The local bending reflex of the leech computes a well-defined sensorim
otor input-output function in which each of several unique patterns of
sensory input elicits a unique pattern of motoneuron activity. Intern
eurons in the reflex respond to most input patterns and contribute to
most motor patterns, suggesting a distributed processing mechanism for
the reflex. This suggestion is supported by models in which connectio
n strengths are adjusted by a neural network optimization algorithm to
reproduce the local bending input-output function. In addition, compu
tational parallels between the local bending network and the perceptro
n, a major class of artificial neural networks, brings the functional
role of local bending interneurons into question and suggests new phys
iological experiments.