With turnover totals fixed and enough contacts for each loss, somatic inhibitory puncta had excess two-day gain and loss agreement in six mice

Do inhibitory contact gains and losses share time beyond their total rates? We tested baseline CA1 imaging data from six mice in Klimmt et al. (2026). We checked all 433 observed count rows and used 61 complete neuron-compartment groups. The planned pooled test gave no clear effect (p = 0.090). A planned somatic secondary test gave excess timing agreement. We then found that some reference tables assigned more losses than the available contacts. We added exact rejection sampling with the same neuron gain totals, mouse-compartment interval gain totals, all loss counts, and enough starting contacts for each loss. In 27 somatic groups, the same-interval share of all possible gain and loss pairs was 19.78%, versus 18.09% in this reference. The excess was 1.69 percentage points. The two-sided conditional p value was 0.0060, or 0.018 after correction for three compartment tests. The final pooled test remained unclear (p = 0.130). The descriptive 95% mouse interval for the somatic excess included zero. Somatic mean squared count change was 38.6% below the physical reference. This gives the same decrease in squared total synaptic weight change only if all contacts have equal fixed weight. It does not show stable current or causal replacement. The optical puncta have no source axon labels. The package preserves source data, provenance, local plans, code, four figures, reference draws, and a fresh replay check.

inhibitory synapses · structural plasticity · hippocampus · gephyrin · turnover timing · conditional randomization · contact availability · structure and function