Updating of memories during memory consolidation

Summary

Animals, including humans, learn by exploring the environment and assessing the predictive value of a particular experience by trial and error. This basic form of learning, called associative conditioning, enables animals to adaptively respond to sensory signals that are associated with reward or punishment. By learning such an association, an animal can more effectively locate and exploit food resources or avoid threats.

Following the initial formation of an associative memory there is a period of consolidation, during which the memory becomes progressively more permanent. We showed that during consolidation there are brief periods of amnesia during which the progression to long-term memory storage can be blocked or altered by unanticipated changes in the sensory environment. Recently we provided further evidence for this flexibility of consolidation by showing that a more important or recent experience during these lapse periods can fundamentally change the fate of the original association. Indeed it can result in the replacement of the first association with a second one. Although lapses may look like non-adaptive deficits in memory storage, we claim that they are important in providing opportunities for alteration of the memory trace. The updating of the memory at lapse points has the advantage of preventing the costly process of consolidation of an obsolete memory while providing opportunity for new or more important experience to be incorporated into or even replace the original memory.

The aim of this proposal is to advance our understanding of this fundamental type of memory updating that results in the swapping of one memory by another.

To achieve this, we will exploit the advantages offered by the relatively simple brain and behaviour of the mollusc Lymnaea. While having a CNS of only ~20,000 neurons, this animal nevertheless shares all of the basic features of associative memory formation displayed by far more complex animals, including humans. This allows us to study a universal form of behavioural adaptation and flexibility at a level of cellular and molecular detail that would not be possible in more complex animals. Because we know that the process of memory formation and its underlying neural and molecular mechanisms are evolutionarily highly conserved, this research is likely to reveal principles and mechanisms of behavioural flexibility that apply equally to both simple and complex animals.

In the longer term this research may lead to the discovery of new drugs or procedures that can modify the flexibility of memory formation including the potential benefit that this may have for the treatment of memory disorders including post-traumatic stress disoder (PTSD). In PTSD the memory of a traumatic event becomes very strong and almost unsusceptible to interference. All the attempts to supress this memory have targeted the reconsolidation process that can be evoked long time after the stressful experience when the memory has already been stored for the long-term. Our work on memory blocking and replacement during the process of consolidation, may offer alternative opportunities for fundamental research in this clinical area.

Reports of temporary amnesia (or lapses) during memory consolidation are widespread in animals, including humans, but it is not clear whether these lapses have a functional role. Previously it was thought that lapses simply represent vulnerable points in memory expression. However, we have shown recently that they provide opportunities for the alteration in the trajectory of memory consolidation in response to unpredictable changes in the sensory environment and thus serve as choice points for updating of memory traces. This flexible feature of memory systems results in blocking of one memory and its replacement by a second memory following a recent and more salient sensory experience. In our example from Lymnaea, the animal changes its feeding preference from one type of chemical to another when a second kind of training is applied at a lapse point. We know that lapses are inevitable consequences of transitions, when a memory is being transferred from its dependence on one underlying molecular mechanism to another but we know nothing about how this leads to memory replacement. Our aim is to understand at the behavioural, electrophysiological and molecular levels how re-training at lapse points allows a more recent memory to be formed at the expense of an earlier memory. We have developed a number of different preparations in Lymnaea, that allow single-trial associative training of snails 'in the dish' and simultaneous recording of neurons in the learning circuit, and molecular analysis of single key neurons at various stages of memory formation and replacement. We will test the hypothesis that memory replacement is a competitive process based on inhibitory synaptic interactions between sensory pathways and neurons underlying alternative memories. There is a general realization that memory formation and retention are highly labile processes and this project will provide fundamental information on what underlies this flexibility at the cellular and molecular levels.

1. Academic community: Our recent research shows that memory lapses are not simply faults in memory formation but they provide opportunities for the alteration in the trajectory of memory consolidation. We have shown that lapses serve as choice points for the updating of memory traces allowing modification or replacement of the original memory. This is a new concept that would create high interest in the field of learning and memory research and our findings on the mechanisms that potentially allow replacement of a memory during its consolidation will be instructive to the wider academic community working on brain mechanisms in general. In terms of capacity-building, the PDRAs and research technician will greatly benefit from learning the advanced, multi-disciplinary research approaches utilized in the project. (i.e. elctrophysiologycal manipulation of synaptic connections and patterns of activity by using DynamicClamp, behavioural pharmacology, histology and molecular biology techniques).
2. Public Health Sector: The benefit here will arise from the long-term impact of our work on understanding how memory consolidation works and can be altered at specific time points. There are a number of conditions such as depression, dementia and Post-Traumatic Stress Disorder (PTSD) that affect our ability to learn and remember. Our research shows that lapses of memory recall during consolidation can provide opportunities for interference of an emerging memory trace and decoding the mechanism can lead to novel treatment of these mental disorders and can contribute to improvement of public health. At the fundamental level of neural networks and neurotransmitters this is a very poorly understood area of significant clinical importance. Indeed considerable costs are associated with less than fully effective treatments for these conditions. Our research, using a model invertebrate system, will answer some of the most fundamental questions about the brain's control of actions - what neural architectures and neurotransmitter systems are involved and how? This research, carried out at a high level of neuronal resolution, will help the interpretation of much lower resolution information available to clinicians dealing with affected patients. Periods when lapses of memory occur after learning could be of targets to medical intervention providing windows of high sensitivity to treatments.
3. Commercial Private Sector: Two distinctly different commercial areas have an interest in our research on simpler neural networks in Lymnaea - the pharmaceutical industry and the IT sector (including robotics). Eli Lilly have supported our work on the Lymnaea acetylcholine binding protein and we have initiated collaboration with Gedeon Richter Pharmaceuticals to use the Lymnaea system to screen memory enhancing drugs. In the IT sector a number of companies (NaturalMotion Ltd, NeuroRobotics Ltd, MultiChannelSystems MCS GmbH) have expressed an interest in our work on simple model nervous systems because an understanding of how neural networks generate adaptive behaviour can inspire new and more powerful artificial networks for a variety of applications, e.g. in entertainment (games and graphics) and robotics (autonomous robots).
4. Wider Public: Public understanding of science issues - we engage in school 6th form lectures, popular articles and books explaining neuroscience research, contribute to local Café Scientifique meetings and the Brighton Science Festival. We will also continue to publish articles in the online journal Scholarpedia which is a refereed online medium for the general dissipation of knowledge (e.g. Benjamin & Kemenes 2010).
Animal welfare & issues of public concern: By conducting our experiments on an organism not covered by the A(SP)A 1986, our research will have a positive impact on the principles of the 3Rs - Replacement, Refinement and Reduction, principles adopted by all research councils and major charitable funding bodies.