Research map / Probabilistic and scientific modeling / Dynamical systems
Spike-to-Field Mechanisms of Turbulence-Like Dynamics in Spatial Spiking Neural Networks
Zequan Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Builds a spatial leaky integrate-and-fire spiking network, maps spikes to activity-density and Hilbert-phase fields, and links spike-level mechanisms to neural-avalanche and turbulence-like macroscopic dynamics.
Problem
Neural avalanches and turbulence-like brain dynamics are measured on coarse-grained continuous fields, leaving unclear which spike-level mechanisms generate these macroscopic observables.
Approach
Explicitly transforms spikes from a spatial LIF network into activity-density and Hilbert phase fields, and tests how a delayed, distance-dependent, passive-dendrite coupling scheme shapes the resulting field-level turbulence-like and avalanche dynamics.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add distance-dependent delays (ablation) | passive-dendrite SNN · Mechanistic controls | mean vortices: 3.70 -> 6.05 | Improved |
| add passive dendrites (ablation) | delayed SNN · Mechanistic controls | mean vortices: 2.29 -> 6.05 | Improved |
| spatially shuffle spikes per frame (global-rate shuffle null) (ablation) | high-rate baseline spikes · Rate-matched nulls | mean vortices: 6.05 -> 0.69 | Hurt |
| randomly permute cell positions (spatial permutation null) (ablation) | high-rate baseline spikes · Rate-matched nulls | mean vortices / S2 slope: vortices 6.05 -> 5.72; S2 slope 0.94 -> 0.72 | Mixed |
| use PV-only inhibition instead of lumped inhibition (ablation) | SNN microcircuit · Inhibitory microcircuits | mean vortices: 6.15 -> 0.35 | Hurt |
| use SST-only dendritic inhibition (ablation) | SNN microcircuit · Inhibitory microcircuits | long tracks: 89.58 -> 94.33 | Improved |
| add VIP disinhibition to PV+SST (ablation) | SNN microcircuit · Inhibitory microcircuits | mean vortices: 1.07 -> 3.77 | Improved |
| apply pure short-term depression (ablation) | SNN · STP | mean vortices: 0.41 mean vortices, 5.67 long tracks | Hurt |
| add SST-like I->E facilitation (ablation) | SNN · projection-specific STP | vortices / firing rate: 8.05 vortices, 120.42 long tracks at 15.62 Hz | Improved |
| use feedforward acyclic graph instead of recurrent (ablation) | low-rate STP SNN · Topology control | pairwise transfer bits/opportunity: 3.35e-4 -> 3.50e-3 | Improved |
| use recurrent graph instead of feedforward acyclic (ablation) | low-rate STP SNN · Topology control (rate-matched) | rho re-entry probability: 0.588 -> 0.736 | Improved |
| use frozen turbulence-like SNN states as readout features | SHD readout · SHD subset (5 classes) | test accuracy: 0.95 -> 0.45 | Hurt |
Closest papers by meaning
- On the Intrinsic Structures of Spiking Neural Networks (JMLR 2024 · similarity 0.87)
- The Dormant Spiking Neuron: A State-Driven Mechanism for Efficient Spiking Neural Networks (NeurIPS 2026 · similarity 0.87)
- Understanding spiking networks through convex optimization (NeurIPS 2020 · similarity 0.87)
- P-SPIKESSM: HARNESSING PROBABILISTIC SPIKING STATE SPACE MODELS FOR LONG-RANGE DEPENDENCY TASKS (ICLR 2025 · similarity 0.86)
- From Cortical Synchronous Rhythm to Brain Inspired Learning Mechanism: An Oscillatory Spiking Neural Network with Time-Delayed Coordination (NeurIPS 2026 · similarity 0.86)
- Latent Diffusion for Neural Spiking Data (NeurIPS 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
