Organized representation of spectrotemporal features in songbird auditory forebrain.

Much of our understanding of the functional organization of auditory cortex is based on relatively simple stimuli, and the cortical representation of complex sounds remains poorly understood. With their rich vocal communication and learning behaviors, songbirds can offer insights into the neural processing of complex acoustic signals analogous to human speech. In the primary forebrain auditory area field L (primary auditory cortex analog) of zebra finches, previous studies identified a limited set of spectrotemporal receptive field (STRF) types, but whether these were spatially organized remained unclear. Here, we investigated the spatial organization of field L in multiple anatomical planes, using multielectrode array recordings and a stimulus that captures aspects of spectrotemporal modulations of song. Our data demonstrate two separate axes along which spectrotemporal aspects of sound are mapped: width of spectral tuning changes systematically from narrow to broad mediolaterally, whereas width of temporal tuning changes markedly, from narrow to broad, from the input to the output layers. These spatial patterns, which are evident in STRFs computed from both multiple and single units, result in a mapping of basic STRF types to subregions of field L. This highly organized representation of spectrotemporal features of sound contrasts with current views of mammalian auditory cortex, where no strong spatial organization of STRF shapes has been seen thus far. Our data identify a coherent initial cortical representation of sound features and suggest that forebrain filtering for basic perceptual qualities of sound occurs in a spatially organized and segregated manner in the songbird auditory system.