Periodic Reporting for period 1 - MemOpt (Optimising memory: Understanding the role of sleep in selective memory strengthening)
Reporting period: 2022-06-01 to 2024-05-31
An essential function of the brain is to selectively retain important information. Being able to select which information to remember and what to forget allows for flexible updating of behavioural goals. By operating in an adaptive fashion, the brain is able to prioritise in memory the most important and relevant aspects of our experience and filtering out unnecessary and redundant information. For memories to be successfully retrieved over long periods of time, they need to undergo a period of consolidation, whereby memories are strengthened and integrated into long-term storage. Sleep plays a critical role in this strengthening process, and there is evidence that memory strengthening during sleep is selective, based on ‘tags’ set during initial encoding. The overarching goal of this project was to understand the mechanisms that govern memory prioritisation and selective remembering during sleep. This is an important question to address, because maladaptive memory biases and co-occurring sleep problems are present in multiple mental health conditions including major depression and post-traumatic stress disorder, as well in memory disorders such as Alzheimer’s disease. Hence, understanding the mechanisms of selective memory processing is crucial for identifying novel therapeutic targets with potentially wide-ranging benefits. The overall objectives of this project were as follows: 1) Identify a neural tagging mechanism that selects particular memories for strengthening during periods of sleep; 2) Discover how memories are reprocessed during sleep to selectively strengthen high priority mechanisms, and; 3) Assess the impact of selective memory processing during sleep on memory representations. Through the experiments performed, the project uncovered neural oscillations involved in the selection of certain memories for memory consolidation during sleep, supporting the conclusion that there are selection mechanisms in the brain that govern the “fate” of memories across periods of sleep-related consolidation.
The first experiment of the project looked to ascertain whether there are particular brain signatures that are associated specifically with successful memory formation following a period sleep, compared to memories that are remembered across a period of wake. A total of 44 participants to take part in the experiment . Each participant visited the sleep laboratory twice. On one visit, participants learned a set of word-image pairs and had to recall as many as possible after taking a nap. During the other visit, they stayed awake after learning the items. EEG was recorded throughout the experiment. Each visit took approximately 5 hours to complete. I found that theta oscillations (i.e. 4-8Hz rhythmic neural oscillations) presented during initial memory encoding uniquely predicted subsequent memory strengthening across a period sleep, suggesting that this neural activity acts as a “tagging” mechanism that prioritises certain memories to undergo selective strengthening during sleep.
In a second experiment, I isolated the mediating mechanism that performs this selective strengthening. Thirty-one participants to take part in the experiment. Each participant visited the sleep laboratory twice, and made one visit to the York Neuroimaging Centre. During the sleep lab visits, participants again encoded word-image pairs as per WP2. This time however, half of the items were associated with a high-value financial bonus (up to £10 maximum) upon successful recall. The other half were associated with a low-value bonus (up to £1). Their memory was tested before and after a period of sleep in the lab. I recorded high-density EEG throughout the experiment, allowing me to isolate the source of the high and low value reward memories, as well as assess ongoing neural activity during the nap itself. I found that across a period of sleep, a “reward bias” emerged, that is the high-reward memories were prioritised in memory. This reward bias correlated with sleep spindle activity, a waxing and waning neural oscillation and defining neural signature of sleep.
The work from this project has been disseminated via peer-reviewed articles, presentations at international conferences, and invited guest lectures at multiple universities.
In a second experiment, I isolated the mediating mechanism that performs this selective strengthening. Thirty-one participants to take part in the experiment. Each participant visited the sleep laboratory twice, and made one visit to the York Neuroimaging Centre. During the sleep lab visits, participants again encoded word-image pairs as per WP2. This time however, half of the items were associated with a high-value financial bonus (up to £10 maximum) upon successful recall. The other half were associated with a low-value bonus (up to £1). Their memory was tested before and after a period of sleep in the lab. I recorded high-density EEG throughout the experiment, allowing me to isolate the source of the high and low value reward memories, as well as assess ongoing neural activity during the nap itself. I found that across a period of sleep, a “reward bias” emerged, that is the high-reward memories were prioritised in memory. This reward bias correlated with sleep spindle activity, a waxing and waning neural oscillation and defining neural signature of sleep.
The work from this project has been disseminated via peer-reviewed articles, presentations at international conferences, and invited guest lectures at multiple universities.
Prior to this project, it was unknown the mechanisms by which the brain selects relevant information for preferential memory strengthening during sleep. I have found the first evidence of such a mechanism in place that governs a memory’s subsequent evolution. These results could be exploited through therapeutic interventions that seek to modify existing memory biases, and potentially even incorporating sleep into existing therapies. For example, therapies that seek to modify maladaptive memory biases in depression might benefit from a nap following the therapy session, where sleep will adaptively consolidate the newly learned memory strategies. In later analyses, I expect to find that highly rewarding memories are reactivated more often during sleep, and that non-invasive techniques can be used to modify what gets reactivated in the sleeping brain. This could have implications for sleep-based interventions to improve memory.