Author : Rajiv Narayan
Publisher :
ISBN 13 :
Total Pages : 200 pages
Book Rating : 4.:/5 (466 download)
Book Synopsis Discrimination of Complex Sounds by Cortical Neurons by : Rajiv Narayan
Download or read book Discrimination of Complex Sounds by Cortical Neurons written by Rajiv Narayan and published by . This book was released on 2008 with total page 200 pages. Available in PDF, EPUB and Kindle. Book excerpt: Abstract: Many animal species produce spectro-temporally complex communication calls that are crucial for species-specific interactions. While studies have revealed precisely timed responses to vocalizations in the auditory cortex (ACx) of many species, few have directly examined the contributions of spike timing to the discrimination of natural sounds by cortical neurons. Songbirds are an attractive choice for examining the processing of complex sounds because they communicate in complex acoustic settings, using vocal calls that bear striking similarities to human speech. They also possess a combination of well-studied vocal communication behavior and well defined auditory neural circuitry. Neurons in field L (the avian analog of ACx) exhibit strong responses to birdsong that could play an important role in song discrimination. In this work, we investigate neural discrimination of conspecific songs in zebra finch field L. We describe a classification framework for quantifying song discrimination based on the responses of single neurons. The technique is first applied to responses of model neurons constructed from experimentally derived spectro-temporal receptive fields of field L. We demonstrate that model neurons can accurately discriminate between songs if the responses are read at relatively fine timescales ([approximate] 10ms). We extend our analysis to electrophysiological recordings from field L neurons. We quantify the performance accuracy of single neurons and examine the timescales relevant for discrimination. A comparison with avian behavioral performance reveals that neural performance matches behavioral accuracy for the best neurons using a spike-timing based code. Finally we examine neural identification of birdsong (targets) in the presence of competing sounds (maskers). We identify two distinct forms of neural interference: addition of spurious spikes during the silent gaps between the target syllables and suppression of informative spikes during the syllables. These effects systematically decrease discrimination performance as the target intensity decreases relative to that of the masker. The behavioral performance of songbirds in a comparable task also degrades in a similar manner. Our results reveal interference effects that could explain the perceptual difficulties that occur when identifying target sounds from a mixture of competing sounds, which is often refered to as the cocktail party problem.