This project builds on the WebGL examples for GPGPU computing to develop in-browser WebGL implementations of Wilson-Cowan style neural field equations. These examples can be browsed as a website here. This project is ever in-development; Please open an issue if you'd like a specific stable configuration to demonstrate a particular effect.
Older versions are here. The most recent WebGL implementation can be found here, and should look like this:
This project predates the widespread availability of floating-point shaders as supported in WebGL3, and therefore uses 8- or 16-bit fixed point representations. This restriction is not so severe for neural field models phrased in terms of a firing rate in [0,1]. If the more recent 16-bit demos are too slow on your device, the original 8-bit WebGL implementation, which is less accurate but friendlier to older hardware/browsers. If your browser doesn't support WebGL, you can try the original CPU implementation here.
This project grew out of the following papers, which use these equations to model hallucinations, responses to cortical stimulation, and retinal waves:
- Rule, Michael, Matthew Stoffregen, and Bard Ermentrout. "A model for the origin and properties of flicker-induced geometric phosphenes." PLoS computational biology 7.9 (2011): e1002158.
- Heitmann, Stewart, et al. "Optogenetic stimulation shifts the excitability of cerebral cortex from type I to type II: oscillation onset and wave propagation." PLoS computational biology 13.1 (2017): e1005349.
- Rule, Michael E., et al. "Neural field models for latent state inference: Application to large-scale neuronal recordings." PLOS Computational Biology 15.11 (2019): e1007442.
Unless otherwise specified, media, text, and rendered outputs are licensed under the Creative Commons Attribution Share Alike 4.0 license (CC BY-SA 4.0). Source code is licensed under the GNU General Public License version 3.0 (GPLv3).