How do I know if my model meets biological constraints?

I’ve come up with what I think is a pretty cool model using the default Nengo parameter settings. I want to make sure that my model matches biological constraints so that I can make comparisons with real data. How can I figure out if my model is biologically plausible?

2 Likes

We’ve tried to build biological plausibility into Nengo directly. In most cases, you have to explicitly use a Node or nengo.Direct neurons to do something that is not biologically plausible.

When it comes to setting the parameters on an Ensemble or Connection, the biological plausibility of your model is going to depend on the system you’re modeling. If you’re modeling a specific area of the brain, then you should look up as much as you can about the neurophysiological details of that region in general, and of the neuron and synapse types found in that region.

If your model could exist in different regions, or you don’t have a specific hypothesis as to where your model might be in the brain, then we often use settings that correspond to neocortical pyramidal cells, with glutamatergic synapses in most connections, and GABAergic synapses in inhibitory connections.

The CNRGlab’s website contains several pages (listed below) to help you find specific numbers for certain quantities. It would be great to keep adding to these pages if you find concrete numbers in your research.