Connectivity statistics treat the cosmic web as a graph: count the filaments meeting at each halo, the junction multiplicities, the node degrees, and compare against what hierarchical growth from a Gaussian initial field predicts. In ΛCDM the distribution of connections per node is well defined once mass and environment are fixed, rising slowly with halo mass as more filaments find their way to deeper potential wells (Bond et al. 1996; Cautun et al. 2014; Codis et al. 2018).
The measured graph runs over-connected at the top. Observations and high-resolution simulations report massive nodes with more filaments than the growth picture comfortably supplies, excess multi-filament junctions, and connectivity-mass and connectivity-environment relations that deviate from the halo-model baseline (Codis et al. 2018; Darragh-Ford et al. 2019). The discrepancy is structural rather than numerical: in the standard picture, connectivity is something a node accumulates as shear funnels matter toward it, so a node cannot easily out-connect its growth history, and an over-connected population pushes toward non-Gaussian initial conditions or missing dynamics in how the web's graph assembles.
The standing is young but consistent across methods, and tied to the rest of the web sector: the same catalogs showing long coherent filaments and flat width-mass relations also show busy junctions. DESI and Euclid will map connectivity across orders of magnitude in node mass, testing whether the graph is the smooth output of growth or carries the intersection structure of discrete events.