|Abstract/Results: ||Episodic memories pertain to specific events for which both content and context can be consciously recalled and verbalized. Between encoding and retrieval, episodic memories are thought to be maintained within hippocampal and neocortical circuits, then to be gradually consolidated into robust memories which can eventually be accurately and comprehensively retrieved in the long term.|
However, the mechanisms by which a recent episodic memory is initially maintained in the brain are not yet fully understood.
The main aim of this thesis is to examine whether learning an engaging task modifies spontaneous brain activity during subsequent resting state wakefulness and sleep onset, and whether such replay of brain activity has an effect on offline memory processing. We examined healthy human volunteers behaviorally and with non-invasive neuroimaging techniques in two different studies.
In the first study, we examined the emergence of task-related hypnagogic hallucinations during a daytime nap with EEG. In the experimental group (n=16), participants played Tetris in the morning for two hours during three consecutive days, while in a first control group (n=13, controlling the effect of experience), participants did not play any game, and in a second control group (n=14, controlling the effect of anticipation), participants played Tetris after the nap. During afternoon naps, participants were repetitively awakened during sleep stage 1 and were asked to report their mental content. Reports content was scored by three judges (inter-rater reliability 85%). In the experimental group, 48 out of 485 (10%) sleep-onset reports were Tetris-related. These reports mostly consisted of images and sounds with very little emotional content. They exactly reproduced Tetris elements or mixed them with other mnemonic components. By contrast, in the first control group only 1 report out of 107 was scored as Tetrisrelated (1%), and in the second control group only 3 reports out of 112 (3%) (between groups comparison; p=0.006). Hypnagogic hallucinations were more consistently induced by experience than by anticipation (p=0.039) and they were predominantly observed during the transition of wakefulness to sleep. The observed attributes of experience-related hypnagogic hallucinations are consistent with the particular organization of regional brain activity at sleep onset, characterized by high activity in sensory cortices and in the default mode network.
In the second study, we tested if brain activity generated during learning is spontaneously repeated during subsequent resting wakefulness. Healthy participants (n=11) were scanned in a randomized within-subject cross-over design composed of two conditions in which fMRI rest sessions flanked an active fMRI session. We applied machine learning based multivariate pattern analysis on these fMRI time series to calculate the percentages of taskrelated scans before and after a memory task (encoding a two dimensional trajectory of face, building and animal pictures) and before and after a control task (an auditory oddball). Activity patterns elicited by visual stimuli were spontaneously expressed in the visual ventral stream before any cognitive challenge. We found that 1.11% more scans were task-related after learning than before, and that the larger the difference between the proportions of spontaneous brain activity linked to a task before and after this task, the better the memorization of task features by the subject, suggesting that reactivations during post-task rest are linked to the memorization of the task. On the other hand, 0.26% less scans were related to the memory task after a control task than before, suggesting that the control task had a repressive effect on the scans related to the memory task. Memory encoding modulated the temporal organization of the spontaneous activity in episodes of longer duration (4 to 6 seconds). Our results show that human memories are maintained during resting wakefulness by temporally organized repetitions of regional brain activity. However, because our other analyses (dynamic causal modeling, cross-correlations and explained variance, and spatial networks based on independent component analysis) could not confirm these findings, we suggest that fMRI might not be the most appropriate methodology to further characterize memory consolidation directly.
Our results add to the growing body of evidence suggesting that offline processing of explicit memories occurs during wakefulness. They extend our understanding of human spontaneous brain activity and its relationship to offline memory processing, although more research is warranted to further explore the details of the memory consolidation process. Interesting perspectives would be to pursue this work during sleep or with other neuroimaging methods such as high density EEG or simultaneous fMRI and EEG.