A factor 1.10 amplitude bound leaves no common strongest target in 14 of 116 selected Allen output groups

We tested whether one target had the largest fitted voltage response in all three train states for outputs of the same source neuron. Allen Synaptic Physiology r2.1 small supplied 247 selected connections in 116 mouse and human source groups. The states were the first pulse, second pulse and late responses during 50 Hz trains. Twenty-four groups had no common strongest target. Fourteen cases remain if each amplitude can change independently by a factor from 1/1.10 to 1.10. Five cases remain at factor 1.25 and none at factor 1.50. We minimized the worst-state deficit over fixed target choices and derived an exact joint error-box rule. Independent linear programs passed 696 feasibility checks. Most groups had a common strongest target. Small human samples, missing outputs, fit error, different trial sets and slice selection prevent a population routing claim. The error bound is assumed, not measured. Code, aggregate results, a figure, source hashes and external retrieval instructions support reproduction. External record tables are excluded.

Allen Institute · synaptic dynamics · output routing · fixed target · error bounds · short term plasticity