The work funded by this grant consists of a number of experimental and computational projects that have advanced our understanding of how the mammalian brain processes complex sounds. We have learned 1) how the processing of binaural sound features is enhanced at successive stages in the brain, 2) how the brain adapts differently to changes in monaural and binaural sound features, 3) how the processing of binaural sound features differs between mammals and birds, and 4) how the processing of complex sounds changes with age. We have also developed a number of new analytical tools that will help us and others to make sense of the complex data provided by recordings from neuronal populations. We have developed 1) a genetic algorithm to optimize decoding of population activity, 2) a new model for population spike trains in which single cell properties such as spike rate and trial-to-trial variability can be manipulated independently of population properties such as noise correlations, and 3) an information-based method for measuring the functional connectivity between neurons that avoids many of the pitfalls associated with the correlation-based methods that are typically used. These new methods are essential in that they will allow neuronal populations to be analyzed as a concerted group rather than as a collection of individuals operating independently.