v [This] book is... a proposed solution of a specific problem:
the origin of the nervous system's unique ability to produce adaptive behaviour.
p.1 I hope to show that a system can be both mechanistic in
nature and yet produce behaviour that is adaptive. I hope to show that the essential difference between the brain
and any machine yet made is that the brain makes extensive use of a method hitherto little used in machines. I hope
to show that by the use of this method a machine's behaviour may be made as adaptive as we please, and that the method
may be capable of explaining even the adaptiveness of Man.
p.11 The direct intervention of consciousness is evidently not necessary for adaptive learning.
p.28-29 we must note that theories are of various types. At one extreme is Newton's theory
of gravitation - at once simple, and precise, and exactly true. When such a combination is possible, Science is indeed lucky!
Darwin's theory, on the other hand, is not so simple, is of quite low accuracy numerically, and is true only
in a partial sense - that the simple arguments usually used to apply it in practice... are gross simplifications
of the complex of events that will actually occur.
The theory attempted in this book is of the latter type.
p.38 'In most cases the change which induces a reaction is brought about by the organism's own movements.
These cause a change in the relation of the organism to the environment: to these changes the organism reacts...' (Jennings.)
'The good player of a quick ball game, the surgeon conducting an operation,
the physician arriving at a clinical decision - in each case there is a flow from signals interpreted to action carried
out, back to further signals and on again to more action, up to the culminating point of the achievement of the task'.
(Bartlett.)
'Organism and environment form a whole and must be viewed as such.' (Starling.)
p.42 That an animal should remain 'alive', certain variables must remain without certain 'physiological'
limits. What these variables are, and what the limits, are fixed when the species is fixed.
p.42-43 Every species has a number of variables which are closely related to survival and which
are closely linked dynamically so that marked changes in any one leads sooner or later to marked changes in the others...
These important and closely linked variables will be referred to as the essential variables
of the animal.
How are we to discover them, considering that we may not use borrowed knowledge
but must find them by the methods of Chapter 2? There is no difficulty. [JLJ - Yeah right...] Given a species,
we observe what follows when members of the species are started from a variety of initial states. We shall find that large
initial changes in some variables are followed in the system by merely transient deviations, while large initial changes in
others are followed by deviations that become ever greater til the 'machine' changes to something very different from what
it was originally. The results of these primary operations will thus distinguish, quite objectively, the essential variables
from the others.
p.43 We can now define 'survival' objectively and in terms of a field: it occurs
when a line of behaviour takes no essential variable outside given limits.
p.58 I propose the definition that a form of behaviour is adaptive if it maintains the essential
variables within physiological limits.
p.58 I shall outline the facts underlying Cannon's concept of 'homeostasis'...
(1) Each mechanism is 'adapted' to its end.
(2) Its end is the maintenance of the values of some essential variables within physiological limits.
(3) Almost all the behaviour of an animal's vegetative system is due to such mechanisms. [JLJ - The vegetative
systems are those body systems that support the living process, that neither require nor greatly benefit from conscious attention.]
p.64 We can now recognize that 'adaptive' behaviour is equivalent to the behaviour of a stable system,
the region of the stability being the region of the phase-space in which all the essential variables lie within their normal
limits.
p.64 'Every phase of activity in a living being must be not
only a necessary sequence of some antecedent change in its environment, but must be so adapted to this change as to tend to
its neutralisation, and so to the survival of the organism... It must also apply to all the relations of
living beings. It must therefore be the guiding principle, not only in physiology... but also in the other
branches of biology which treat of the relations of the living animal to its environment and of the factors determining its
survival in the struggle for existence.' (Starling.)
p.80 Chapter 7 The Ultrastable System
p.82 To be adapted, the organism, guided by information from the environment,
must control its essential variables, forcing them to go within the proper limits, by so manipulating the environment (through
motor control of it) that the environment then acts on them appropriately.
p.82 It is axiomatic (for any Black Box when the range of its inputs
is given) that the only way in which the nature of its contents can be elicited is by the transmission of actions
through it. This means that input-values must be given, output-values observed, and the relationships in the paired values
noticed. In the kitten's case, this means that the kitten must do various things to the environment and must
later act in accordance with how these actions affected the essential variables. In other words, it must proceed by trial
and error.
p.82-83 Adaptation by trial and error is sometimes treated in psychological
writings as if it were merely one way of adaptation, and an inferior way at that. The argument given above shows that the
method of trial and error holds a much more fundamental place in the methods of adaptation. The argument shows, in fact, that
when the organism has to adapt (to get its essential variables within physiological limits) by working through an
environment that is of the nature of a Black Box, then the process of trial and error is necessary, for only such
a process can elicit the required information.
p.83 The process of trial and error... the process may be playing
the invaluable part of gathering information, information that is absolutely necessary if adaptation is to be successfully
achieved. It is for this reason that the process must enter into the kitten's adaptation.
As the kitten proceeds by trial and error, its final behaviour will
depend on the outcome of the trials, on how the essential variables have been affected. This is equivalent to saying that
the essential variables are to have an effect on which behaviour the kitten will produce
p.85 any system that has essential variables with given limits,
and that adapts by the process of testing various behaviours by how each affects ultimately the essential variables, must
have a second feedback formally identical (isomorphic) with that described here. This deduction holds equally for brains living
and mechanical.
p.85 We will now take the mechanism and ask: Given such a mechanism, in
whatever material form, will it necessarily show adaptive behaviour?
p.119 Even if the ultrastable system is suitably
arranged - if the critical states are encountered before the essential variables reach their limits - it usually cannot
adapt to an environment that behaves with sudden discontinuities.
p.120 The Homeostat's method for achieving adaptation is thus essentially
useless when its environment contains such 'lethal' discontinuities.
p.120 there can be little doubt that on many occasions living organisms
have missed success either by abandoning a trial too quickly, or by persisting too long with a trial that was actually
useless.
p.121 If we grade an ultrastable system's environments according
to the difficulty they present... at the difficult end are those that contain many variables richly cross-linked to form a
complex whole.
p.122-123 Chapter 9 Ultrastability in the Organism... Many
simple bodies have one diagram that is so obtrusive that one is apt to think of it as the specification of how the
parts are joined... But we must beware of thinking that it is the only pattern; for there are also systems
whose parts or variables have no particular position in space relative to one another, but are related dynamically in some
quite different way... the 'system' is a set of concentrations, and the diagram of immediate effects shows how the
concentrations affect one another.
p.127 What is necessary that any material entity should serve as a step-mechanism in ultrastability?
Only that it should be a step-mechanism, that it should be able to be changed by the essential variables, and that it should
have an effective action on the reacting part R.
p.131 The ultrastable system is thus the appropriate form for the organism if the disturbances that come
to it from the world around fall into two clearly defined classes:
(1) Frequent (or even continuous) small impulsive disturbances to the main variables.
(2) Occasional changes, of step-function form, to its parameters.
The ultrastable system... may, in some cases, actually be the optimal mechanism by which an organism can
ensure its survival.
p.192 We... can go on to consider the problem of how a large and complex organism can adapt to a
large and complex environment without taking the almost infinite time suggested by S. 11/5.
p.193 Many of the variables, often the majority, are constant over appreciable intervals
of time, and thus behave as part-functions... If one looks around one, only in the most chaotic surroundings will
all the variables be changing. This constancy, this commonness of part-functions, must, by S. 12/14, be due to commonness
of states of equilibrium in the parts that compose the terrestrial environment. Thus the environment of the living organism
tends typically to consist of parts that are rich in states of equilibrium... Associated with this constancy... is the fact
that most variables of the environment have an immediate effect on only a few of the totality of variables...
We are, in fact, led again to consider the properties of a system whose connexions are fluctuating and conditional - the type
encountered before in S. 11/12 and therefore treatable by the same method. I suggest, therefore, that most of the environments
encountered on this earth by living organisms contain many part-functions. Conversely, a system of part-functions adequately
represents a very wide class of commonly occurring environments.
p.195 the set of variables active at one moment will often be different from the set active at another.
The pattern of activity within the environment will therefore tend... to be fluctuating and conditional rather than
invariant. As an animal interacts with its environment, the observer will see that the activity in the environment is limited
now to this set, now to that. If one set persists active for a long time and the rest remains inactive and inconspicuous,
the observer may, if he pleases, call the first set 'the' environment. And if later the activity changes to another set he
may, if he pleases, call it a 'second' environment.
p.210 In other words, within a multistable system, subsystem adapts to subsystem in exactly the same
way as animal adapts to environment. Trial and error will appear to be used; and, when the process
is completed, the activities of the two parts will show co-ordination to the common end of maintaining the essential variables
of the double system within their proper limits.
p.214 We are now in a position to summarize the answer, given by the intervening chapters, to the
objection, raised in S. 11/2, that ultrastability cannot be the mode of adaptation used by living
organisms because it would take too long. We can now appreciate that the objection was unwittingly
using the assumption that the organism and the environment were richly joined both within themselves and to each other. Evidence
has been given, in S. 15/2, that the actual richness is by no means high. Then Chapters
15 and 16 have shown that when it is not high, adaptation by ultrastability can occur in a time that is no longer impossibly
long.
