Résumé |
(1) Being interested in the mechanisms by which recognition occurs, and
given the relative paucity of research in ecological acoustics (aside from a
growing body of research on speech perception), this chapter will primarily
adopt the information processing approach in the discussion that follows,
making reference to possible contributions by the ecological stance where
these are appropriate. (2) The purpose of the present chapter is to examine
aspects of auditory representations and the nature of the processes operating
on them that result in the recognition of sound sources and events. For the
purpose of illustration, the discussion will be primarily confined to musical
instruments and a few simple, `natural' acoustic events. Relatively little
work has been done on systematically investigating the various stages of the
recognition process in audition. Therefore, I am obliged to examine a
number of experiments that were directed at other experimental issues in an
attempt to glean indirectly information that will help us piece together a
picture of nonverbal auditory recognition. While the purpose of this chapter
is to examine auditory recognition in general, most of the experiments that
have directly dealt with the recognition of musical instruments and natural
sound events have been confined to identification experiments. (3) The
structure of the rest of this chapter is as follows. First I will consider in
abstract terms the stages of processing that may be imagined to contribute to
recognition as conceived within the information processing approach and
extract from this consideration a number of important issues that should be
analyzed. Next I will examine experimental data from the literature on
perception and identification of musical instrument tones and natural acoustic
events. These experiments variously involved asking listeners to
discriminate between sounds, to judge the degree of similarity among them,
to classify them, or to identify them by name or by an arbitrarily chosen
label. The data obtained from these experiments will be used to complete a
more concrete, if tentative, picture of the stages of auditory processing
previously discussed in abstract terms. Finally, I will discuss the properties
of a number of models of recognition drawn from the domains of nonverbal
audition, speech, and visual form recognition and analyze them in terms of
what they might contribute to further clarification of the process of auditory
recognition.
|