Stimulation patterns are added the coupling variables of BNM

stimulus = numpy.zeros((self.model.nvar, self.number_of_nodes, 1))
LOG.debug("%s: stimulus shape is: %s" % (str(self), str(stimulus.shape)))

The state variables affected by a stimulus should be a configurable parameter as it is the case of the noise in stochastic integration.

For some models the current implementation could be considered as sufficiently correct, but for other models, this isn't the case.

In the same way we define a cvar for the coupling, there could be a svar, for stimulation.
The stimulus strength [weighting] could have different values for nodes, state variables and modes.

In the current implementation the stimulus is added to the coupling variable