Hippocampal theta-gamma phase-amplitude coupling in memory encoding and retrieval
[eng] Deciphering the mechanism of human memory formation and retrieval is one of the still unsolved mysteries of neurosciences. Theoretical models of underlying mechanisms of memory formation and retrieval have led the memory field for the past decade (Hasselmo, Bodelon, & Wyble, 2002). However...
| Autor: | |
|---|---|
| Formato: | tesis doctoral |
| Estado: | Versión publicada |
| Fecha de publicación: | 2024 |
| País: | España |
| Recursos: | Universidad de Barcelona |
| Repositorio: | Dipòsit Digital de la UB |
| OAI Identifier: | oai:diposit.ub.edu:2445/211951 |
| Acesso em linha: | https://hdl.handle.net/2445/211951 http://hdl.handle.net/10803/691119 |
| Access Level: | acceso abierto |
| Palavra-chave: | Neurociències Memòria Record (Psicologia) Hipocamp (Cervell) Neurosciences Memory Recollection (Psychology) Hippocampus (Brain) |
| Resumo: | [eng] Deciphering the mechanism of human memory formation and retrieval is one of the still unsolved mysteries of neurosciences. Theoretical models of underlying mechanisms of memory formation and retrieval have led the memory field for the past decade (Hasselmo, Bodelon, & Wyble, 2002). However, with methodological and technological limitations these models have yet to been shown in humans, especially deep brain areas such as the hippocampus, traditionally not easily recorded with non-invasive techniques. It has been reported in rodent studies, that theta-gamma interactions modulate memory processes and that phase-states and interactions between oscillations may be of importance for the successful formation and recall of episodic memories (Manns, Zilli, Ong, Hasselmo, & Eichenbaum, 2007). It is well accepted that memory formation relies on the coupling and interaction of theta-gamma oscillations (Tort, Komorowski, Manns, Kopell, & Eichenbaum, 2009) in humans, but phase-preference mechanisms similar to those observed in rodents have yet to be found in humans. According to the latest evidence in human studies, memory formation of humans may rely on differential communication between brain areas (Griffiths et al., 2019) and to different phase states in the hippocampus (Kerren, Linde-Domingo, Hanslmayr, & Wimber, 2018). Here, we are aiming to advance the understanding of how theta-gamma interactions in the human hippocampus may be reflective of underlying mechanisms that could differ between encoding and recall states, and how they could predict the success of memory formation and retrieval. We conducted an iEEG study on intracranially implanted epileptic patients, in order to analyse direct recordings from the hippocampus and relate them to an episodic sequential memory task. We believe that the analysis of theta-gamma interactions in the hippocampus are key to differentiating mnemonic processes and predicting successful memory retention. In Study 1, we describe the Mean Opposition Vector Length (MOVI) a new method to analyze phase-preference differences between neurophysiological datasets. This method was developed during this thesis. We have tested it on simulated synthetic data in order to really understand its potential and limitations, and have compared it with the Jensen Shannon Divergence, a more established method of analysis of phase-preference opposition between distributions. To assess the validity of our new index, we compared MOVI and JSD using Matthew’s Correlation Coefficient (MCC) that aids in determining the accuracy of a test when the outcome is known. We found that our new index is more sensitive than the established techniques, more resistant to noise, and detects specific variations in the data that other methods ignore. In this way we proposed a novel technique of analysis for assessing distribution opposition between datasets that can be used in future studies on phase-preference. In Study 2, we explored neural signatures of epileptic patients implanted with deep brain electrodes. During the recording of their brain activity, they performed a task of encoding and recall of series of congruent images to emulate life-like episodic memories. We first explored the specific neural signatures found during encoding and recall in our sequential task by analyzing increases and decreases in power via a time-frequency analysis. Then we explored relationships between frequencies by analyzing theta-gamma phase-amplitude coupling. Finally we explored theta-gamma phase preference during encoding and recall. We have shown that theta-gamma interactions in the hippocampus are predictive of memory processes and can differentiate between encoding and recall. More importantly, we found that this phase-coupling preference was predictive of the successful retention of sequential mnemonic traces. Our results from these two main studies suggest that the study of phase amplitude coupling (PAC) and phase opposition is essential to research that focuses on the mechanistic processes underlying memory formation and retrieval. Additionally, we bridge the gap with rodent studies and show a mechanism underlying mnemonic processes so far never observed in humans, but consistent with previous findings on the communication between different brain areas. Altogether these findings shed a light on the neural mechanisms that support memory formation, early recall, and delayed recall, and on the different analytical methods that can be used to assess phase-amplitude interactions. This advances the understanding of human memory, of the neurophysiological mechanisms underlying different processes, and on the analytical methods used to understand them. |
|---|