Joint response reliability left a similarity slope change uncertain after cell controls in one MICrONS mouse

We tested whether the product of two response correlation ceilings changes the relation between measured movie response similarity and synaptic connection. Public MICrONS v1 tables linked activity and anatomy in one mouse. The candidate panel had 295,371 ordered pairs and 8,128 connections. A Poisson mean model with source and target cell effects used 163,870 informative pairs. The interaction coefficient was 0.0265, with a nominal shared-cell 95% interval of -0.0011 to 0.0541. The test left the slope change uncertain. The response-quality interval was just above zero; different-scan and extra-geometry intervals included zero. This is a follow-up to prior reliability analyses, with joint reliability and both cell effects. One animal, selected cells, measurement error and a mean model that can exceed one limit inference. Code, scalar data, fitted outputs, checks and a static figure support reproduction.

MICrONS · response reliability · structure and function · cell effects · uncertainty