This talk, of course, was really close to my heart. Klaude Weiss' lab is also working on the feeding behavior of Aplysia, which is the behavior I use to operantly condition the animals. In this system, the radula (a tongue-like organ) moves in and out of the buccal cavity. Depending on when the radula is closed, it moves either food into the esophagous (ingestion) or inedible objects out of it (rejection; see YouTube video for an ingestive behavior).
The interneutons responsible for generating the neural patterns (buccal motor programs, BMPs) interact to generate a continuum of programs to handle any feeding situations the animal might encounter. This is in contrast to the distinct classes of BMPs one usually uses to describe the behavior.
The properties of the buccal network are history-dependent, such that repeated stimulations trigger different responses depending on the previous stimulation history. We learned that the network has basically two different states and repeated stimulations (via cerebral-buccal interneuron 2) lead to transitions between these two states. Thus, depending on what state the network is in, the same stimulation will lead to different behavioral responses.
Transition between these states appears to be modulated by neuropeptides (e.g. small cardioactive peptide).