[
International Worm Meeting,
2021]
A neural bottleneck is characterized by a convergence of multiple inputs onto a smaller number of neurons. This configuration implies that the network compresses information encoded in the incoming signals. However, it is unclear how biological neuronal networks may perform such a compression and how the information content may change. The pair of RIP neurons in C.elegans represent a simple implementation of a neural bottleneck. The RIPs receive converging sensory inputs and provide the only connections to the network controlling pharyngeal pumping through a pair of gap junctions with the pharyngeal I1 neurons. To investigate the role of such a bottleneck motif, we supply controlled touch stimuli to a population of worms while observing changes in pumping behavior. Estimating information compression requires large amounts of measurements of both the input and output signals. We therefore implemented a high-throughput assay taylored to supplying vibrations as touch stimuli while observing feeding behavior in large worm populations. We present a custom image analysis tool termed 'PharaGlow' that can detect pumping events in several animals moving on standard cultivation plates. As expected, we find that pumping is inhibited by vibrations and this inhibition is abolished in mechanosensory defective mutants. Genetically disrupting the RIP-I1 gap junctions also abolished this inhibition. These results show that this connection is necessary for pumping inhibition by vibrations and provide us with the necessary tools to study sensory information encoding and compression in the RIPs neural bottleneck.