Association of Electroencephalographic Statistical and Complexity Features with Cognitive Function Under Acute Partial Sleep Deprivation: Evidence from a Numerical Stroop Task

Background: Sleep deprivation adversely impacts cognitive performance. Despite the growing body of studies on electroencephalography (EEG) analysis, the association between complexity and statistical features across EEG electrodes and cognitive function of individuals has not been completely explored. In this study, we investigated the correlation between these features of EEG signals and the cognitive function of individuals assessed in a numerical Stroop test following acute partial sleep deprivation.
Methods: In this study, 19 male and female participants completed both the NST and EEG on the first day to establish their baseline cognitive and neural activity. During the subsequent 24-hour period, participants were instructed to restrict their sleep to a four-hour window between 3:00 a.m. and 7:00 a.m., remaining awake for the rest of the time. On the following day, they had a resting-state EEG recording which followed by a NST.
Results: After acute partial sleep deprivation, higher numerical Stroop error rates were significantly associated with increased EEG complexity measures, including Higuchi’s and Katz’s fractal dimensions and sample entropy, particularly in frontopolar, frontal, and temporal regions (strongest at Fp1). Kurtosis showed a positive association only at the temporal site (T7), while skewness demonstrated mixed correlations. Standard deviation was positively related to errors at frontal (Fp2) and parietal (P7) electrodes. Overall, increased EEG complexity and variability were linked to poorer cognitive performance.
Conclusion: Increased EEG complexity and variability, particularly in frontal and temporal regions, may serve as potential biomarkers of cognitive impairment during acute partial sleep deprivation.