| Contents |
| Heteroclinic networks or neural ensembles that implement winnerless competition dynamics have recently been proposed to underlie encoding, coordination and execution of information at sensory, central and motor nervous systems. These kinds of networks are appropriate to describe information processing with transient neural dynamics by successive metastable states. Heteroclinic networks largely rely on topologies with asymmetric inhibitory connections. Using an activity rate model description, in this work we show that large-scale heteroclinic networks with inhibitory synapses and electrical gap-junction diffusive coupling give rise to coherent spatio-temporal patterns built of sequential activations. |
|