Analyses of Pupil Oscillation Frequency in Relation to Psychomotor Vigilance Test Lapses of Attention across Different Doses of Sleep Deprivation




Professor of Psychiatry

Project Summary

For the past academic year, I had the privilege of working with Dr. David Dinges, Dr. Sarah McGuire, and Dr. Mathias Basner in the Unit for Experimental Psychiatry in the Perelman School of Medicine. My laboratory experience was a phenomenal and enriching educational experience, and I gained new skills that will undoubtedly be important in my future career.


My project investigated how pupil oscillation may reflect the degree of vigilant decrement measured by the PVT during sleep deprivation. Past studies have shown that chronic sleep disruption in male mice has been correlated with reduced neural counts of the locus coeruleus, a noradrenergic nucleus that projects to cortical areas and is involved in sleep-wake regulation. Additional research has shown that pupillary changes in mice correlate with activity changes of the locus coeruleus. My project explored the extent to which pupillary changes acquired at 60 Hz resolution track the degree of vigilance decrement measured by daily 20-min. psychomotor vigilance tests (PVT). 90 healthy adults underwent either experimental sleep restriction (two bouts of 5-nights of sleep restriction (4 h time in bed [TIB]) separated by 1, 3, or 5 days of recovery sleep [4 h TIB] or a no-sleep-restriction (control) protocol (17 days of 10 h TIB). Fast Fourier transform waveform analyses of pupillary radius measurements, Pupillary Unrest Index calculations, and mean low-frequency pupil oscillation power (0.0-0.4 Hz) were generated using a MATLAB algorithm after removing artefacts. Afterwards, an ANOVA and the Tukey’s range post-hoc pairwise test were performed in R.


It was found that sleep deprivation is correlated with significantly greater pupillomotor unrest after just two days of sleep deprivation. Additionally, pupillomotor unrest is significantly correlated with mean reaction time, mean lapses, and mean false starts on the PVT. These results help reveal how the degree of sleep homeostatic drive is reflected in both low-frequency pupillary oscillations and pupillomotor instability. With future statistical analyses, the results of this study are powerful findings in elucidating how sleep deprivation may have direct physiological effects on the locus coeruleus.