Lernen und Konsolidierung von salienter Information: der kausale Einfluss von Theta-Oszillationen
Zusammenfassung der Projektergebnisse
Why do we remember some memories easily for the rest of our lives while forgetting others within a few minutes? It has been shown that memories are selective and that the salience of information critically influences the retention of memories. In this line, it has been shown that memory enhancement for salient information correlates (i) with increased activity in the cortico-mesolimbic dopaminergic circuit (in particular, the substantia nigra/ ventral tegmental area complex [SN/VTA] together with the hippocampus) and (ii) with enhanced theta oscillations. This project aimed to further pinpoint the neurocognitive processes that are associated with encoding and consolidation of salient information, with a focus on the role of theta oscillations for salient information. In order to address this crucial question from multiple angles, I used a multi-modal approach with studies using functional magnetic resonance imaging (fMRI), brain stimulation via transcranial direct current stimulation (tDCS), scalp electroencephalography (EEG), and intracranial EEG. In a first experiment using fMRI (published in: Gruber et al., 2016, Neuron), I tested the effect of salient information on post-learning neural dynamics and subsequent memory for salient information. I found that post-learning increases in communication (resting-state functional connectivity [RSFC]) between the SN/VTA and hippocampus predicted preferential retention of salient information. In addition, multivariate pattern classification revealed that hippocampal representations of high-salience contexts were preferentially reactivated during post-learning rest, and the number of hippocampal reactivations was predictive of preferential retention of salient information. These findings are novel in humans and are consistent with findings in rodents. Further behavioral experiments indicated that post-learning rest periods (compared to active no-rest periods) enhanced source memory for salient information, again consistent with the idea of enhanced consolidation processes during post-learning rest. In a series of three further studies, I then asked how theta oscillations affect encoding and consolidation of salient information. In a first study in which we measured intracranial EEG from the prefrontal cortex and hippocampus of two pharmaco-resistant epilepsy patients, I found phase synchronization (an index of communication between two distant brain areas) in the theta frequency band between the prefrontal cortex and the hippocampus. Importantly, this theta phase synchronization was enhanced for salient information (i.e. contextually unexpected, novel information). Therefore, this experimental paradigm was most suitable to test how brain stimulation of the prefrontal cortex might affect hippocampus-dependent learning of salient information. Thus, in a key experiment combining tDCS-EEG, the experimental group of participants received tDCS over left prefrontal cortex for 20 minutes ahead of the learning phase of the experiment. Control group participants received a sham stimulation for 20 minutes ahead of learning. EEG was recorded during the learning phase (i.e. salient and non-salient information), a post-learning rest phase, and a memory retrieval phase. This protocol allowed to investigate brain stimulation effects on theta oscillations in all critical stages of memory (i.e. encoding, consolidation, and retrieval). The behavioral and EEG data with a focus on theta oscillations are currently analyzed. Potentially promising, preliminary evidence suggested that brain stimulation might improve memory performance, but this memory improvement was independent of the salience of information. Currently, more analyses are needed to better understand a potentially causal role of theta oscillations for successful learning of information. In a final experiment, we tested whether theta oscillations can be enhanced via creating intrinsically salient states (via curiosity). Indeed, preliminary analyses suggested that intrinsically salient states elicit theta oscillations and that the magnitude of such theta oscillations correlated with memory benefits for neutral information during intrinsically salient states. In conclusion, this project demonstrated that salient information leads to enhanced consolidation and that theta oscillations play an important role during both salient information and salient states. Further analyses on the current data sets will help to better understand the role – and potentially the causal role – of theta oscillations in support of salient memories.
Projektbezogene Publikationen (Auswahl)
-
(2016). Postlearning hippocampal dynamics promote preferential retention of rewarding events. Neuron, 89, 1110-1120
Gruber, M., Ritchey, M., Wang, S.-F., Doss, M., & Ranganath, C.
