|
|
@@ -0,0 +1,37 @@
|
|
|
+# Feature competition in ellipsoid
|
|
|
+
|
|
|
+## Recurrent inhibition can amplify a signal
|
|
|
+
|
|
|
+- Peter Dayans Book, mostly based on visual
|
|
|
+- [Wangs Paper on decision making](https://www.ncbi.nlm.nih.gov/pubmed/12467598) also uses recurrent inhibition
|
|
|
+ * takes IF neurons
|
|
|
+ * strong recurrent connectivity
|
|
|
+- find a simple rate based model and simulate it
|
|
|
+- Song and other papers on recurrent inhibition
|
|
|
+ * references 8,9,10 in this [paper](https://www.nature.com/articles/nn.3704), 8 and 12 show that gird cell networks seldomly recurrently excite each other
|
|
|
+ - [8](https://www.sciencedirect.com/science/article/pii/S0896627312011191?via%3Dihub) connect oscillatory patterns with grid cell formatioon via recurrent inhibition, exponential integrate and fire neurons, nice [source code online](https://senselab.med.yale.edu/ModelDB/ShowModel.cshtml?model=150031&file=/GridCellModel/grid_cell_model/grid_cell_network_brian.py#tabs-1)
|
|
|
+
|
|
|
+- Renshaw cells are inhibitory cells in the spinal cord, which are recurrently connected to motor output units, the activation of the inhibitory neurons then allows to modulate the gain of the motor neurons
|
|
|
+- in principle, one could assume a slightly different recurrent inhibition, namely that the inhibition has a stonger effect on the neuron that does not drive the interneuron, but how one generalizes this for multiple neurons is unclear
|
|
|
+- another effect of recurrent inhibition are oscillations, see [here](https://www.cell.com/trends/neurosciences/fulltext/S0166-2236(96)20048-7), what would this mean for different interneuron shapes?
|
|
|
+- Hannay Monyer found that interneurons receive input from very different grid cells, in a [commentary](https://www.nature.com/articles/nn.3704) from Moser, they discuss the implications ofr recurrent inhibitory models
|
|
|
+## Compare connectivity between modules for polar and circular interneurons
|
|
|
+- generate connectivity matrices by assuming that inside the density of an interneuron, all pyramidal cells are recurrently connected
|
|
|
+- compare connectivity matrices for the two cases
|
|
|
+- what happens to the connectivity if modules are introduced
|
|
|
+ * canonical way of introducing modules?
|
|
|
+
|
|
|
+
|
|
|
+# Meeting
|
|
|
+1. What is the typical I-E connectivity?
|
|
|
+2. Differences in connectivity?
|
|
|
+3. How to introduce features? Spatially modulated sine...
|
|
|
+
|
|
|
+## Todo
|
|
|
+1. Get a distance dependent effective coupling function, ex-in-ex via the same interneuron inhibits, ex-in-ex-in-ex
|
|
|
+in contrast leads to disinhibition
|
|
|
+* do I want to do this with graph-tool
|
|
|
+* how to identify the right network motif?
|
|
|
+* which other figures could be useful?
|
|
|
+* save and load connectivity
|
|
|
+
|
