Page 216 - Foundations of Cognitive Psychology : Core Readings
P. 216

The Auditory Scene  221

               from our memories, they tend to coalesce into a single entity and we experience
               an angry green dog and not merely anger, greenness, and dogness taken sepa-
               rately. Although the combination of these qualities has never occurred in our
               experience, and therefore the individual qualities must have been dredged up
               from separate experiences, those qualities can be experienced visually only as
               properties of an object. It is this ‘‘belonging to an object’’ that holds them
               together.
                 The streamplays thesameroleinauditorymentalexperienceasthe object
               does in visual. When we want to talk about auditory units (the auditory coun-
               terparts of visual objects), we generally employ the word ‘‘sound.’’ We say that
               a sound is high pitched or low, that it is rising or falling, that it is rough
               or smooth, and so on. Again I am convinced that this is not simply a trick of
               language, but an essential aspect of both our conceptual and our perceptual
               representations of the world. Properties have to belong to something. This
               becomes particularly important when there is more than one ‘‘something’’ in
               our experience. Suppose there are two acoustic sources of sound, one high and
               near and the other low and far. It is only because of the fact that nearness and
               highness are grouped as properties of one stream and farness and lowness as
               properties of the other that we can experience the uniqueness of the two indi-
               vidual sounds rather than a mush of four properties.
                 A critic of this argument might reply that the world itself groups the ‘‘high’’
               with the ‘‘near’’ and the ‘‘low’’ with the ‘‘far.’’ It is not necessary for us to do it.
               However, it is not sufficient that these clusters of properties be distinct in the
               physical happenings around us. They must also be assigned by our brains to
               distinct mental entities. In auditory experience, these entities are the things that
               I am calling streams. As with our visual experience of objects, our auditory
               streams are ways of putting the sensory information together. This going to-
               gether has obvious implications for action. For example, if we assign the prop-
               erties ‘‘far’’ and ‘‘lion roar’’ to one auditory stream and the properties ‘‘near’’
               and ‘‘crackling fire’’ to another one, we might be inclined to behave differently
               than if the distance assignments had been reversed.
                 When people familiar with the English language read the phrase ‘‘The gray
               wagon was on the black road,’’ they know immediately that it is the wagon that
               is gray, not the road. They know it because they can parse the sentence, using
               their knowledge of the English syntax to determine the correct ‘‘belongingness’’
               relations between the concepts. Similarly, when listeners create a mental repre-
               sentationofthe auditory input, they toomustemployrules aboutwhatgoes
               with what. In some sense, they can be said to be parsing this input too.
               The Principle of Exclusive Allocation
               Any system that attempts to build descriptions of a natural world scene must
               assign the perceptual qualities that it creates to one organization or another.
               The quality ‘‘loud’’ is assigned to the organization that represents the roar of
               the lion. The quality ‘‘far’’ is assigned as the distance of that same event. The
               Gestalt psychologists made this point by introducing the principle of belong-
               ingness. In describing the visual organization of drawings like the one in figure
               9.5, they pointed out that the lines at which the drawn irregular figure overlaps
               the circle (shown as a dark line in part B of the figure) are generally seen as part
   211   212   213   214   215   216   217   218   219   220   221