C03
The ability to anticipate and plan events in a certain order is essential for many daily activities. Our goal is to reveal trainable and transferrable representations of serial order for memory, navigation, and skilled movement in the human brain, and identify their reserve potential in the face of tau pathology in medial temporal lobe. Following a multidimensional approach including magnetoencephalography, transcranial magnetic stimulation, and PET imaging, we will uncover macroscopic circuits representing serial order that are shared across domains in young and elderly individuals.
Principal Investigators
Dr. Elena Azañón
Dr. med. Max-Philipp Stenner
Co-Workers
Anwesha Das
Alexandros Karagiorgis
Serial order in memory & action
The contents of our memories, and the movements that constitute our actions, often follow a certain serial order. Despite the fundamental role of serial order for memory and action, it is unclear how neurons encode order. At a behavioral level, memory and action are subject to the same characteristic ordering errors. This could indicate that memory and action follow similar principles of serial order encoding, or even share part of the neuronal circuitry involved in serial order processing, with important implications for reserve and transfer. Can motor sequence training alleviate the deficits in serial order memory typically observed at old age?
Domain-general encoding of serial order?
Many models of serial order encoding follow the principle of competitive queuing, which can give rise to benchmark serial order memory phenomena. Competitive queuing models assume a parallel planning layer, where representations of multiple items are active simultaneously, and a competitive choice layer, where items compete for selection based on current priority. Compatible with competitive queuing, Kornysheva et al. (2019) identified in human magnetoencephalography (MEG) data a rank-sensitive gradient of simultaneous representations of several upcoming movements in a motor sequence before sequence initiation, source-localised to the parahippocampal area. Preliminary results in our labs indicate a similar gradient when humans retrieve sequences of tones from memory, in the absence of any sequential movements. Besides this shared signature, previous studies point to a potential overlap in macroscopic circuits encoding serial order processing for memory vs. action, specifically in prefrontal cortex and medial temporal lobe (Averbeck et al., 2002; DeVito & Eichenbaum, 2011). This is important because partially overlapping circuits encoding serial order for memory and action could provide a substrate for transfer across domains.
An early marker of pre-clinical tau pathology?
The ability to remember sequences of items in a specific order declines as individuals age (Allen et al., 2015). There is some evidence from neuropsychology indicating that deficits in serial order memory might serve as an early indicator of Alzheimer’s disease, and help distinguish Alzheimer’s disease from frontotemporal lobar degeneration and normative ageing (Bellassen et al., 2012). To establish deficits in serial order processing as an early marker of pathology, a link between serial order deficits and tau pathology, in particular at pre-clinical stages, is needed. This link should consider serial order for memory and action separately, as a step towards understanding the reserve potential of any shared circuitry for serial order processing.
Aim of the project
Our goal for the second funding period is to determine the domain-generality, transferability, and neural circuitry of competitive queuing gradients for serial order representation, together with their vulnerability to pathology. Specifically, we aim to reveal to what extent circuits that encode serial order are shared between action and memory, whether training effects transfer across domains, and to what extent pre-clinical tau pathology in medial temporal lobe impacts on serial order representations for memory and action.
Our approach
Our approach combines human machine learning with human MEG data obtained during motor sequence tasks and serial order memory tasks. In addition, we will conduct a training intervention study across several days, using an established protocol that improves the planning of sequential movements (Ariani et al., 2021), and assess transfer of training effects to serial order memory performance, as well as the associated competitive queuing gradients in MEG before and after training. Finally, we will quantify tau pathology in medial temporal lobe of healthy elderly via positron emission tomography and assess its effects on serial order memory tasks.
Allen, T. A., Morris, A. M., Stark, S. M., Fortin, N. J., & Stark, C. E. L. (2015). Memory for sequences of events impaired in typical aging. Learning and Memory, 22(3), 138–148. https://doi.org/10.1101/lm.036301.114
Ariani, G., Shahbazi, M., & Diedrichsen, J. (2023). Cortical areas for planning sequences before and during movement. BioRxiv . https://doi.org/doi: https://doi.org/10.1523/ENEURO.0085-21.2021
Averbeck, B. B., Chafee, M. V., Crowe, D. A., & Georgopoulos, A. P. (2002). Parallel processing of serial movements in prefrontal cortex. Proceedings of the National Academy of Sciences, 99(20), 13172–13177. https://doi.org/10.1073/pnas.162485599
Bellassen, V., Iglói, K., de Souza, L. C., Dubois, B., & Rondi-Reig, L. (2012). Temporal order memory assessed during spatiotemporal navigation as a behavioral cognitive marker for differential Alzheimer’S disease diagnosis. Journal of Neuroscience, 32(6), 1942–1952. https://doi.org/10.1523/JNEUROSCI.4556-11.2012
DeVito, L. M., & Eichenbaum, H. (2011). Memory for the order of events in specific sequences: Contributions of the hippocampus and medial prefrontal cortex. Journal of Neuroscience, 31(9), 3169–3175. https://doi.org/10.1523/JNEUROSCI.4202-10.2011
Kornysheva, K., Bush, D., Meyer, S. S., Sadnicka, A., Barnes, G., & Burgess, N. (2019). Neural Competitive Queuing of Ordinal Structure Underlies Skilled Sequential Action. Neuron, 101(6), 1166-1180.e3. https://doi.org/10.1016/j.neuron.2019.01.018
Our hypotheses
We expect that serial order processing for memory vs. action involves partially overlapping macroscopic circuits in the dorsolateral prefrontal cortex and medial temporal lobe, and a shared neuronal population signature – a parallel, order-sensitive representation of upcoming (anticipated) events in a sequence, consistent with competitive queuing. In addition, we hypothesize that there is transfer of improvements in serial ordering of movements during motor sequence training to serial order memory. Finally, we expect elderly individuals with high tau accumulation in medial temporal lobe to exhibit diminished competitive queuing gradients, and reduced serial order memory performance, compared to age-matched individuals with low tau concentrations.