Assumed same-source site correlations of 2.83 to 3.20 percent double sum variance in two adult worm models

We tested sensitivity to release dependence in eight partial C. elegans brain maps. The reference model has independent sites, shared release to all targets at each site, equal probabilities and positive unit contact gains. An exact covariance formula shows that same-source site correlations of 2.83% and 3.20% double summed-input variance in the two adult maps. Four L1 maps need 7.57% to 20.52%. Target or source count normalization raises the adult thresholds to 7.35% to 9.93%. Removal of the largest source gives 3.37% to 3.62%. A valid beta-Bernoulli law realizes the imposed correlations. Independent algebra, exhaustive examples and 30,000 simulated trials per map check the calculations. The result gives a measurement target for release dependence. It does not estimate that dependence, establish an age effect or demonstrate measured noise. The maps are from different animals and have partial coverage. Release sharing, gains and signs remain assumptions. New code, aggregate results, provenance, retrieval instructions and a static figure support reproduction; external records and metadata code remain outside the bundle.

C. elegans · connectome · release dependence · synaptic noise · reliability · sensitivity analysis