Schemata

   1) Once Established, Schemata Can Control Later Observations

   Once we have accepted a configuration of schemata, the schemata themselves provide a richness that goes far beyond our observations. . . . In fact, once we have determined that a particular schema accounts for some event, we may not be able to determine which aspects of our beliefs are based on direct sensory information and which are merely consequences of our interpretation. (Rumelhart, 1980, p. 38)

   2) The Nature of Schemata

   Through most of its history, the notion of the schema has been rejected by mainstream experimental psychologists as being too vague. As a result, the concept of the schema was largely shunned until the mid-1970s. The concept was then revived by an attempt to offer more clearly specified interpretation of the schema in terms of explicitly specified computer implementations or, similarly, formally specified implementations of the concept. Thus, Minsky (1975) postulated the concept of the frame, Schank and Abelson (1977) focused on the concept of the script, and Bobrow and Norman (1975) and Rumelhart (1975) developed an explicit notion of the schema. Although the details differed in each case, the idea was essentially the same. . . . Minsky and the others argued that some higher-level "suprasentential" or, more simply, conceptual structure is needed to represent the complex relations implicit in our knowledge base. The basic idea is that schemata are data structures for representing the generic concepts stored in memory. There are schemata for generalized concepts underlying objects, situations, events, sequences of events, actions, and sequences of actions. Roughly, schemata are like models of the outside world. To process information with the use of a schema is to determine which model best fits the incoming information. Ultimately, consistent configurations of schemata are discovered which, in concert, offer the best account for the input. This configuration of schemata together constitutes the interpretation of the input. (Rumelhart, Smolensky, McClelland & Hinton, 1986, pp. 17-18)