It seems reasonable that the description of scene analysis in human listeners could apply to other species as well. The notion of scene analysis itself could be a most important property of all sensory systems, in all organisms, as they attempt to determine the actual sources, under the usual conditions of multiple, simultaneous sources, some of obvious `biological
signifi-cance,' and some not. We suggest that having capacities to segregate individual,
simultaneous sources is a state of fundamental adaptive value. In this view, all sounds audible to a species are significant in the sense that fitness of the species would depend upon the ability to segregate the relevant sources from the irrelevant ones. In the case of auditory scene analysis, the environmental constraints and factors that operate to shape the development of any species' auditory system and sense of hearing are simply the facts of physics ^ the principles of mechanics that make one sound source di¡erent from another. These principles are essentially the same for all organisms, and for all time. Thus, adaptations to these physical
principles could be expected to be similar among organisms. This similarity may not necessarily be in anatomical structure or physiological mechanism, but in functional result. Thus, one might expect the sense of hearing to have broadly similar functions among organisms because they all have essentially the same fundamental problems to solve: that of sound source segregation and determination. Predator and prey `detection,' presumably two of the earliest and most fundamental of auditory functions, should be recognized as complex tasks requiring information to be acquired and processed about these sources independently of the rest of the
auditory scene. This sort of sound source determination would also seem to be required for communication; in many cases, the physical characteristics of the sources are the most
important message in communication. Success in gen-eral sound source determination, including acoustic predator and prey detection and communication, would seem to require the acquisition of the same sorts of information about sound sources in usually complex acoustic environments that are used in auditory scene analysis by humans, and as recently demonstrated, in starlings and goldfish . In summary, we argue that the task of determining sound source characteristics in the usual environments containing multiple sources, scatterers, and noise has given rise to the most important general pressures operating throughout vertebrate evolution in initially
shaping and maintaining the vertebrate sense of hearing.