A major issue of science today concerns the delineation of neural mechanisms underlying behavior, including mechanisms for sensory information processing. One important task performed by sensory systems is the recognition of specific signal patterns. For example, mechanisms for sound pattern recognition are important for the perception of auditory communication signals, including human speech. This project will use psychophysical techniques to study the neural computations for sound pattern recognition in the echolocating lesser bulldog bat. The bats will be presented with artificial echo signals and a behavioral response of the bat to the signals will be used as a bioassay for complex sound pattern recognition. The project will test the hypothesis that a critical computation step for pattern analysis involves neural summation of an array of frequency specific filters. Results will be used to generate a model of the neural computations for pattern recognition in this species. The model could then potentially be applied to both non- human and human central nervous systems, and perhaps be used to provide insights for generating models for pattern recognition and information processing in artificial systems. Many aspects of the bat auditory system are similar to those of other mammals. Therefore the proposed project may increase our understanding of speech perception in humans. Moreover, an understanding of how the bat processes its biosonar signals could result in improvements in computer based artificial pattern recognition systems. Types of pattern recognition systems that could benefit from this knowledge include sonar and radar systems, speech recognition systems, and various medical imaging devices.
