Selected MICrONS pairs leave a slope change at zero measured movie correlation uncertain

We tested whether the relation between movie response correlation and synaptic connection rate changes at zero correlation. We used public MICrONS v1 tables from one mouse, with 295,371 candidate directions and 8,128 connections. A continuous two-slope Poisson model controlled for source cell, target cell and contact opportunity. The primary slope change was 0.0371 log-rate units per 0.1 correlation. Its nominal 95 percent interval was -0.0282 to 0.1024, with p = 0.263. The negative-side slope was 0.0317, with interval -0.0279 to 0.0913. The positive-side slope was 0.0688, with interval 0.0502 to 0.0874. The three measured-response checks also left the slope change uncertain. A secondary model-response test gave a slope change of 0.0860, with interval 0.0110 to 0.1611. This test does not prove a linear rule or an absence of negative-correlation effects. One mouse, selected reconstructed cells, response error, unmeasured geometry and network dependence limit transfer. Code, a derived pair table, source hashes, an independent fit check and a figure support reproduction.

MICrONS · response correlation · connectivity · nonlinear relation · null result · cell fixed effects