Influence of Sleep Deprivation on Feedforward and Recurrent Processing

This preregistration is linked to the project “Influence of spontaneous fluctuation in arousal on feedforward and recurrent processing.” Consequently, much of the analysis plan closely follows our previous preregistration for that project. In the previous study, we examined spontaneous fluctuations in arousal on a second‑to‑minute timescale, also referred to as the local arousal state (Beerendonk, Mejias et al., 2024, 2025). Many studies have shown that local arousal often exhibits an inverted U‑shaped relationship with performance, commonly known as the Yerkes–Dodson law (Yerkes & Dodson, 1908). In that work, we found evidence that the neural mechanisms underlying this inverted U‑shaped relationship most likely arise at later stages of visual processing, i.e., during recurrent processing. The next step is to understand how neural processing behaves across different global arousal regimes, such as those induced by sleep deprivation. Previous work has shown that when global arousal is manipulated pharmacologically, the arousal–performance curve shifts towards higher arousal levels while maintaining its inverted U‑shaped form (Beerendonk, Mejias et al., 2025). In the current project, we manipulate participants’ global arousal state via sleep deprivation and investigate how these changes affect the neural processing of visual stimuli. Specifically, we aim to assess how sleep deprivation alters feedforward and recurrent processing. To this end, we will use multivariate pattern analysis (MVPA; decoding) of EEG signals elicited by Kanizsa illusions and by local contrast manipulations (figure rotation). These visual features have been shown to differentially index recurrent and feedforward processing, respectively (Fahrenfort et al., 2017; Noorman et al., 2024). In addition, we will use pre‑stimulus pupil size as an index of arousal (Beerendonk et al., 2024, 2025; de Gee et al., 2017; Joshi et al., 2016; Lloyd et al., 2023; Murphy et al., 2014; Reimer et al., 2016; Strauch et al., 2022).