The local regulation of human sleep: anatomo-functional bases and implications for behavior

The traditional view of sleep and wakefulness as two distinctand mutually exclusive states has been recently challenged by thediscovery that they actually are locally regulated and that islandsof sleep- and wake-like activity may often coexist in a sameindividual at a given time. Importantly, it has been suggested thatthe local regulation of sleep may be involved in many of theessential functions of sleep in physiological conditions. Particularattention has been given to the study of the so-called 'slow waves'of sleep, which represent in particular the main hallmark of nonrapideye movement (NREM) sleep. In fact, local changes in slowwave activity have been shown to occur in brain regions that aremore actively used during wakefulness, ultimately reflectingwake- and experience-dependent plastic processes. In addition,electrophysiological events similar to sleep slow waves have beenfound to occur also during wakefulness and have been suggestedto reflect neuronal functional fatigue and the accumulation ofsleep need. Of note, while experimental research has started toshed light on the mechanisms involved in the local regulation ofsleep-like activity, and on how they affect cognition and behavior,many aspects still remain to be fully clarified.Given these premises, in the present Thesis, my aim was toadvance the current knowledge on the local regulation of sleep inhumans, with a specific focus on slow-wave-like activity. To thisaim, I performed three different experiments. In the first study, Iinvestigated the role of cortico-cortical white matter connectionsin the generation and propagation of sleep slow waves. To thisaim I analyzed overnight high density (hd)EEG data collected inan extremely rare population of ‘split-brain’ patients and in twoadditional groups including neurologic patients and healthycontrol subjects. Obtained results demonstrated that the travelingof sleep slow waves is significantly affected by the resection of thecorpus callosum, which leads to a reduced proportion of crosshemisphericslow waves. This result demonstrates that the waysleep slow waves propagate can inform us regarding the status ofbrain connections and may thus offer a valuable marker forfunctional or structural alterations caused by traumatic orneurodegenerative disorders. On the other hand, our analyses showed that the lack of inter-hemispheric connections is notassociated with dissociations characterized by sleep rhythms inone hemisphere and wake-like activity in the other half of thebrain. In addition, while we found that sleep slow waves tend tooriginate more often in the right than in the left hemisphere, suchan asymmetry was found not to differ between split-brain patientsand subjects with an intact corpus callosum. Overall, these resultsindicate that global state changes are coherently modulated acrossthe cortical mantle by non-cortical (bottom-up) mechanisms.In two additional studies, I investigated the local regulation ofsleep-like activity during wakefulness and its possible effects oncognition and behavior. In particular, in one experiment I applieda single-subject multi-session design to explore whether theregional distribution of morning-to-evening increases in localsleep-like activity is dependent on the degree of experiencedependentactivation or rather it mainly reflects inter-regionaldifferences in vulnerability to neuronal fatigue. In fact, it has beenshown that low-frequency power increases during wakefulnessand decreases after a night of sleep, and such changes are onaverage more pronounced over frontal areas. Our results showedthat changes in low-frequency activity may peak in different brainregions. In particular, we observed at least two main morning-toeveningvariation patterns: one, more common and stronger,involving centro-frontal cortical areas, and one, less common,mainly involving sensory cortices. This observation does notsupport an inherent vulnerability of frontal areas and is insteadpotentially compatible with a use/experience-dependentregulation of electrophysiological indices reflecting functionalfatigue and sleep need.Finally, I investigated whether the occurrence of local sleeplikeepisodes may influence behaviors with a social relevance,such as the ability to regulate one’s own emotional reactions. Inparticular, my aim was to test whether the occurrence local sleeplikeactivity within brain areas involved in emotional regulationcould account for failures in the suppression of emotionalexpressions. Obtained results demonstrated, for the first time, thatsleep-like activity in frontal and parietal areas precede emotionregulation failures. Moreover, I found that the incidence ofbehavioral failures is negatively correlated with a shorter sleep duration the night preceding the experiment, in line with previousevidence linking local sleep-like episodes and sleep loss. Takentogether, these results indicate that transient, local 'neuronal sleep'may represent a direct functional cause of impairment in complexand socially relevant human behaviors.