Quantify how connectivity features beyond EI cell-type structure influence network dynamics

Determine the extent to which connectivity features beyond the excitatory–inhibitory cell-type-specific mean connectivity—specifically, second-order synaptic connectivity motifs such as chain, reciprocal, convergent, and divergent motifs—affect the dynamics of recurrent excitatory–inhibitory networks, in contrast to classical models that assume independent synapses.

Background

Classical EI network models typically assume synaptic strengths are independent conditioned on cell type, enabling mean-field reductions. Recent anatomical datasets, however, show robust higher-order structure in local connectivity, including over-represented pairwise motifs that violate independence. Establishing how much these additional connectivity features alter stability and responses remains crucial for interpreting perturbation experiments and for extending balance theories beyond first-order statistics.

References

While decades of influential research have demonstrated the strong role of the basic EI cell type structure, to which extent additional connectivity features influence dynamics remains to be fully determined.

— Identifying the impact of local connectivity patterns on dynamics in excitatory-inhibitory networks  (2411.06802 - Shao et al., 2024) in Introduction

Whether such an effective representation can reliably predict local dynamics remains an open question and will require validation against full-network simulations and established reduction methods.

— Reducing Boolean Networks via Analysis of Dynamic Network Subgraph Behavior  (2608.19292 - Zakeri et al., 19 Aug 2026) in Discussion section, paragraph beginning “The present catalogue does not yet establish a direct reduction method for large Boolean networks”