Emdin, M. | Search bioRxiv

bioRxiv2025

Immersive NREM dreaming preserves subjective sleep depth against declining sleep pressure

Perceived sleep depth is a key determinant of subjective sleep quality, traditionally thought to reflect unconsciousness and reduced cortical activation. Here, we combined high-density EEG with a serial awakening paradigm during NREM sleep to examine its neural and experiential correlates. As expected, deeper sleep was associated with reduced cortical activation, reflected in a lower high-to-low frequency power ratio. Yet, this relationship weakened in the presence of dreaming, indicating that immersive conscious experiences can counteract the impact of cortical activation on perceived depth. Indeed, perceived sleep depth was lowest during states with a mere sense of presence and highest during immersive dreaming or deep unconsciousness. Across the night, physiological sleep pressure and subjective sleepiness declined, but perceived sleep depth rose alongside increasing dream immersiveness. These results challenge the view that deep sleep stems solely from reduced brain activity and suggest that immersive dreaming sustains perceived sleep depth as homeostatic pressure wanes.

neuroscience↗

bioRxiv2025

Chemosensitive brainstem and diencephalic components in breath-hold fMRI

ObjectiveThe knowledge on breathing control and central chemoreception, key subcortical functions involved in several neuropathologies, is still mainly based on animal studies. In humans, functional MRI (fMRI) offers the needed spatio-temporal resolution and non-invasiveness, but the lack of specific tools and preprocessing solutions hinders its use in brainstem studies. We hereby propose an original fMRI analysis pipeline aimed at unravelling central chemoreception mechanisms, by integrating acquisition, spatial coregistration, noise removal and a novel data-driven analysis solution to compare network activation levels across tasks or conditions in fMRI. ApproachNovel analysis methodologies are integrated with the optimization of known preprocessing approaches to physiological noise correction and brainstem-focused coregistration. We couple independent components of fMRI data, separately estimated from healthy subjects during Free Breathing (FB) and Breath Hold (BH), by means of spatial correlation. We then identify statistically significant differences between BH and FB in CO2-dependent components by means of voxel-wise comparisons of components percent signal change. Components were localized using the Brainstem Navigator Atlas for enhancing network interpretability. Main ResultsUsing the pipeline we characterized CO2-related BOLD oscillations within the central control system of breathing. We corroborated the primary chemoreceptive role of medullary raphe in healthy subjects. We observed that BH over-activated ascending sensory-motor projections through the postero-lateral thalamus, descending projections through the putamen, and peripheral sensations entry points in the dorsal medulla. We highlighted the role of latero-dorsal tegmentum in the response to hypercapnia-induced aversive effects. SignificanceOur method allows to non-invasively locate primary chemoreception and related arousal triggers, characterizing alterations and therefore fostering the identification of therapeutic targets in abnormal breathing. Moreover, the proposed strategy addresses the issues of inter-task comparison among homologous independent sources, of their characterization and interpretation. Its extension could benefit all similarly challenging brainstem-focused studies, including those on Parkinsons and Alzheimers diseases.

bioengineering↗