Causal Discovery via complexity-gated spiking neural networks

We introduce X-SPK (Cross-channel SPiKing), a fully unsupervised, gradient-free spiking method for causal discovery in multivariate time series. Each channel is encoded by an independent fixed leaky reservoir and converted to equal-rate events; directed evidence is accumulated in a shared $$\begin{gathered} D \\ timesD \end{gathered}$$ gate matrix via winner-take-all spike-timing plasticity and read out through the competition ratio $W_{ij}/W_{jj}$. The underlying mechanism is geometric: if $i\to j$, conditioning on events of $i$ increases and concentrates $j$'s near-future occupancy of its threshold-crossing region; this yields more threshold crossings, higher decayed trace mass, and a systematic WTA advantage in the causal direction. The method uses unified model settings with limited dataset-specific preprocessing and is evaluated on CausalRivers, Lorenz, Mackey-Glass, and TimeGraph (including non-stationary and confounded variants), reporting standard metrics against PCMCI+/GPDC, DYNOTEARS, VARLiNGAM, NeuralGC, CCM, and Rhino/RhinoLite across linear, nonlinear, and non-stationary regimes.