Page 62 - Foundations of Cognitive Psychology : Core Readings
P. 62

The Appeal of Parallel Distributed Processing  61



































               Figure 4.2
               Some ambiguous displays.The first one is from Selfridge, 1955.The second line shows that
               three ambiguous characters can each constrain the identity of the others.The third, fourth, and
               fifth lines show that these characters are indeed ambiguous in that they assume other identities in
               other contexts.(The ink-blot technique of making letters ambiguous is due to Lindsay and Norman,
               1972).


               so the identity of one letter can influence the identity assigned to another
               in reading.A famous example of this, from Selfridge, is shown in figure 4.2.
               Along with this is a second example in which none of the letters, considered
               separately, can be identified unambiguously, but in which the possibilities that
               the visual information leaves open for each so constrain the possible identities
               of theothersthatweare capableofidentifying allofthem.
                 At first glance, the situation here must seem paradoxical: The identity of each
               letter is constrained by the identities of each of the others.But since in general
               we cannot know the identities of any of the letters until we have established the
               identities of the others, how can we get the process started?
                 The resolution of the paradox, of course, is simple.One of the different pos-
               sible letters in each position fits together with the others.It appears then that
               our perceptual system is capable of exploring all these possibilities without
               committing itself to one until all of the constraints are taken into account.
               Understanding through the Interplay of Multiple Sources of Knowledge  It is clear
               that we know a good deal about a large number of different standard situa-
               tions.Several theorists have suggested that we store this knowledge in terms
               of structures called variously: scripts (Schank, 1976), frames (Minsky, 1975), or
   57   58   59   60   61   62   63   64   65   66   67