bioRxiv · 2026 · Preprint

Sparse cortical dynamics reveal flexible condition-dependent spike-order codes

Guozhang Chen, Wolfgang Maass, Franz Scherr

Combining mouse visual cortex recordings with a data-driven microcircuit model reveals how local connectivity and heterogeneity support flexible spike-order codes.

The scientific question

How do sparse cortical circuits organize spikes into flexible, condition-dependent temporal codes?

Neocortical circuits compute with remarkably few spikes, yet how recurrent wiring avoids rigid temporal pipelines to adapt spike timing remains unclear.

Conceptual overview of sparse, condition-dependent spike-order computation.
Figure 1 from the preprint ↗ · CC BY-NC-ND 4.0

The core idea

Comparing V1 Recordings and Microcircuit Models

We analyze single-trial spike-latency sequences in mouse V1 recordings alongside a task-trained data-driven V1 microcircuit model, performing structural ablations and targeted silencing of high-information early neurons.

What we found

Heterogeneity and Local Wiring Enable Flexibility

Neuronal diversity and distance-dependent local connectivity suppress rigid sequential firing to reproduce condition-dependent spike-order codes, while silencing early informative neurons disrupts low-dimensional task trajectories and decisions.

Continue the question

Related reading

Explore our research

Cite this work

Guozhang Chen, Wolfgang Maass, Franz Scherr. Sparse cortical dynamics reveal flexible condition-dependent spike-order codes. bioRxiv, 2026. doi:10.64898/2026.05.13.724761.