Report

Teaser, summary, work performed and final results

Periodic Reporting for period 2 - NEUROMEM (A Neurocomputational Model of Episodic Memory)

Teaser

Our memories define us, and their disruption in psychiatric and neurological conditions can be devastating. However, how we are able, e.g., to remember our wedding day and re-imagine the scene that was around us, remains one of the great mysteries of the human mind. NEUROMEM...

Summary

Our memories define us, and their disruption in psychiatric and neurological conditions can be devastating. However, how we are able, e.g., to remember our wedding day and re-imagine the scene that was around us, remains one of the great mysteries of the human mind. NEUROMEM is an integrated experimental and computational attempt at a fundamental breakthrough in this problem. Building on recent insights into how environmental location and orientation is encoded by neurons in the mammalian brain, I aim to develop a mechanistic understanding of how events we experience are stored, recalled and imagined, i.e. a neurocomputational model of how specific memories result from patterns of activity in neuronal populations. NEUROMEM will provide mechanistic answers to 3 long-standing questions: 1) What is the link between memory and space, and role of spatial context in re-imagining episodes? 2) How are the multiple diverse elements of complex life-like events recollected together? 3) How can remembered events be read-out as visuospatial imagery? Work will comprise psychological and functional neuroimaging experiments using sophisticated designs including use of virtual reality, and corresponding simulations of how such behaviour can be driven by neuronal activity. The computational modelling will directly contact neurophysiological data such as the firing of place and grid cells in the hippocampal formation, and provide quantitative behavioural predictions, while neuroimaging provides a read out of population activity during this processing in the human brain. NEUROMEM will generate new hypotheses and explanations at the cognitive level, of interest to all scholars of the complexity of the human mind, and allow neurophysiological interpretation of behavioural data - providing a vital link between cognitive theory and neuroimaging and neurological data. Its implications extend beyond memory, including the mechanism for imagining views that have not been experienced.

Work performed

NEUROMEM aims to investigate the neural mechanisms that support the spatial structure of memories and the way in which the context and content of an experience are bound together and later remembered. We have carried out a number of experimental studies and complementary computational modelling work to provide a comprehensive account of the neural machinery that supports human memory function.

We have provided a detailed model of spatial memory and imagery, incorporating representations of objects into egocentric parietal and allocentric medial temporal representations to combine the content and context of an experience within flexible representations (Bicanski and Burgess, 2018). This new model offers an account on how the brain stores complex representations in memory, via associating the content of an experience with the surrounding context, and can flexibly use these representations to generate imagery to guide future behaviour. This model also incorporates the provision for imagined movement via connections between grid cells in entorhinal cortex and hippocampal place cells. The model is important in consolidating our understanding and making predictions about how memories are formed, retrieved and updated within a complex system of brain regions.

To investigate and model the interactions between place cells and grid cells to support dynamic imagery and memory representations, we have recently proposed a model of recognition memory in which grid cells encode translation vectors between features of an attended stimulus and thus guide eye movements between expected features to accumulate evidence to identify experienced stimuli (Bicanski and Burgess, 2019). The model is timely in providing a neural account of recent empirical data demonstrating the grid cells show responses coupled with eye movements in non-human primates and humans.

To examine the associative structure of long-term memory and its reactivation at retrieval via hippocampal pattern completion, we have provided an account of how negative experiences can affect memory and contribute to memory disturbances often seen in posttraumatic stress disorder (PTSD; Bisby and Burgess, 2017). This review consolidates the current empirical evidence and makes novel predictions about how negative experiences can affect amygdala- and hippocampal-dependent memory in opposing ways, strengthening memory for the negative content and weakening the associative and contextual structure of the memory representation. The result of these opposing interactions are thought to support distressing intrusive imagery in PTSD. We have expanded this work further, carrying out behavioural experiments to assess the associative structure of negative events in memory and the way in which they are retrieved. The results of this work support our initial proposals, that negative events weaken the associative structure of memory representations to disrupt pattern completion processes leading to memories being retrieved in a less holistic manner (Bisby et al., 2018). This work also utilises computation modelling to account for the neural mechanisms how negative events can alter hippocampal function to reduce memory coherence via a Hopfield model of associative learning in the hippocampus.

In collaboration with research projects with other funding streams, we have also assessed the way in which sequence information is incorporated within memory representations. To do this, we have combined a sequential motor learning task with magnetoencephalography (MEG) to investigate the temporal dynamics of brain areas required to retrieve sequence information associated with action responses (Kornysheva et al., 2019). We show that the brain learns and controls multiple sequences of complex actions by flexibly combining distinct representations of the actions required, the interval timing and the sequence position. In addition, we have examining oscillatory changes within the hippocampus, utilising desktop vi

Final results

NEUROMEM has produced a number of studies that are progressive and beyond the state of art in both the techniques they utilise and the results that they have produced. For example, Bicanski and Burgess (2018) provides the only neural systems model of spatial imagery and memory that explains cognition and behaviour at the level of interacting populations of neurons in multiple brain regions - this makes a series of new predictions for fMRI and behaviour (and electrophysiology neuropsychology). Bicanski and Burgess (2019) provides the only model so far linking visual behaviour (saccadic eye movements) to both the neural level (grid cell firing patterns) and behaviour in terms of memory and attention. We have also provided a novel behavioural paradigm for looking at binding within multi-element episodic events, and how this relates to negative emotional experience and fMRI activity in multiple brain areas (Bisby et al., 2018; Bisby and Burgess, 2017). The results of this work provide valuable information on the role of the hippocampus in spatial memory storage and retrieval and how it is affected by psychological issues such as the experience of negative events, with implications for psychological disorders such as generalised anxiety disorder or posttraumatic stress disorder. We have combined our virtual reality spatial memory tasks with fMRI to provide novel insights into memory processing in controlled but pseudo-realistic situations (Suarez-Jimenez et al., 2017). In collaboration with projects funded by other sources we have also combined our memory tasks with other methodologies (MEG, iEEG) to provide further novel insights into the neural mechanisms involved (Bush et al., 2017; Kornysheva et al., 2019).

For the remaining period of the project, we will continue to use novel approaches in behavioural tests, neuroimaging and computational modelling and advanced analysis methods to understand the brain systems involved in human memory. We envisage that future studies within the project will generate results that have further our understanding of the role of the hippocampus in memory storage and retrieval and how sub-regions differently contribute to these processes. One particular focus will be on finding out how the spatial system (comprising place cells, grid cells etc) supports sequential memory - a seemingly non-spatial function which is critical to episodic memory. We also expect to further our previous work on how negative events can influence the brain systems involved in memory and how alterations might lead to intrusive memories, a prominent feature in posttraumatic stress disorder.