Human hippocampal theta and high-gamma oscillations in spatial encoding and consolidation in a virtual Morris watermaze task
thesisposted on 28.03.2022, 00:56 by Yi Pu
The formation of spatial memories has been proposed to proceed in two stages. In the initial stage, which occurs during active navigation in a new environment, hippocampal cell assemblies are activated to encode spatial information. The activation of encoding assemblies is accompanied by low frequency theta band neuronal oscillations. In the second stage, which occurs during rest or sleep, hippocampal assemblies activated during the encoding phase are reactivated to consolidate the newly formed but labile memory traces. These reactivations are accompanied by high frequency neuronal rhythms. In Buszaki's (1989) two-stage model of spatial learning, theta rhythms are proposed to provide a mechanism to bind sequential hippocampal place-cell assemblies over time; while the high-frequency oscillations are hypothesized to potentiate and consolidate the sequential activation of cell assemblies. These oscillatory mechanisms are well-established in animal models, but evidence in the human hippocampus is lacking. In the current thesis, I aimed to bridge this gap between animal and human models of spatial memory formation. I used non-invasive magnetoencephalography (MEG) recordings, and a virtual Morris water maze (vMWM) task to investigate whether human hippocampal low and high frequency oscillations play roles in different stages of spatial leaning. In Chapter One, I briefly review research on the functions of hippocampal rhythms and put forward the research questions I aimed to address in the thesis. In Chapter Two, I review evidence and make the case that MEG is a suitable technique for addressing these questions. Based on this, in Chapter Three, using MEG I examined whether low frequency human hippocampal theta oscillations play a role in spatial encoding during virtual navigation. Consistent with previous work, the results showed left hippocampal theta power increase during goal-directed navigation, supporting the contention that MEG can reliably detect and localize human hippocampal signals. Further, my analyses showed that right hippocampal theta oscillations were modulated by environmental novelty and were correlated with navigation performance, providing strong support for the idea that theta plays an important role in environmental encoding during navigation. In Chapter Four, I analysed the same MEG dataset to examine high frequency gamma activity during the inter-trial rest periods. The results confirmed significantly increased right hippocampal high-gamma during the inter-trial period in the new environment relative to the familiar one; and that gamma was positively correlated with theta power measured during navigation; and also with subsequent navigation performance in the familiar environment. In Chapter Five, I examined theta and high-gamma oscillations in a group of age-matched females, and compared these to the male data described in chapters three and four. Since there are clear and well-established gender differences in spatial behaviour, this study was designed to determine if these are reflected in the neurophysiological measurements. Consistent with the previous literature, the behavioural results showed clear gender differences. Males scored higher on a psychometric test of spatial ability, were faster in navigating the vMWM, and showed significant speed improvements in familiar versus new maze environments, while females did not improve. The MEG analyses confirmed corresponding gender differences in both the theta and gamma rhythms, strongly reinforcing the functional importance of these two rhythms in spatial learning. In the concluding Chapter Six, I summarize the results and discuss how they contribute to our understanding of the neurophysiological mechanisms of memory formation in the human brain. I conclude that MEG provides sensitive, reliable, and behaviourally relevant measurements of human hippocampal function during spatial navigation. Consequently, MEG is a crucially important technique for bridging the gap between animal and human models of hippocampal function. Future developments in MEG sensor technology are likely to further enhance its sensitivity and utility in this regard.