Yuxiu Shao, Bridging Network and Circuit models: a low-rank approach

by Yuxiu Shao (Beijing Normal University (China))

Wednesday Oct 8, 2025, 10:30 AM → 11:30 AM Europe/Paris

Understanding how connectivity structure shapes network dynamics is paramount in the field of neuroscience. Theoretical investigations of multi-population neuronal networks often consider statistically homogeneous populations and incorporate either only the population-averaged mean or i.i.d. fluctuations in synaptic couplings. Advanced connectome dataset from multiple species highlighted the strong presence of motifs – specific connectivity patterns between pairs and triplets of neurons–beyond the scope of mean connectivity.  However, it is a priori not clear which of the experimentally identified connectivity motifs exert a strong influence on neural dynamics. While most previous works focused on reciprocal motifs, here we show that another feature of connectivity, chain motifs, has a much stronger impact on the dynamics of neural activity.

We compared the effects of chain and reciprocal motifs within two-population excitatory-inhibitory networks using an analytical framework that approximates the connectivity in terms of low-rank structures that incorporate motifs. We mathematically derived  the dominant eigenvalues and characterized the statistics of corresponding eigenvectors. We then used these results to perform a low-rank approximation that predicts the effects of connectivity motifs on linear network dynamics.

Our results show that chain motifs have a much stronger impact on dominant eigenmodes than reciprocal motifs. Moreover, an over-representation of chain motifs induces an additional eigenmode with an eigenvalue of sign opposite to the dominant one, thus modifying the network's effective rank. This additional eigenmode substantially influences network dynamics, offering a new perspective on how local EI motifs shape the network’s excitability. Our exploration of the physiological connectivity dataset for the first time revealed the significant impact of EI chain motifs on altering the network's effective rank, permitting the discovery of richer dynamics associated with these specific connectivity motifs.