DYNAMIC REORGANIZATION OF NETWORK TOPOLOGY IN PRIMARY SENSORY CORTICES DURING CROSS-MODAL PROCESSING

Processing diverse sensory modalities in natural environments leads to cross-modal modulation of information processing in the brain. However, the underlying mechanism and direct evidence for cross-modal modulation between primary sensory cortices remain unclear. Here, we characterize the single-unit-level network structure of cross-modal modulation between primary sensory cortices. Using two multielectrode probes, we simultaneously recorded single-unit activity from the mouse primary visual and auditory cortices during visual and/or auditory stimulation. We estimated a directed functional network through pairwise temporal relationships of spike timing by computing cross-correlograms of binned spike trains. When we quantified the network’s graph-theoretical organization, we found that multisensory conditions significantly decreased modularity relative to unisensory conditions, suggesting a shift from intra-modular segregation to inter-modular integration via cross-modal modulation. This inter-modular integration enhanced small-world network properties, characterized by higher local clustering with shorter path lengths. To further investigate how cross-modal modulation shapes information routing, we analyzed the network under different stimulation contexts and found that distinct types of single units routing stimulus information acted as hubs that reorganized dynamically according to the stimulus context. Together, our results indicate that cross-modal modulation of network topology provides a neural substrate for context-dependent routing, potentially serving as a key mechanism for efficient multisensory integration.
Acknowledgements: This work was supported by KIST (26E0021) and the National Research Foundation of Korea (NRF) funded by MSIT (NRF-2021R1C1C2012843). J.-H.H. was supported by MSIT/IITP ITRC (IITP-2026-RS-2022-00156225) and NRF (RS-2024-00415812).