Although crickets are a model system for acoustic communication since more than 50 years so far very little attention has been paid to the neural networks generating the singing motor pattern.
During singing, male crickets rhythmically open and close their front wings producing a sound pulse during each closing movement. Three to five pulses are grouped to chirps that are continuously repeated with a rate of 2-3 Hz. The motor pattern underlying singing behaviour basically consists of alternating activity of wing opener and closer motoneurons. Singing behaviour is controlled by an identified command neuron descending from the brain. Its tonic spike activity drives the central pattern generator (CPG) for stridulation. Singing motor activity can be released in de-afferented preparations, when the command neurons are activated by micro-injecting cholinergic transmitter agonists into the brain.
As the front-wing motoneurons are located in the mesothoracic ganglion it was assumed that also the neural network constituting the singing CPG is housed in the thoracic ganglia. In order to identify the interneurones of the singing network, we intracellularly recorded and stained interneurones first in the thoracic and then in the abdominal ganglia of fictively singing crickets., The singing pattern is monitored by recording the motoneurone activity in a truncated wing nerve. We test if neurons are part of the singing CPG by perturbing their rhythmic activity with intracellular current injection and analyzing if they entrain or reset the ongoing singing pattern.
Contrary to previous assumptions we found that some interneurons of the singing CPG are housed in the abdominal ganglia. Interneurons ascending from the first abdominal ganglion have a strong impact on the singing pattern. Upon electrical intracellular stimulation they reset or elicit the generation of opener and closer motoneuron activity in the pulse pattern of the song. The participation of the abdominal ganglia in the pattern generation explains why crickets stop singing, once the abdominal connectives are cut.
We now focus this experimental approach to analyse how the central motor pattern is forwarded to the corollary discharge interneurone (CDI), that modulates the sensitivity of the auditory pathway during singing. The CDI interneuron is activated in phase with the closer motoneurons and we expect it to inhibit other sensory pathways during singing. Recently we found an inhibitory input to cercal interneurons in singing crickets.