Bradford J. May - US grants

Sound localization cues are created by filtering properties of the head and pinna, which are known collectively as the head-related transfer function (HRTF). Because the HRTF is directionally dependent, unique patterns of energy peaks and troughs are added to the amplitude spectrum reaching a listener's eardrum as the source of a complex sound moves from one spatial location to another. The adaptive premium placed on accurate sound localization has exerted strong evolutionary pressure to develop auditory mechanisms for processing these spectral cues. Our longterm objective is to understand how such specializations influence auditory information processing within the central nervous system. Vocal communication systems, including human speech, convey meaning through similar patterns of spectral variation. In this context, proposed experiments will lead to a better understanding of how other biologiCally relevant sounds are first transformed by filtering properties of the ear and then processed by neural circuits within the brain. Expected results will build upon our extensive knowledge of localization behaviors in cats by examining how performance in sound localization tasks is influenced by manipulations such as blocking one ear, lesioning afferent or efferent pathways, and introducing background noise. Experiments of Aim l will test the hypothesis that binaural processing of spectral cues is necessary for accurate orientation toward sound stimuli. Experiments of Aim 2 predict that the dorsal cochlear nucleus performs a critical spectral analysis of the localization cues inherent in HRTF-filtered sounds. Experiments of Aims 3 and 4 will combine electrophysiological and behavioral techniques to test the hypothesis that the olivocochlear efferent system enhances the auditory nerve representation of spectral localization cues, particularly in the presence of environmental noise. Animal psychophysical experiments are the most direct means for testing hypotheses about the function of neural systems. Our research design is particularly powerful in this respect because it integrates well-characterized sound localization behaviors, recently developed neurophysiological models, and natural sound patterns to address fundamental issues of auditory processing.

This project will use extracellular electrophysiological recording to study the effects of background noise on the neural representation of pure tone stimuli. Neural responses will be measured in terms of average discharge rates for single units. Dynamic range properties of single-unit responses will be derived from measures of rate-level functions that are obtained in quiet and in the presence of continuous background noise. Dynamic range measures will be sampled in the dorsal cochlear nucleus, inferior colliculus and primary auditory cortex of behaving cats. Inhibitory mechanisms that shape the neural representation of tones in noise at each of these three levels of auditory processing will be described by frequency response maps, responses to bandpass noise, and responses to broad band noise with notched spectra. This research is based on independent observations that inhibitory influences increasingly reduce auditory responses to continuous noise at higher levels of the auditory system. A reduction in noise-driven activity potentially enhances the rate representation of signals that occur in the presence of background noise by decreasing the negative effects of noise adaptation The primary objectives of this proposal are to demonstrate changes in noise adaptation across three levels of auditory processing, to reveal neural inhibitory mechanisms that contribute to changes in noise-driven activity, and to measure the functional consequences of these changes in terms of the quality of neural representations for signals in noise. Although some of these objectives have been addressed in the nervous system of anesthetized and decerebrate cats, a comprehensive analysis of the neural encoding of pure tones in noisy environments has not been performed in behaving cats. Data obtained in behaving cats are critical to our understanding of noise adaptation because anesthesia substantially alters activity of descending efferent projections to the auditory periphery and inhibitory mechanisms at each subsequent level of auditory processing.

Description (provided by applicant): The long-term goal of this research is to describe the auditory processes that allow us to hear in background noise. Each sensory system shows a hierarchical organization with functional specializations for separating biological signals from less important backgrounds. Our studies have characterized neural response types within the central auditory system that can be interpreted as mechanisms for extracting auditory signals from background noise. These neurons appear to connect in series across the major auditory nuclei to create parallel pathways with exceptional noise-cancellation capabilities. Our proposed studies will address the following questions: What are the special physiological properties of central auditory neurons that improve the encoding of signals in noise (local processing)? Does the improved representation remain functionally segregated as it ascends the auditory system (parallel processing)? Is the neural representation further enhanced when it reaches higher structures (series processing)? How do descending pathways influence the quality of the representation (efferent processing)? Answers to these questions will be pursued with a combination of single-unit electrophysiological techniques and animal psychophysical paradigms. Ties between physiological and behavioral results will be strengthened by performing single-unit studies in awake preparations and using psychophysical procedures that incorporate the essential stimulus conditions of physiological experiments. In addition to experiments in cats that build directly upon our previous studies, the mouse will be introduced as a model for auditory signal processing. Cross-species comparisons of our physiological and behavioral results will distinguish general auditory processes from species-specific specializations. A better understanding of the brain's signal processing solutions will lead to improved designs for assistive listening devices that now show poor performance in noisy environments.

DESCRIPTION (provided by applicant): Our long-range goal is to understand the neural mechanisms of hearing. All information about sound is encoded by patterns of activity in the auditory nerve. By virtue of its position as the obligatory recipient of auditory nerve input, the cochlear nucleus is a key site to study how acoustic stimuli are translated into neural codes that are then sent to the brain. Cochlear nucleus neurons are organized into definable groups that share common inputs, cellular mechanisms, and axonal targets. By linking structure and function for different cell types, we seek to understand their roles in hearing. Ultimately, this kind of knowledge should allow us to interpret structural anomalies resulting from deafness or noise-induced damage in terms of their effect on acoustic processing. In this application, we propose to study multipolar cells in the ventral division of the cochlear nucleus. The axons of these cells target other neurons in the cochlear nucleus. They also project to nearly every brain stem and midbrain structure in the auditory pathway. Multipolar neurons are comprised of several distinct subclasses of cells. We will use pathway tracing and immunocytochemical techniques to study the structure and neurochemistry of different classes with respect to their axonal targets. We will employ an in vitro preparation of the isolated whole brain to study how the activity within a subclass influences its axonal targets. Information obtained from both types of experiments will result in links between structural and functional data for different cell types. This combined anatomic and physiologic approach to the study of multipolar cells will allow us to determine how their encoded messages are distributed in the brain, will impact models of binaural hearing, and may suggest new designs for human brain stem implants.