With fixed contact filters, a shared radius gave isolated dopamine access to at most 1.81% of HVC spine sites in one Area X map

We tested when distance alone can give an HVC spine access to dopamine contacts on its own target cell while excluding contacts on other targets. We used 796,003 HVC spine-head sites on 8,096 medium spiny neurons and 85,328 putative dopamine contacts in one adult zebra finch Area X export. The main filters required a contact score of at least 0.6, area of at least 0.1 square micrometres and source skeleton length of at least 50 micrometres. A contact on another target was nearer at 765,098 sites (96.12%). For any common nonnegative radial response that does not increase with distance, these sites cannot give a larger individual-contact effect to a contact on their own target. Exact optimization over every shared radius gave at most 14,395 isolated sites (1.8084%). The earliest maximizing radius was 3.2388 micrometres. Separate radii for each site could isolate only 3.8825%. Lower dopamine area cutoffs raised the shared-radius maxima to 3.43–3.72%. Direct distance checks matched exactly. An offline replay reproduced the main results, and a clean source rebuild reproduced all seven numerical products byte for byte. The result limits a geometric access gate in one selected point catalogue. It does not test summed signals, receptor placement, selective release, local eligibility or learning. The package preserves code, derived distances, source hashes, the proof, figures, results and repeat steps.

connectome · dopamine · Area X · song learning · spatial access · distance · eligibility trace · exact optimization