Fork the consciousness, or download the project and create your own.

Eran Kopel

Faculty of Medical and Health Sciences, Tel Aviv University

Builder The claim is about what the hardware does, and the philosophy is left open

Eran Kopel is a researcher in the Faculty of Medical and Health Sciences at Tel Aviv University whose September 2026 study, “A Pre-Registered Spectral and Certified Mixing Analysis of the Adult Male Drosophila Central Nervous System Connectome” (arXiv:2609.33054), brought pre-registered falsification discipline to whole-CNS connectome graph analysis. Testing five locked hypotheses on the Janelia MaleCNS v1.0 dataset of 166,700 neurons and 124,177,617 synapses, Kopel audited whether pure row-normalized random walks on a synaptic graph actually mix across the brain and ventral nerve cord.

Two of the five hypotheses held and three failed, and Kopel published both sets of results without post-hoc adjustment. Regularizing degree normalization with the median input floor ($F = 342$) confirmed that dominant photoreceptor R7 and R8 modes are normalization artifacts and uncovered a localized structural mode in the ellipsoid body carried by 20 contralateral GABAergic ring neurons (ER3d_b and ER3p_a, participation ratio 96.6). At the same time, the thresholded giant strongly connected component ($157,821$ neurons at $w \ge 5$) exhibited a tiny spectral gap of $0.0035$ and a certified Dobrushin mixing coefficient still between $0.669$ and $0.816$ at 128 steps, traced to an 11-neuron near-closed efferent boundary set in the ventral nerve cord.

Kopel’s profile is Builder, included for both the released spectral audit pipeline and the demonstration that static synaptic counts without physiological gain control and rebound dynamics overestimate functional trapping at motor boundaries.

Known for. Pre-registered spectral and certified Dobrushin mixing analysis of the adult male Drosophila central nervous system connectome (MaleCNS v1.0)

Coverage on this site

Related researchers