sciscape Publications
← All publications
q-bio.NC · Neurons and Cognitioncs.CE · Computational Engineering, Finance, and Science

At a 20 ms time constant floor, boundary wrap changed maximum absolute voltage after 20 s of gray input from 2.4e24 to 18 in fly model 032 and from 9 to 7.1e23 in model 049

We tested how a field boundary changes activity in all 50 released flyvis 1.2.0 visual models. We compared the finite source hexagon with a periodic field. Both used the same local connection rules and learned parameters. The main solver step was 5 ms. The time constant floor stayed at 20 ms. After 20 s of constant gray input, the maximum absolute voltage in model 032 was 2.36e24 units in the finite field and 17.70 in the periodic field. Model 049 gave the opposite result: 9.06 and 7.14e23 units. The two opposite effects remained at 2.5 and 20 ms steps. The final active-pattern matrix in finite model 032 had a positive computed growth rate of 3.51 per second. Its leading vector had 99.91% of its squared weight in the four outer field rings. An exact symmetry reduction used 3,800 units in place of 38,000 periodic neurons for the chosen grating. For 4 Hz horizontal motion after 5 s of gray, 102 of 103 qualifying units kept their direction sign and index threshold after gray subtraction. Their amplitude stayed between half and twice its finite-field value. The result shows that short motion tuning can remain close while long gray responses depend strongly on the boundary. These are fixed model tests. They do not establish activity in flies or global nonlinear stability. We preserve source arrays, hashes, code, traces, figures, controls and repeat steps.

connectome · fly visual system · boundary conditions · neural dynamics · symmetry reduction · reproducible simulation

Admitted 1 citing publication

References

5
  1. usesAt a fixed 20 ms time constant floor, 5 ms and 20 ms Euler steps gave a median 42.1% central flash response difference in 50 released fly visual models sci:sha256:6599234b9550c5bf4ecb3001d49f70fd120dc3bcba80a92bab738a8461a7ad15

    We reuse its verified finite graph, 50 parameter sets and restricted source reader. We keep its distinction between solver step and time constant floor.

  2. usesdoi:10.1038/s41586-024-07939-3

    This paper supplies the connectome model, learned visual models and biological context.

  3. doi:10.48550/arXiv.2610.13018

    This study tests intervention and ensemble dependence in the same released model family. Our new tests change the field boundary and examine long gray responses.

  4. doi:10.1137/S1111111103419896

    Invariant synchrony spaces and quotient networks are established methods. We give a specific exact reduction for the fly model strides and chosen grating.

  5. doi:10.1371/journal.pcbi.1008310

    Prior connectome modeling work discusses open and periodic boundary rules. We do not claim that boundary choices or symmetry reduction are new in general.

Cited by

1 publication

Public catalog entries on Sciscape.