The overall goal of this project is to develop a comprehensive computer model of neural coding of auditory space in the auditory thalamocortical system. A self-organizing neural network model of auditory cortical maps will be studied. The output of the network will be that of a simulated planar cortex that will afford direct comparisons of simulated virtual space receptive fields (VSRF) with those of actual VSRFs of direction- sensitive neurons in primary auditory cortex (AI). These computational simulations will greatly benefit from the ability to synthesize virtual auditory space from measured Head-Related Transfer Functions (HRTFs). The same stimuli used for microelectrode recordings in auditory cortex of cat will be used in the development and stimulation of the neural-network models. These network models assume that primary auditory cortex is subject to experience-dependent changes. Currently available computational models of neural signal processing in the auditory periphery and brain stem will be used to provide a neural representation of binaural stimuli to the self-organizing thalamocortical model. The simulation of self-organizing processes operating on input data with intrinsic structure leads to the emergence of topographical maps. These maps afford the opportunity to examine overlays of functional organization. Currently, the ability to corroborate the emergence of a spatial auditory map with a detailed map of an auditory cortical field is quite limited, but the simulated development of maps will demonstrate how global topographic order can emerge, in principle, from local cooperative and competitive interactions within the cortical field. It is anticipated that these simulation studies will help guide neurophysiology research with regard to deciphering the neural code of auditory space. Specifically, the simulations may suggest where to probe the cortex with microelectrodes and with what types of stimulation. Interactions between the tonotopic frequency organization and orthogonal iso-frequency organization will be investigated computationally. This computational modeling work may provide a better understanding of the representation of complex sounds in general at higher levels in the auditory system. Given the nature of the model to re-organize, effects of cochlear lesions can be studied and thus aid in the study of sensorineural hearing impairment.
| Jenison, R L (2000) Correlated cortical populations can enhance sound localization performance. J Acoust Soc Am 107:414-21 |
| Jenison, R L; Reale, R A; Hind, J E et al. (1998) Modeling of auditory spatial receptive fields with spherical approximation functions. J Neurophysiol 80:2645-56 |
| Wotton, J M; Jenison, R L (1997) A backpropagation network model of the monaural localization information available in the bat echolocation system. J Acoust Soc Am 101:2964-72 |
| Wotton, J M; Jenison, R L; Hartley, D J (1997) The combination of echolocation emission and ear reception enhances directional spectral cues of the big brown bat, Eptesicus fuscus. J Acoust Soc Am 101:1723-33 |
