Perceptions and Representations

The Theoretical Bases of Brain Research and Psychology

Keith Oatley

• Publisher: Routledge
• Year: 2017
• ISBN:
  9781315515991
• Edition: 1
• Topic: Psychology
• Subtopic: History & Theory in Psychology
• Index: Psychology

Originally published in 1978, this study examines the shortcomings of some theoretical approaches to psychological and neurophysiological mechanisms at the time. Keith Oatley illustrates the extent of these shortcomings by showing how inefficient brain researchers – using their present approaches – would be in trying to understand a computer, which is considerably simpler than the human brain. He concludes that we need better theories than those usually espoused in psychology, and goes on to expound a theory of cognitive representation and inference in perception, which began with Helmholtz more than a hundred years ago but which can now be given substance and formal structure in artificial intelligence programs. The author deploys this theory to give an account of some fundamental problems, such as how we see a three-dimensional world, and how the brain copes so well with incomplete sensory data and with damage to its own components.

1 Lashley and some fundamental problems in psychology

---

 

The three themes of this book are the relation of the brain’s structure to psychological function, the problem of how people perceive successfully in a world where evidence presented to the senses is rather degenerate and fragmentary, and the role of theory in understanding neural and mental processes.

We live in a world in which we perceive objects fulfilling particular functions that serve our purposes. A chair, for instance, is easily recognizable. It is an artificial, solid object in the outside world with a surface raised above the ground, and suitable for supporting the human bottom. It has a back against which a person’s back can rest while he or she is sitting. But a chair which we might see is also a mental construct: it is a mental representation of an object capable of supporting the sitting body.

Although phenomenally we seem to be able to experience such objects in the outside world with immediacy, we do not ‘sense’ them directly. We experience them via the mental representations we have of them. Expanding on this notion, it is possible to show that the world contains an indefinitely large number of objects which we would recognize as chairs, yet they are either in detail or in gross appearance different from one another. Furthermore, we can recognize chairs from an indefinitely large number of viewpoints, in an indefinitely large number of contexts, and with a very substantial degree of occlusion of the chair from our viewpoint. In fig. 1.1, for instance, we have no difficulty in recognizing a curved dark area which indicates the existence of a chair in this scene. Putting it another way, what is presented to our sensory receptors is fragmentary, and very different from the interpretations which constitute our experience and which are in terms not of signals from the retina, or fragments of lines, but in terms of solid objects in the world. No finite set of retinal patterns would specify the class of chairs, yet with a finite brain we do see definite chairs.

Figure 1.1 Fragmentary evidence and context allow us to recognise a chair in this scene.

This kind of puzzling difference between what seems to be the domain of our retinal images and the world of our mental awareness may make one wonder whether ideas are more real than the sensory appearances of the material world. These sensory appearances are at best degenerate facsimiles of reality. Certainly any particular retinal image of a chair from a particular viewpoint is for us a degenerate facsimile, a fleeting two-dimensional projection on the retina of a three-dimensional object. Yet our perception of a particular chair is not of the chair’s image, but of the chair, and is such that we know a great deal about it, whether it would be comfortable, hard or soft, cheap or expensive, what it would look like from the other side and so on.

Plato in book VII of The Republic (e.g. edited by Hamilton and Cairns, 1961) has a simile of man’s predicament. Living consciously in a world of ideas yet physically in a world of appearances, people are like men in a cave; fettered with their eyes looking at one wall, and with their backs to a fire which casts shadows of themselves and of various objects on the wall in front of them. They might see ‘men carrying past the wall implements of all kinds … and human images and shapes of animals as well’. The point Plato wanted to make from this was that the men in the cave think the shadows (which correspond to the degenerate appearances in the physical world) are reality. In Plato’s simile people have to become freed of their fetters, constraining them to look at the shadows, and force their minds to the bright but painful glare of light at the cave’s entrance and make a long difficult journey upwards and out of the cave to the daylight where they can see the real world.

However, the story could go another way. The shadows are like the patterns falling on our receptors. But with a theory of the process of casting shadows, and knowledge of the kinds of objects that could be casting the shadows, we can understand from the distorted and flickering phantoms of shadow what is going on in the middle world of people between the fire and the shadows. Our ideas, or interpretations, guided by the theory of shadow casting, bring us into touch with a reality, indeed closer to a reality, of more practical use to us than the shadows. When conducting ourselves in the world we do not see the flickering shadows on our retina. We actually see our interpretations of what is happening in the world, people, objects, fields, trees and so on, disposed in three-dimensional space. Any demonstration, such as fig. 1.1, indeed a glance around a room where many objects will be partly obscured and all seen from only one direction, will illustrate this. Incomplete figures or cartoons are not as special as they may seem; they merely demonstrate the ability of our minds to cope with fragmentariness. Some of the best demonstrations of this are kinetic ones, such as those of Johansson (1971) of which a single still picture appears in fig. 1.2a. The spots of light which are all that appear on a film which Johansson has made are in fact carried by a person as in fig. 1.2b, and on watching a moving film of just the spots one sees not just moving spots. Orie recognizes a person, with an almost perceptible shadowy body and limbs performing easily recognizable movements of walking, dancing, etc.

An important fact about visual images (retinal or otherwise) is that such appearances are fragmentary (maximally from one viewpoint we can only see one side of an object), yet we experience a very solid complete world. This book is about the question of what processes or mechanisms could be capable of mediating between the shadows that fall on our sense organs and the conscious experience of meaningful reality which we believe we have. How can neural structure subserve this function? What kinds of theories in psychology illuminate the issue of how the brain can perform this type of function, and what organizations of process can begin to achieve interpretations of fragmentary flickering shadows?

Figure 1.2 The pattern of dots in (a) is unrecognisable, but when they move, the outline of a Person is Projected or hallucinated as in (b), (redrawn from Johansson 1971)

Some of the inspiration for this book comes from the work of Karl Lashley. He might be argued to be the founder of physiological psychology. He identified some crucial issues in that subject, and, in particular, the following: the relation of structure to function, the problem of fragmentariness and inconstancy of sensory input, and the issue of how theories of brains illuminate the difficult problem of understanding the mind These three issues are rather aptly illustrated by three quotations from Lashley.

(1) Most of our (psychology) textbooks begin with an exposition of the structure of the brain, and imply that this lays a foundation for a later understanding of behaviour … The chapter on the nervous system seems to provide an excuse for pictures in an otherwise dry and monotonous text. That it has any other function is not clear. (1930)

(2) Here is the dilemma. Nerve impulses are transmitted over definite restricted paths … yet all behaviour seems to be determined by masses of excitation, by the form or relations or proportions of excitation within general fields of activity without regard to particular nerve cells. It is the pattern and not the element that counts. (1942)

(3) I am less impressed with the analogies between various machines and neural activity, such as are discussed in Cybernetics. There has been a curious parallel in the histories of neurological theories and of paranoid delusional systems. In Mesmer’s day the paranoid was persecuted by malicious animal magnetism, his successors by galvanic shocks, by the telegraph, by radio, and by radar, keeping their delusional systems up to date with the latest fashions in physics. Descartes was impressed by the hydraulic figures in the royal gardens and developed a hydraulic theory of the action of the brain. We have since had telephone theories, electrical field theories, and now theories based on the computing machines and automatic rudders. I suggest that we are more likely to find out how the brain works by studying the brain itself and the phenomena of behaviour than by indulging in far-fetched physical analogies. The similarities in such comparisons are the product of an oversimplification of the problems of behaviour … (1951)

Given these three quotations I follow some of Lashley’s recommendations. There is no chapter on the structure of the nervous system. But this book does contain some discussion of how far anatomy and related studies are important in explanations of behaviour. I will also try to bring up to date the problem that Lashley regarded as a central issue in his experimental work, the question of how the brain operates successfully despite damage to its parts, as well as fragmentariness and variation of details of patterns on the retina.

I make no attempt at making complete presentations of empirical data. There are plenty of texts that tackle that task. Instead I concentrate on theories and explanations rather than experiments, and though I touch on some of the issues of learning and motivation, for the most part I shall use the issues of visual perception to carry the arguments forward.

Lashley very much gave the impression of having followed his own advice of the third quotation given above, and ‘studied the brain itself’. As a result of this he was able to delineate at least some of the important conditions that must be fulfilled by theories of how the brain produces behaviour. Yet despite the fact that his advice is often good, I take exactly the opposite view to Lashley at this point, and will pursue more or less farfetched analogies with machines of various kinds, particularly computers.

Though Lashley studied the brain and behaviour, and was an astute theoretician, I will argue that his principal difficulty was precisely that he was hampered by the lack of apt physical analogies for the processes he studied. His biological work hinted at the existence of principles of organization necessary for behaviour, and this work predated the understanding of any such principles embodied in physical systems. But these processes remained mysterious, because without making analogies with systems that we are already familiar with or can become familiar with, we find it difficult or impossible to grasp the unfamiliar brain. Analogies are theories, are explanations in familiar terms, are metaphors, are understandings. Without them studying the brain and behaviour may be tantalizing, but ultimately it is sterile. Somewhat more than twenty years after Lashley’s death the ‘similarities in such comparisons’ with various kinds of machines may no longer be just the product of oversimplification, but may allow us to discern the nature of principles that make behaviour possible. Any understanding of the brain is a theory, is an analogy. Oversimplication has certainly occurred, and even to an absurd degree. No doubt it will continue. But not because analogies have been drawn from the latest fashion in physics. Rather it has taken place because the particular analogies or theories upon which brain researchers have drawn (apparently unconsciously at times) are too impoverished to give any insight into the workings of a complex brain. Lashley himself sought an apposite metaphor from the physical world, but the best he could come up with to give some idea of how it is the ‘pattern and not the element that counts’ was a system of diffraction patterns of waves on an inhomogeneous liquid. This is not just an arbitrary example of someone searching for the form of a theory by drawing on a metaphor from the furniture of our ordinary world. I will argue that this kind of physical analogy is typical, and that brain research has been continuously handicapped as much by the lack of sufficiently powerful metaphors or analogies as by any other factor. Now with the beginnings of understanding of how to create intelligent processes in computers we are armed with a set of analogies or metaphors altogether richer and more adequate for thinking about brain function than before.

Lashley was, of course, right to rail against oversimplification; and the oversimplification which he principally railed against was that of behaviour produced by reflex connections, conditioned or otherwise. We are now in a position to go further with the kind of theory that belatedly should replace the reflex, and as it happens the attempt to design aritficial intelligence in computers has helped to begin to see what form it should take, whether computers constitute the latest fashion in physics or not.

Chapters 2 and 3, amongst other things, involve an argument that in one important sense of the word ‘understand’ we only understand physiological mechanisms in so far as we have good psychological theories, i.e. theories of the logical structure of behaviour. The book thereafter concentrates upon matters in which we have some understanding of the necessary logical structure for particular types of behaviour or mental process, and I discuss how our psychological theories affect our interpretation of physiology.

A number of matters often considered to be part of physiological psychology, or brain research, or neurobiology, do not come within the scope of this account; I hope because they are either tangential to understanding behaviour and mental processes, or at best contribute to it only rather slightly. Thus the laborious pursuit of the conclusion that brain area A seems to have something to do with behaviour X or the research that allows us to admire the intricacy of some elaborate network in the brain without being able to say what it does or how it does it is not discussed to any large extent.

Two things which would seem appropriate in a book on physiological psychology would be to discuss first how known properties of neurones and neural nets might constrain, suggest and ultimately constitute explanations of how behaviour is produced, and second, how our understanding of mental processes and behaviour inform our understanding of their physiological embodiment. Though not dismissing the importance of this first concern, I concentrate largely on the second. These two seem to (or perhaps ought to) be principal concerns of physiological psychology.

Perhaps the central theme of the book is the exploration of how theory pervades scientific activity in brain research and, indeed, any activity. In brain research some kinds of experiment, e.g. recording electrical activity from a nerve cell, seem to lead to the conclusion that we can have direct contact with facts about the brain, without the impediment of theory or speculation which will necessarily distort or deceive: what we need to understand about the brain are plenty of these facts. I have tried to show within the context of brain research and experimental psychology that such a stance is shaky, if not misleading. To talk about such observations as facts is still to hold a theory, but a theory that is not expressed or even recognized as such is so much the worse for its non-recognition.

It may seem surprising but the implicit theory which still pervades most brain research and much psychology is the theory of reflexes, and this theory can be shown to be distinctly limited.

Reflex theory has had a good long run (more than 300 years) and we have by now understood quite a lot of its limitations. There are very many. Rather than pretending that the terms of this theory are fact, and at the same time being continually unable to come to grips with the more interesting aspects of the brain, what we need are new and better ways of thinking about the subject. Such ways are at last becoming available in a form that can be used.

2 The brain as a physical system

---

Lashley’s prescription, as discussed briefly in chapter 1, was that in order to find out how the brain works we should study the brain itself. By this he presumably meant that in order to understand how the brain works to embody learning, perception, language, cognition and various other interesting, intelligent mental processes and behaviour, we should study the brain itself. Lashley made it clear in his work that he did not want to understand the brain as a factory for the manufacture of strange...