Professor William Kristan describes difficulties in building excitatory and inhibitory networks, which can either run out of control or remain inactive.
So, excitatory and inhibitory networks, most networks are a combination of the two, but in fact a lot of times they are predominantly one or the other. So excitatory just means that all the cells, all the neurons in a network are exciting each another, exciting the connections. The synapses they make on one another cause more activity, so they cause the cells to be more active and inhibitory would turn them off. Surprisingly, you would imagine one of the problems with excitatory networks is that they tend to run away, one will excite another, which will excite another, and they will tend to explode. Inhibitory networks tend to turn themselves off, so they have to have a constant level of excitation or inherent activity. So, the problems that one gets into in either building these kinds of networks or in studying them, is a bit different because in one case they tend to get out of hand, and in another case they tend to go nowhere.