Previously, we developed two similar 3-variable models to explain the patte
rn of spontaneous rhythmic activity generated by the spinal cord of chick e
mbryo. The models differ in the implementation of a slow depression variabl
e, but each produces patterns of activity that qualitatively resemble exper
imental recordings. For both models, the ratio inter-episode interval/episo
de duration (i/d ratio) was too small compared to experiments. The introduc
tion of a parameter representing the connectivity of the network has differ
ent effects on each model, therefore allowing them to be distinguished expe
rimentally. Also, the model that seems to better satisfy experimental const
raints is able to generate a large i/d ratio. (C) 1999 Elsevier Science B.V
. All rights reserved.