In walking down the street with a friend, one simultaneously engages in at least five movement processes: walking (including maintaining posture), breathing, talking, gesticulating, and scanning the shop windows and passers-by. But each of these processes involves the cooperation of multiple processes: for example, stepping is determined
inter alia by high-level route-selection processes ('turn left at the town hall'), visual
feedback about the location of obstacles, and tactile feedback from the soles of the feet. And each of these in turn requires activity in a neural network linking an array of receptors with an array of motor neurons.
These behaviours involve not only 'externally directed' movement, but also a variety of 'exploratory' movements that help update an 'internal model of the world' (Craik 1943). In a new situation, we can recognize that familiar things are in new relationships, and use our knowledge of those individual things and our perception of those relationships to guide our behaviour on that occasion. It thus seems reasonable to posit that the 'internal model of the world' must be built of units which correspond, roughly, to domains of interaction — a phrase carefully chosen to include objects in the usual sense, but to include many other things besides, from some attention-riveting detail of an object all the way up to some sophisticated domain of social or linguistic interaction for purposeful beings. We may use the word 'schema' to correspond to the unit of knowledge — the internal representation of a domain of interaction — within the brain.
The intelligent organism does not so much respond to stimuli as select information that will help it achieve current goals — though a well-designed or evolved system will certainly need to take appropriate account of unexpected changes in its environment. To a first approximation, then, planning is the process whereby the system combines an array of relevant knowledge, to determine a course of action suited to current goals. In its fullest subtlety, planning can involve the refinement of knowledge structures and goal structures, as well as action
per se. While an animal may perceive many aspects of its environment, only a few of these can at any time become the primary locus of interaction.
In general, our thesis is that
perception of an object (at least at the pre-verbal level) involves gaining access to routines for interaction with it, but does not necessarily involve execution of even one of these subroutines. Our image for the control of the ensuing behaviour is context-dependent interpretation (in the sense described in
programs and planning), in that new inputs (such as coming upon an unexpected obstacle) can alter the elaboration of the high-level structures into lower-level tests and actions which in turn call upon the interaction of motor and sensory systems. We study programs which are part of the internal state of the system prior to action, and which can flexibly guide that action in terms of internal goals or drives and external circumstances.
To better appreciate the intimate relation between perception and action, consider the
perceptual cycle (Neisser 1976). The subject actively explores the visual world — for example, by moving eyes, head, or body (or manipulating the environment). Exploration is directed by anticipatory schemas, which Neisser defines as plans for perceptual action as well as readiness for particular kinds of optical structure. The information thus picked up modifies the perceiver's anticipations of certain kinds of information which — thus modified — direct further exploration and become ready for more information. For example, to tell whether or not any coffee is left in a cup we may reach out and tilt the cup to make the interior visible, and keep tilting the cup further and further as we fail to see any coffee, until we either see the coffee at last or conclude that the cup is empty.
Head and Holmes (1911) were perhaps the first to study systematically patients' perceptions of the spatial aspects of their own bodies. They referred to the basis of this perception as the 'postural schema'. This integrated representation of prior movements was held to be updated by each change of position, and to provide a postural model into which all incoming sensations might be integrated.
F. C. Bartlett (1932), who had been much influenced by
Henry Head, introduced the term 'schema' into the psychological literature in the sense of an active organization of past reactions, or of past experiences, which must always be supposed to be operating in any well-adapted organic response. With this, the emphasis shifts from the postural frame to cognitive aspects as revealed in Bartlett's memory experiments.
Workersin
artificial intelligence organize their schema for 'understanding', separating the problem of sensory representation from that of directing action (see Bobrow and Collins 1975 for a sampling of approaches). Minsky (1975) has advanced his concept of 'frames' as a unification of these studies. Here, the stress on recognition of overall contexts which subsume the particularities of the current situation complements the schema-assemblage emphasis on the building up of a representation from familiar sub-parts.
Neisser (1976), influenced by
Gibson, takes a holistic approach with little concern for mechanism and with the schema usually viewed as corresponding to a total situation rather than some localized element of it. He takes explicit account only of those actions which are directed to sampling sensory data. Neisser's use of schema seems to be that of Bartlett, augmented by the Gibsonian view of information pick-up and the resultant stress on the perceptual cycle discussed above.
Another root of the use of 'schema' in current psychology is found in the work of
Jean Piaget. The Piagetian schema is the internal representation of some generalized class of situations, enabling the organism to act in a coordinated fashion over a whole range of analogous situations. Reviewing his approach to the genesis and development of knowledge, Piaget (1971) relates his schemas to the innate releasing mechanisms of the ethologists and thus, via Konrad Lorenz, to the schema of
Kant in the
Critique of Pure Reason (1787). Yet Oldfield and
Zangwill (1942–3) assert that the Head–Bartlett concept of schema has no connection with that of Kant!
The concept of schema has also developed a special meaning in the motor skills literature — for instance, in the work of R. A. Schmidt. Schmidt's schemas seem suited to the performance of a single motion in the laboratory or in sports (such as swinging a bat) rather than to a complex manipulation or to goal-oriented performance in a dynamic environment. Each such schema is broken into two parts: the
recall schema seems akin to feedforward (cf. Fig. 2 in
feedback and feedforward), being responsible for the complete control of a rapid movement, even though environmental feedback may later signal errors. The
recognition schema is responsible for the evaluation of response-produced feedback that makes possible the generation of error information about the movement. It thus seems to combine on-line feedback and identification procedures which may operate even after a movement is completed to better tune the schema for its next activation.
(Published 1987)— Michael A. Arbib
Bibliography- Arbib, M. A. (1972). The Metaphorical Brain.
- Bartlett, F. C. (1932). Remembering.
- Bobrow, D. G., and Collins, A. (1975). Representation and Understanding: Studies in Cognitive Science.
- Craik, K. J. W. (1943). The Nature of Explanation.
- Gibson, J. J. (1977). 'The theory of affordances'. In Shaw, R. E., and Bransford, J. (eds.), Perceiving, Acting and Knowing: Toward an Ecological Psychology.
- Head, H., and Holmes, G. (1911). 'Sensory disturbances from cerebral lesions'. Brain, 34.
- Minsky, M. L. (1975). 'A framework for representing knowledge'. In Winston, P. H. (ed.), The Psychology of Computer Vision.
- Neisser, U. (1976). Cognition and Reality: Principles and Implications of Cognitive Psychology.
- Oldfield, R. C., and Zangwill, O. L. (1942–3). 'Head's concept of the body schema and its application in contemporary British psychology'. British Journal of Psychology, 32–3.
- Piaget, J. (1971). Biology and Knowledge: An Essay on the Relations between Organic Regulations and Cognitive Processes.