Abstract View

SPECTRO-TEMPORAL RECEPTIVE FIELD (STRF) SEPARABILITY OF PRIMARY AUDITORY CORTICAL (AI) NEURONS. H.L.Read*; B.J.-S.Liu; M.A.Escabi 1. Dept. of Psychology, Univ. of Connecticut, Storrs, CT, USA 2. Biomed. Engin., Univ. of Connecticut, Storrs, CT, USA 3. Univ. of Connecticut, Storrs, CT, USA Physical filters generally demonstrate trade-offs in the ability to resolve spectral versus temporal components of a sound. About 30% of neurons in the inferior colliculus have spectral-temporal non-separable receptive fields suggesting that they too show trade-offs in the ability to resolve spectral versus temporal components of sounds. Singular value decomposition (SVD) methods were used to assess separability of cortical STRF's. Single-unit and reverse correlation techniques were used to construct STRF's to the envelop of moving ripple and ripple noise acoustic stimuli. Linear recombination of gabor estimates of the SVD spectral and temporal components sufficiently predicted cortical STRF's for more than 70% of neurons. Spectral bandwidths were typically under 2 octaves with peaks aligned but sometimes phase-reversed for the two ears as observed in inferior colliculus (IC, Qui et al., 2003). Temporal components ranged from 25 to >100 msecs in cortex whereas those found in IC were generally <25 msecs. Increased temporal component durations were generally due to prolonged inhibition or adaptation. Conserved STRF properties include: separability, spectral bandwidth and binaural properties. These data indicate that temporal encoding undergoes a significant transformation from the levels of IC to cortex. Citation: H.L. Read, B.J.-S. Liu, M.A. Escabi. SPECTRO-TEMPORAL RECEPTIVE FIELD (STRF) SEPARABILITY OF PRIMARY AUDITORY CORTICAL (AI) NEURONS. Program No. 182.20. 2003 Abstract Viewer/Itinerary Planner. Washington, DC: Society for Neuroscience, 2003. Online.