How deep is your sleeping mind? Unravelling the extent of information processing during sleep

One third of our lives is dedicated to sleep, a highly vulnerable state where we cannot interact with our environment. Still, it is thought that the reduced connection the sleeper maintains with her environment is a key condition for memory consolidation and the incorporation of daily information into new or existing schema (Diekelmann & Born, 2010; McClelland, McNaughton, & O’Reilly, 1995; Walker & Stickgold, 2010). The replay of memories during sleep was shown to be a crucial element for such processes (Oudiette & Paller, 2013). In order to reactivate memories, neural ensembles in which they are coded need to be available. Moreover, studies showed that these same neuronal ensembles may also process external stimuli during sleep, which contrasts with the apparent unresponsiveness of the sleeper. In particular, the sleeping brain is more reactive to relevant stimulations than meaningless ones. New mothers wake up more when they hear the cry of their own baby than the cry of another (Formby, 1967). Using electroencephalography (EEG), it was experimentally demonstrated that, during sleep, the brain reacts differently when hearing the sleeper’s own name than the name of someone else, and can detect semantic incongruities between pair of words and in sentences (Bastuji, Perrin, & Garcia-Larrea, 2002; Ibáñez, López, & Cornejo, 2006; Perrin, Bastuji, & Garcia-Larrea, 2002; Perrin, Garcı́a-Larrea, Mauguière, & Bastuji, 1999). More impressively, auditory words can reactivate learnt associated motor patterns (Andrillon, Poulsen, Hansen, Léger, & Kouider, 2016; Kouider, Andrillon, Barbosa, Goupil, & Bekinschtein, 2014) and new associations can be generated using auditory and olfactory conditioning procedures (Arzi et al., 2012, 2014). It therefore appears that complex information can be processed during sleep even though the activity of the brain changes substantially (Iber, 2007; Massimini et al., 2005; Nir & Tononi, 2010).

In my thesis work, I would like to assess the extent to which information can be processed during sleep. In particular, contextual information was shown to be taken in account by the caudal part of the dorsolateral prefrontal cortex in decision-making (Koechlin, Ody, & Kouneiher, 2003). During sleep , the activity of the dorsolateral prefrontal cortex is generally decreased (Braun et al., 1997) but on the other hand, contextual information like odors showed successful reactivation and consolidation of associated memories (Rasch, Büchel, Gais, & Born, 2007; Rihm, Diekelmann, Born, & Rasch, 2014). The purpose of this thesis is to shed light on the extent and the richness of contextual information of a memory that can be reactivated by external stimulations (and therefore potentially by spontaneous reactivations).