NeuromodulationΒΆ

A neuromodulatory event influences synaptic behaviours by simulating additional neurontransmitters. It is implemented by linking from projection fibers to existing synapses as illustrated below:

_images/neuromodulation.png

Projection edges (connection type in SONATA: neuromodulatory) are created between virtual neurons and existing connectome synapses, as defined by SONATA docs.

  • A neuromodulator sends a spike, and the relevent post cell receives the spike at synapses e.g. S1, S2 and S7.

  • The synapse model at S1, S2, and S7 has to manage both presyanaptic spikes (nc_type = 0) and neuromodulatory spikes (nc_type = 10).

  • Neuromodulatory spikes may change parameters in the synapse models (e.g. Use), but not the membrane current i.

  • The neuromodulators influence one or more, but not all synapses from a PRE neuron (e.g. S3 is not changed)

  • The neuromodulators influence synapses coming from different neurons (e.g. S7 that comes from somewhere else)

For each neuromodulatory projection, Neurodamus finds its closest post cell synapse by the location parameter, the exact afferent_section_id and the closest afferent_section_pos (< 0.05).

Spikes from neuromodulators are injected via replay.

The neuromodulatory strength (neuromodulation_strength) and decay time (neuromodulation_dtc) are pre-defined in projection edges, and users can override them via a simulation config file, see connection_overrides.

Source

neuromodulation.rst