As observers we expext the environment to change and try to describe those features that remain unchanged with the passage of time. An unchanging from of events due to the acticvity within an assembly is called a behaviour. The behaviour of a steam engine is a recurrent cyclr of steam injection and piston movements that remains invariant. The behaviour of a cat is made up of performances like eating and sleeping and, once again, it is an invariant form selected from the multitude of things a cat might possibly do. The behaviour of a statue is a special case, for the statue is immobile, or to use an equivalent formalism, it changes at each instant of time into itslef. We shall neglect the special case entirely. An "assembly" is the dynamic part of an observer's environment, a piece of the real world, which is freely supplied with energy. Although the energetics do not immediately concern us, the assembly embodies one or many more or less regular modes of dissipating the energy -- steam expansion or metabolism -- as a result of which it prodecues an unlimited supply of observable events. (ATC 18)

It is easy to cite brain models which are merely imitations; most well-behaved robots, most of the tiny automata that imitate a naughts and crosses player, nearly all of the maze solving puzzle machines (though there are some, like Deutsch's Rat, which are used explicitly to illustrate an organizational principle rather than to imitate a response). There are not so many cybernetic models to choose from, butone of them, made by Ashby and called the Homeostat, admirably issustrates the distinction. It is made up of four interacting regulators and an independent switching mechanism which changes the interconnections between these elements until a stable arrangement is reached. It can (from the viewpoint of psychology and engineering respectively) be dubbed a "brain-like analogue" and a "device for solving differential equations", for it does, rather imperfectly, display a brain-like bahaviour and it will, rather eccentrically, solve differential equations. Its imperfections as an equation solver (which it is not meant to be) are obvious from its construction and have met with a good deal of heavy-handed criticism. Its imperfections as a brain-like analogue (which, once again, it is not meant to be) occur because at the level of functional analogy the organization of a homeostat is not particularly brainlike. It is only when we come to the level intended in the cybernetic abstraction that the self-regulation in a homeostat is identical with the self-regulation in a brain, and with reference to this feature the homeostat is a cybernetic model of all brains. (ATC 17)

Given a well-defined set of elements, it is possible to measure the amount of uncertainty with reference to this set. The reference frame provides a set of states, hence a measure of uncertainty is possible and is called the variety of the set. The simplest case is the system where at any instant, each state is equally likely to occur. Since there are n states, an observer is initially uncertain about "which of n", or conversely, the appearance of one particular state removes this uncertainty and conveys an "amount of information", selecting one of n possibilities. Information and uncertainty, if expressed in an additive form as logarithmic measures, are very simply related indeed;

The controller must "keep the student's attention" which is a special case of "requisite variety". The student is a system with given variety of behaviour, say u; that is, he must attend to something. u is a measure of the rate at which data of some kind must be processed, or decisions of some kind made, in order that the system shall have the status of a "student". Suppose, then, that he does attend to the problem display. The variety of a problem, with reference to the student -- for short -- its "difficulty", is the amount of decision making needed to reach a solution (imagine a choice process, whereby uncertainty about a response is reduced until one response is actually made). Now to keep the student's attention the controller must select a sequence of problems which have an average "difficulty" at least equal to u. Unless it does, the student will daydream. Unfortunately, if it does, there is no gaurantee that he will not. But, given the matching condition to be cited in 2, t is an estimate of overall difficulty and the defect selection tends to satisfy the requisite variety condition. (ATC 93)

A great deal of cybernetics is concerned with how stability is maintained with "control mechanisms". One of the first of these to be treated explicitly was Watt's invention of the governor (a theoretical analysis was offered by Maxwell in 1865). The device illustrates a principle called negative feedback. A signal, indicating the speed of a steam engine, is conveyed to a power amplifying device (in this case, a steam throttle) in such a way that when the engine accelerates the steam supply is reduced. Hence, the speed is kept stable. The signalling arrangement is independent of energetic considerations, and it is legitimate to envisage the governor as a device which feeds back information in order to effect speed control. (ATC 12)

At the level of systems, there is no difference between biological and mechanical control. But sometimes the biological controller, as well as the control system, has a ready mechanical analogue. When a limb is moved from position X to another, Y, the muscular contraction depends upon the frequence of nerve impulses arriving at the muscular end plates. Stretch receptions in the muscle signal the degree of contraction along "proprioceptive" fibres and this feedback to various parts of the brain which are concerned specifically inhibits motor activity and stabilizes the motion. The whole process is monitored by a further, often visual, feedback which conveys a difference signal Y-X which is 0 when the act is completed. (ATC 71)

For the most part, the subsequent discussion is confined to conversations that highlight learning and have educational significance. In this context the main tenet of relativism was previewed at the end of Section 1.4; there is no theory of learning apart from a theory of teaching; no theory of teaching apart from a theory of learning. There is a theory of learning and teaching together and that is all. Results from many studies support this point of view as, also, does the evidence of commonsense. So, to be dogmatic (but with confidence) learning implies teaching and teaching implies learning. Sometimes the teacher and learner responsible for th joint process are obvious (a student at a desk and another person wearing an academic cap and gown). Sometimes, the teacher and learner are not so obviously distinct and turn out to be unexpected but, once-indicated, intuitively plausible entities. (CCL 33)

Though brains, human or animal, are often associated with cognitive operations, neither biological fabric nor any other kind of fabric is responsible (except in an incidental way) for the peculiar nature of cognition. Without contesting the utility and cogency of arguments from evolution (the gradual development of language, problem solving capability, awareness and the like) they tell, as they stand, a one sided tale about a ubiquitous and many faceted event. The present theory does not "biologicise" man-kind or mind; nor is it reductionist. Cognition may occur at the level of groups of people (so social awareness is taken in earnest, not as a collation of individual awarenesses) or it may characterize the activity of slightly unconventional computing machines. (CCL 1-2)S

Home Cybeneticians Gordon Pask Symbolic generalizations

Symbolic Generalizations | Beliefs, Models, and Analogies | Values | Exemplars | Techniques | Guiding Questions