Dekodieren von Gedächtnisreaktivierung im Schlaf
Zusammenfassung der Projektergebnisse
After encoding, new memories have to be transformed to last for the long-term. During the course of this consolidation, the neural substrate supporting the memories shifts from hippocampal to neocortical regions, a process that has been termed systems memory consolidation. Sleep has been proposed to play an important role in this transition. During my time in Princeton, I worked on three main research projects related to this topic: 1) decoding memory reprocessing from EEG recordings during sleep, 2) content classification of naturalistic stimuli in an existing fMRI data set (piloting stimuli for future research), as well as data collection and analysis for a project that 3) investigated systems consolidation of naturalistic learning material over rehearsal to determine the informational content of the emerging memory representations. I originally applied for a Research Fellowship with the German Research Foundation to work with Prof. Ken Norman to study the covert reprocessing of memory during sleep. Our main goals were to investigate the properties of memory reactivation, e.g. when elicited by targeted external reactivation cues. We found that cue presentation during sleep elicits memory-related processing. This effect is strongest during slow wave sleep and specific to the auditory cue that was previously linked to mnemonic material – we did not observe the same results when the classifier was trained on brain responses evoked by a non-mnemonic control sound. These results agree with published studies in the sleep and memory domain, stressing the role of SWS, and reactivation in this sleep stage specifically, for declarative memory consolidation. Because my work on tracking memory engrams in humans by diffusion and functional MRI has proven valuable in understanding systems consolidation, my host Prof. Ken Norman and I started a collaboration project on tracking memories for naturalistic stimuli over repeated learning and rapid systems consolidation over rehearsal with magnetic resonance imaging. Our main objective was to determine whether a neocortical memory representation emerges over repeated rehearsal of naturalistic stimuli. Specifically, we wanted to test whether we can decode the content of what has been learned from brain activity during memory encoding and retrieval, and by extrapolation, whether this could lead to discriminable structural memory engrams in regions responsible for memory storage. In line with our previous work using univariate approaches, the neocortex showed a general increase in category discrimination across learning runs, whereas the hippocampus showed across-run decreasing category discrimination, with significant content representations only during the first learning repetition. I published three research papers. The papers investigate systems consolidation over rehearsal in wakefulness, the complementary roles of rehearsal and sleep in systems memory consolidation, and the effect of sleep on memory stability.
Projektbezogene Publikationen (Auswahl)
- (2018). Fast track to the neocortex: A memory engram in posterior parietal cortex. Science; 362: 1045-1048
Brodt S, Gais S, Beck J, Erb M, Scheffler K, Schönauer M
(Siehe online unter https://doi.org/10.1126/science.aau2528) - (2019). Rehearsal initiates systems memory consolidation, sleep makes it last. Sci Adv; 5: eaav1695
Himmer L, Schönauer M, Heib D, Schabus M, Gais S
(Siehe online unter https://doi.org/10.1126/sciadv.aav1695)
