Oxford Interdisciplinary Bioscience Doctoral Training Partnership

点击翻译按钮获取中文摘要

Sleep has proven to be an important component of memory consolidation and generalization. Advancing our knowledge of the relationship between sleep and these important neural computations is advantageous to both basic and clinical neuroscience. Research suggests, however, that there may be important differences between sleep in the lab and in a naturalistic, at-home setting. In this thesis, we test the feasibility of conducting research into these topics remotely, in participants' homes. In chapter 2, we test online motor (serial reaction time task; SRTT) and declarative (paired-associates) memory tasks to test whether they can be learned in an at-home setting and whether they benefit from sleep compared to wake. We found that while these memory tasks could be learnt remotely and online, they showed no significant benefit of sleep over wake. In chapter 3, we asked whether task improvement correlated with features of sleep. Using a portable, low density electroencephalography (EEG) device (Dreemband) we explored the covariation between task performance and sleep features as measured by the EEG. We found a significant covariation between spindle-Slow Oscillation coupling and improvement on the SRTT, but no significant relationships between sleep features and performance on the paired-associates task. In chapter 4, we tested whether an underlying sequence structure could be generalized from one online task to another and whether this benefitted from sleep. We found no evidence of generalization and so developed a novel task to probe sleep-associated generalization in the future. In summary, this thesis established online motor and declarative tasks which can be learned remotely, it found that sleep features collected at-home mirror in-lab findings and set up a generalization task that can be learned remotely, although whether it benefits from sleep is yet to be tested.