Sleep and Memory Consolidation

This project intends to provide neurophysiological evidence for the hypothesis that sleep serves to consolidate memories. For this purpose we selected three different memory tasks, for which behavioural data had demonstrated that, as compared to a corresponding retention interval of wakefulness, sleep enhances retrieval. Implicit perceptual memory will be tested with a visual discrimination task, implicit procedural motor memory with a mirror-tracing task and explicit declarative memory with a paired-associate word list task. Multi-channel electroencephalographic (EEG) recordings will be obtained in a parallel group design (3 x 24 healthy subjects) during initial learning in the afternoon (encoding), during the following night (storage/consolidation) and during recall in the morning thereafter as well as after one week (retrieval). EEG data analysis will include event-related desynchronisation and synchronisation (ERD/ERS) of individually determined frequency band power and low-resolution brain electromagnetic tomography (LORETA). Concerning encoding and retrieval, we want (1) to replicate results from ERD/ERS studies as well as (2) to compare the activated cortical brain regions revealed by LORETA with activation patterns revealed by positron emission tomography (PET) or functional magnetic resonance imaging (fMRI) (i.e. validation of LORETA for imaging memory functions). Concerning memory storage/consolidation our aim is to replicate that (1) the selected implicit and explicit memory tasks are consolidated during sleep, (2) both, early slow wave sleep (SWS) and late rapid eye movement (REM) sleep are required for overnight learning (two-step model of memory consolidation) and (3) SWS is more important for explicit, REM sleep for implicit memory consolidation. Furthermore, we want to test the hypotheses that (4) delta and spindle waves contribute locally to memory consolidation, (5) cortical activation of theta power changes experience-dependently during REM sleep and (6) sleep after learning is strongly related to a change in task performance even after a period of one week (i.e. memory consolidation is initiated in the night after learning). Thus, the aim of this project is to investigate when and where memories are consolidated during sleep. An important aspect is to integrate the expertise of three different Austrian research centers: the Department of Neurology (University of Vienna) is specialised in multi-channel sleep EEG-recordings, the Department of Psychiatry (University of Vienna) is specialised in quantitative EEG analysis (including LORETA) and the Department of Physiological Psychology (University of Salzburg) in ERD/ERS and memory research. Additionally, all partners have much experience in national and international scientific co-operations.

This project intended to provide neurophysiological evidence for the hypothesis that sleep serves to consolidate memories. For this purpose we selected two different memory tasks, for which behavioural data had demonstrated that, as compared to a corresponding retention interval of wakefulness, sleep enhances retrieval. Implicit procedural motor memory was tested with a mirror-tracing task and explicit declarative memory with a paired-associate word list task. 48 healthy young students were trained on one of these two tasks in the evening and performance was tested before and after sleeping in the sleep laboratory, and in a one- week follow-up. Brain activation was recorded by multichannel-electroencephalography (=EEG) during the encoding and the retrieval sessions and while subjects were sleeping in the sleep lab. An important aspect was to integrate the expertise of three different Austrian research centers in the analysis of the EEG: the Department of Neurology (University of Vienna) is specialised in multi-channel sleep EEG-recordings, the Department of Psychiatry (University of Vienna) is specialised in quantitative EEG analysis ("LORETA") and the Department of Physiological Psychology (University of Salzburg) in Event Related Desynchronisation/Synchronisation (ERD/ERS) and memory research. Results support the hypothesis of a consolidation process which occurs during sleep. For the declarative task overnight changes of words recalled were found to be significantly correlating with stage 2 spindle activity (special features in the EEG). Accuracy in the implicit mirror tracing task was significantly improved after one night of sleep. Improvements in the speed component of the mirror tracing task were highly related to increases in the percentage of REM. Low-resolution brain electromagnetic tomography (LORETA) revealed that in the night following intensive declarative learning cortical reactivation exclusively occurs during the short but frequently recurring episodes with sleep spindles. The findings are consistent with the hypothesis that a repeated activation of hippocampocortical and thalamocortical networks provides the basis for the reorganization and consolidation of declarative memory and confirm the active role of sleep in synaptic plasticity on a local level. In a follow-on project the study design was extended to after-dinner napping. 88 healthy young students were also trained on a paired-associate word list task or a mirror-tracing task. After that they were allowed to either rest for 90 minutes in a reclining chair or nap within a total bedtime of 90 minutes. Once again we found a very positive effect of sleep on memory consolidation. Performance after napping was superior to the one after the resting condition in both memory tasks.