Spatial localization and the refinement of orienting behavior: What can be learned from the barn owl?

The barn owl is a nocturnal predator that relies on audition for hunting. In addition to being able to localize sound sources with high accuracy, the barn owl is capable of adjusting its orienting behavior in the presence of changes in the sensorimotor conditions. This is the case, for example, when the head and ears change drastically in shape and size during growth, altering the auditory inputs. In the last two decades, the biological substrate of spatial localization and orienting behavior in the barn owl have been carefully investigated, and much data are now available regarding the anatomy and physiology of the neural structures involved. In this paper we review our recent work modeling the principal neural structures responsible for the production of orienting behavior in the brain of the barn owl. In order to expose these models to sensorimotor and environmental conditions similar to those experienced by the barn owl, we coupled the simulation of the neural structures with a robot that emulates the head of a barn owl. This system was composed of a robotic head equipped with two lateral microphones and a camera, and was presented with auditory and visual stimulation. A number of interesting results have emerged from this work. In particular, it has allowed a deeper understanding of how the barn owl reliably localizes a sound source, by elucidating some of the mechanisms underlying the rejection of noise. In addition, it has led to the formulation of a learning scheme accounting for a wide range of biological observations on how the barn owl calibrates orienting behavior. The resulting system was able to orient accurately toward visual and auditory targets, while maintaining accurate performance even in the presence of manipulations of the sensory or motor conditions. This work provides a direct example of how an interdisciplinary approach, based on the coupling of computer simulation of brain structures with robotic systems, can lead to the understanding of basic biological problems while producing robust and flexible control of systems that operate in the real world.