (PDI-21) to assess perceptual stability and psychosis proneness, respectively. Participants were randomized to sleep either as normal (8 to 9 h in bed) (n = 72; M_age = 24.7, SD = 6.2, 41 women) or to stay awake through the night (n = 74; M_age = 24.8, SD = 5.1, 44 women). Sleep deprivation resulted in diminished perceptual stability, as well as in decreases in perceptual stability over the course of the task. However, we did not observe any association between perceptual stability and PDI-21 scores, nor a tendency for individuals with higher PDI-21 scores to be more vulnerable to sleep-deprivation-induced decreases in perceptual stability. The present study suggests a compromised predictive processing system in the brain after sleep deprivation, but variation in psychosis trait is not related to greater vulnerability to sleep deprivation in our dataset. Further studies in risk groups and patients with psychosis are needed to evaluate whether sleep loss plays a role in the occurrence of objectively measured perceptual-related clinical symptoms.

Keywords: perceptual stability; predictive processing; psychosis proneness; random dot kinematogram; sleep deprivation.

Figure 1

(A) A typical trial of the Random Dot Kinematogram (RDK) task. Ambiguous dot kinematogram are presented repeatedly for 600 ms interleaved by a blank screen of 800 ms duration. The sphere can be perceived as rotating either leftward or rightward. Participants are asked to report the perceived direction of the rotation by making a button press. (B) The perceptual time course from an example participant. The stabilizing effect of endogenous predictions are automatically built up during intermittent presentation of the ambiguous stimulus, this results in participants tending to have the same percept across successive presentation cycles. The figure is replicated with permission from Schmack et al. [9].

Figure 2

(A) Conditional effects of sleep condition (sleep deprivation no/yes) on adjusted survival probability and (B) interaction effect of sleep condition (sleep deprivation no (“normal sleep” group) in red and sleep deprivation yes (“sleep deprived” group) in blue) by PDI-21 conviction score. Error bars and bands represent 95% confidence intervals.

Figure 3

(A) Condition effects of previous response missing, (B) previous response left or right, and (C) interaction effect of sleep condition (sleep deprivation no (“normal sleep” group) in red and sleep deprivation yes (“sleep deprived” group) in blue) by completed trial fraction. Error bars and bands represent 95% confidence intervals.