Sektionen
Sie sind hier: Startseite Abteilungen und Arbeitsgruppen Neuropsychologie Forschung
Artikelaktionen

Research

Systems Memory Consolidation via Reactivation

 

After encoding, neural representations of past experience have to be transformed to become stable and lasting memories. The neural substrate supporting new memories gradually shifts from highly plastic hippocampal to slower-learning neocortical regions, a process that has been termed systems memory consolidation. This transformation is thought to be achieved by repeated training of neocortical memory networks, either by active rehearsal or by offline reactivation during sleep.

Until recently, it has been difficult to study the covert processes that support memory consolidation. The advent of novel imaging methods and analysis approaches has now made it possible to locate and track memory engrams in the human brain. Contrary to the long-held believe that the neocortex is a slow learner, studies applying these tools have found that independent neocortical memory representations can be formed rapidly from the outset of learning.

The overall goal of our research is to explore three factors that we think contribute to an accelerated redistribution of physical memory traces from the hippocampus to neocortex –retrieval, distributed practice, and sleep. We apply diffusion-weighted imaging of learning-induced plasticity in the human brain to test how and where memory is stored in the brain and to observe the redistribution of memories over time. To study the role of sleep in this process, we use pattern recognition to detect the covert reactivation of memories on a neural level, or we employ targeted memory reactivation to bias which information our brains consolidate during the night.

 

Spontaneous Memory Reprocessing during Sleep in Humans

 

Sleep holds a crucial role in forming lasting long-term memories (Schönauer et al., J Cogn Neurosci, 2014; Schönauer et al., Sleep, 2014; Schönauer et al., Cortex, 2015; Himmer et al., Neurobiol Learn Mem, 2017). Recently, we developed a way to detect memory reprocessing during sleep in humans, using machine learning approaches, which can find patterns in ongoing brain activity that are related to previous learning (Schönauer*, Alizadeh*, et al., Nat Commun, 2017). This makes it possible to study the covert processes that contribute to memory stabilization during sleep and investigate involved brain areas, as well as the underlying mechanisms. We found that the amount of memory reactivation during sleep is directly related to overnight memory stability, demonstrating the important role of offline repetition in memory consolidation. Our current research focuses on how memory processing during sleep gives rise to systems memory consolidation.

 

Rapid Neocortical Plasticity via Memory Rehearsal

 

Contrary to the long-held belief that systems consolidation takes weeks or months to develop, additional learning repetitions can accelerate the process considerably. In two studies, we showed that extra-hippocampal memory representations can be established rapidly over rehearsal during wakefulness in areas like the occipital and parietal cortex, which previously have not been seen as dedicated memory regions (Brodt et al., Proc Natl Acad Sci U S A, 2016; Brodt et al., Science, 2018). Using a novel diffusion-based imaging method that measures learning-induced plasticity in brain microstructure, we demonstrated that learning-related changes in functional brain activity evoke concurrent rapid changes in brain structure, over the course of a single learning session (Brodt et al., Science, 2018). These structural changes in the neocortex remained stable over time and predicted the long-term retention of the memories, thus fulfilling all necessary criteria of a memory engram. Importantly, while systems memory consolidation can be initiated by rehearsal during wakefulness, we found that sleep is necessary to make it last (Himmer*, Schönauer*, et al., Sci Adv, 2019). Our current research looks at how memory engrams develop from the outset of learning and how they are redistributed over time.

« März 2024 »
März
MoDiMiDoFrSaSo
123
45678910
11121314151617
18192021222324
25262728293031
Benutzerspezifische Werkzeuge