Motor skill learning in patients with Obstructive Sleep Apnoea (OSA): the effect of intervening sleep and sleep fragmentation

There is substantial evidence that sleep benefits the consolidation of memory. This concept can be exemplified by explicit motor sequence learning tasks. On such tasks individuals demonstrate gains in performance over training (‘online’ improvements) as well as further spontaneous ‘offline’ enhancements in performance (in the absence of further training). These ‘offline’ enhancements are typically only exhibited after a retention interval composed of sleep, as such these tasks are thought to demonstrate sleep dependent memory consolidation. Although patients with Obstructive Sleep Apnoea (OSA) have been shown to demonstrate impaired outcomes on these motor learning tasks, it is unclear whether this impairment is associated with an ‘online’-practice based learning impairment, or is due to a specific deficit in ‘offline’ sleep dependent learning processes. In addition it has yet to be determined whether such deficits can be rectified with a treatment intervention. As such, this thesis aimed to determine the nature of this impaired learning outcome in OSA patients, establish whether this outcome can be normalized with treatment via Continuous Positive Airway Pressure (CPAP), and to investigate one of the likely casual mechanisms, the degree of sleep fragmentation. Compared to matched controls, patients with OSA demonstrated impaired motor sequence learning specifically associated with an absence of overnight/offline enhancements in performance, but intact online-learning over training. These overnight enhancements in explicit motor sequence learning were then shown to normalize (relative to controls) in patients with compliant CPAP usage. The extent of these overnight gains in motor learning were also shown to be negatively associated with the degree of sleep fragmentation as characterised by the Arousal Index. Subsequent analyses however suggested that these overnight gains in motor learning, demonstrated by healthy controls and CPAP treated patients, occur largely over a 10 minute rest period immediately following training (prior to sleep onset). Having demonstrated deficient offline learning can be overcome with the successful treatment of OSA, this thesis attempted to investigate these outcomes in healthy controls using an experimental sleep fragmentation design. Consistent with previous findings of this thesis, and divergent from current theoretical models of sleep dependent procedural learning, offline gains in motor learning were demonstrated to occur over a short post-training rest interval. Subsequent motor performance was shown to remain equally stable (with no additional offline gains) across longer retention intervals containing either uninterrupted or experimentally fragmented sleep, as well as intervals solely containing wakefulness. In recent years an increasing body of literature has demonstrated that offline learning processes are impaired in a wide range of sleep and psychiatric conditions. This thesis presents a systematic investigation of these outcomes in patients with OSA, and provides novel findings demonstrating the reversibility of this impairment with CPAP treatment. Of wider interest were the findings resulting from key methodological alterations to an established explicit motor sequence learning paradigm. These findings show that a theoretical model based solely on sleep dependent learning cannot provide full explanation for the offline improvements observed in procedural learning, or explain the deficits in this process exhibited by patients with OSA.