Physiology of Sound Processing in the Avian Auditory System.

Our interests lie at the intersection of systems, cellular and synaptic physiology of sensory systems, in particular the auditory system. How is sound information encoded in the neural activity? What are the biophysical and cellular processes that create the "code"? One of the early stages in this coding process occurs in the auditory brainstem, where the cochlear nuclei receive input directly from the auditory nerve. The avian brain stem has been particularly useful for the study of the encoding of a sound's location in space. Location is computed using (at least) two binaural cues: interaural timing differences and interaural intensity differences.

Cochlear nucleus angularis (NA) is critical for sound intensity coding, while cochlear nucleus magnocellularis (NM) appears to be highly specialized for coding timing difference. NA's sensitivity to intensity may provide a substrate for spectrotemporal processing, critical for sound object identification, birdsong, and speech. How NA is involved in these tasks has not yet been explored, and an understanding of NA physiology will provide a basis for understanding important auditory functions besides sound localization. NA neurons have different in vitro and in vivo firing properties, responses to sounds, and morphology, compared to NM neurons, and show hetereogeneity within the NA population. We are therefore investigating the synaptic physiology of NA using in vitro slice preparation from young chicks, and to relate these properties to known subtypes. We are particularly interested in the short-term plasticity exhibited in these circuits and its relation to information processing.

The biophysical basis of coincidence detection.

An intriguing property of the nucleus which computes interaural time difference, the nucleus laminaris (NL) is the relationship of the length of the bipolar NL dendrites with tonotopy. NL neurons act as coincidence detectors for incoming binaural signals, and are tuned for particular interaural time delays. Previous modeling work suggests that the dendrites confer a computational advantage (compared to no dendrites) by segregating the inputs arriving from either side. Our lab has been investigating why there exists a gradient of dendritic length and the biophysical properties of NL neurons that might explain this computational effect.


We would like to thank the National Institutes of Health, the National Institute for Deafness and Communication Disorders, and the National Organization for Hearing Research for their generous support.