Spatial and non-spatial auditory cortical processing in humans

In complex listening situations typical of everyday life, we are often faced with the problem of identifying and localizing various sound sources in the environment. Four studies examined human brain activity associated with processing auditory identity and spatial information in order to test the hypothesis that distinct streams of brain activity are associated with identifying and localizing sounds (i.e., the auditory dual-pathway model). In Experiment 1, a brief delay match-to-sample task in which sounds varied in pitch and horizontal location provided converging functional magnetic resonance imaging (fMRI) and event-related potential (ERP) evidence that the processing of pitch and location information invokes distinct haemodynamic and electrophysiological patterns of activity. The meta-analysis of Experiment 2 further validated this model when an examination of all published auditory spatial and non-spatial human imaging (fMRI and positron emission tomography) data revealed a clear functional dissociation into "what" and "where" processing streams. Experiment 3 employed a modified version of Experiment 1's delay match-to-sample task in which the delay between the two sounds was increased so that the maintenance and comparison stages of each working memory task could be evaluated. The fMRI activity demonstrated that although much of the identification network was expressed during the maintenance aspect of the task, it was the active comparison of the two locations that invoked much of the spatial network. Finally, an electrophysiological study of sustained attention to either the identity or location of environmental sounds (Experiment 4) investigated the temporal components of the two functional streams. In separate blocks, listeners detected repetitions in sound category or location. Relative to the category task, it was found that target-related localization activity occurred earlier suggesting that the dorsal system processes information faster than the ventral pathway. Taken together, the results from these four studies strongly argue in favour of specialized auditory streams in the human brain. Because of the similarity to "what" and "where" segregation in vision, the current research suggests that "what" and "where" processing may represent a fundamental principle of functional organization in the human brain. Speculation is offered on the purpose of having two specialized streams.