Eye and Brain: The Psychology of Seeing - Fifth Edition / Edition 5 available in Paperback, eBook
Eye and Brain: The Psychology of Seeing - Fifth Edition / Edition 5
- ISBN-10:
- 0691165165
- ISBN-13:
- 9780691165165
- Pub. Date:
- 09/15/2015
- Publisher:
- Princeton University Press
- ISBN-10:
- 0691165165
- ISBN-13:
- 9780691165165
- Pub. Date:
- 09/15/2015
- Publisher:
- Princeton University Press
Eye and Brain: The Psychology of Seeing - Fifth Edition / Edition 5
Buy New
$20.95Buy Used
$11.62
-
$11.62
-
SHIP THIS ITEM
Temporarily Out of Stock Online
Please check back later for updated availability.
-
Overview
Illusion continues to be a major theme in the book, which provides a comprehensive classification system. There are also sections on what babies see and how they learn to see, on motion perception, the relationship between vision and consciousness, and on the impact of new brain imaging techniques.
Product Details
| ISBN-13: | 9780691165165 |
|---|---|
| Publisher: | Princeton University Press |
| Publication date: | 09/15/2015 |
| Series: | Princeton Science Library , #38 |
| Edition description: | Fifth |
| Pages: | 296 |
| Sales rank: | 1,134,076 |
| Product dimensions: | 5.50(w) x 8.40(h) x 0.70(d) |
About the Author
Read an Excerpt
Eye and Brain
The Psychology of Seeing
By Richard L. Gregory
PRINCETON UNIVERSITY PRESS
Copyright © 1997 Richard L. GregoryAll rights reserved.
ISBN: 978-0-691-04837-6
CHAPTER 1
Visions of vision
The eye is a simple optical instrument. With internal images projected from objects in the outside world, it is Plato's cave with a lens. The brain is the engine of understanding. There is nothing closer to our intimate experiences, yet the brain is less understood and more mysterious than a distant star.
We have only to open our eyes, and spread before us lies a banquet of colours and shapes, shadows, and textures: a pageant of rewarding and threatening objects, miraculously captured by sight. All this, from two tiny distorted upside-down patterns of light in the eyes. Seeing is so familiar, apparently so easy, it takes a leap of imagination to appreciate that the eyes set extremely difficult problems for the brain to solve for seeing to be possible. How does it work? How are ghostly images transformed into appearance of solid objects, lying in an outer world of space and time?
From the beginnings of recorded questioning there have been several approaches to how we see. These are very different from current views. An essential problem was how distant objects reached eye and brain, while remaining out there in space. Two and half millennia ago, Greek philosophers thought that light shoots out of the eyes, to touch objects as probing fingers. A different notion at that time was that objects have expanding 'shells' like ripples from a stone dropped on a pool, but maintaining the object's shape to great distances. Called 'sense data' until quite recently by philosophers, they were supposed to be intermediaries—neither matter nor mind—between objects and perceptions. Both of these ideas were serious candidates before it was realized that in the eyes there are images of light, optically projected from the outside world onto the screens of the retinas. Optical images were unknown before the tenth century, and not until the start of the seventeenth were images discovered in eyes. At last it became clear that light does not enter or leave the brain, locked privily in its box of bone. All the brain receives are minute electrochemical pulses of various frequencies, as signals from the senses. The signals must be read by rules and knowledge to make sense. Yet what we see, and what we know, or believe, can be very different. As science advances, differences between perceived appearances and accepted realities become ever greater.
This is far beyond the common account that the eye is a camera; yet this is essentially true, though far from the whole story. It is the uncamera-like features of eyes and brains that most interest us here.
What is striking is the huge amount of brain contributing to vision, giving immense added value to the images of the eyes. Where does this extra richness for vision come from? By some authorities it is simply denied—they see perception as passive acceptance of what is out there, as a window facing the world. But this does not begin to explain how we see objects from the sketchy images of the eyes, even from sparse lines and crude dots of seemingly inadequate pictures. In ideal conditions, object perception is far richer than any possible images in the eyes. The added value must come from dynamic brain processes, employing knowledge stored from the past, to see the present and predict the immediate future. Prediction has immense survival value. It not only makes fast games possible in spite of the physiological signal delays from eye to brain, and brain to hand. Anticipating dangers and potential rewards is essential for survival—made possible by buying time from seeing objects distant in space.
This introduces a particular kind of way of thinking about perception. It is essentially the view of the nineteenth century German polymath—physiologist, physicist, psychologist—Hermann von Helmholtz (1821–1894) who described perceptions as 'unconscious inferences' from sensory data to what might be out there. This is the 'school of thought' accepted here, but there are others. Psychology is unusual among the sciences, in having doubts of its most basic assumptions, with very different alternatives held by different authorities. 'Active' and 'passive' accounts are extremely different ways of describing and explaining phenomena of vision. Few other sciences have such divisions in their basic ideas. There were equally dividing paradigms (as the American philosopher of science Thomas Kuhn calls them) at the time of Darwin, over whether there was evolution of species by natural selection or special creation of each species, but this is now resolved, by almost universal acceptance of evolution. However, although active, essentially Helmholtzian, accounts of perception are now dominant, this was not so a few years ago, and they are not universally held today.
Paradigms of perception
It might be useful to outline some recent kinds of explanations:
Behaviourism was founded by John Broadus Watson (1878–1958) with his manifesto of 1913: 'Psychology as the behaviourist views it'. This set out to deny consciousness, at least as a ploy to make psychology scientifically respectable. Behaviourism was extremely influential in America until the 1980s, especially with the experiments and ideas of B. F. Skinner. It is based on the earlier work of the Russian physiologist Ivan Petrovich Pavlov (1849–1936) with his experiments on conditioned (or 'conditional') reflexes. Pavlov showed that, starting with an innate (inherited) reflex, such as salivating to the sight or smell of meat, dogs would come to salivate to any stimulus (such as a bell) presented at the same time or just before the food. It proved possible to build up chains of conditioned reflexes. For the behaviourists, it seemed that chains of conditioning would explain all learned behaviour, even language.
They listed innate reflexes observed in babies, and measured strengths of drives for rewards. So they developed a scientifically respectable-looking 'atomism' for describing complex behaviour from simple components.
Problem solving, at least for animals such as cats, was supposed to be by trial-and-error—without insight into the nature of a problem. Perception and behaviour were supposed to be controlled quite directly by stimuli, with modifications from internal states of drives such as hunger, so that with sufficient knowledge psychology should become a perfectly predictive science. This has not worked out. Watson's denial of consciousness makes psychology even more like physics; but for most of us today, it threw the baby out with the bath water.
Gestalt psychology was a very different rival school. Founded by a group of German scientists in the 1920s, the emphasis was on dynamics and 'holism'. Many of the Gestalt psychologists (entirely different from recent Gestalt therapy) fled from Nazi Germany under Hitler, to settle in America where they had a major influence.
A 'Gestalt' was a grouping of elements such that the whole is greater than the sum of its parts. Analysis into perceptual components was not supposed to be possible. An important concept was 'pregnance', roughly, 'pregnant with meaning'. Problem solving was supposed to be by 'insight'. A famous example is Wolfgang Kohler's chimpanzee Sultan; presented with a banana out of reach and a number of short sticks, he was described as looking at the sticks for several minutes, then suddenly joining two together to pull down the banana. Throughout there is an emphasis on sudden solutions, with sudden insights.
The Gestalt psychologists described visual perceptions as more than the sum of stimuli, organized according to various laws. These were mainly derived from subjective reports of how arrangements of dots are seen as patterns: which dots 'belong' together, form lines and so on, or are separate. This may seem vague, hardly 'scientific'; but the Gestalt laws of organization have turned out to be important for perception of sight and sound. They have been taken up by the artificial intelligence (Al) community, especially for programming computers to recognize patterns and objects. The laws include:
(1) closure—tendency for a roughly circular patterns of dots to be seen as 'belonging' to and forming an object;
(2) common fate—parts moving together, as leaves of a tree, seen as an object;
(3) contiguity of close-together features; and a preference for smooth curves.
The laws of organization were supposed to be inherited, but as they correspond to common features of almost all objects, learning could be involved, to give us all much the same visual organizations. Gestalt notions of brain physiology (electrical fields and so on) have been abandoned.
Cognitive psychology, in its various forms, denies that perception and behaviour are controlled by stimuli, emphasizing the importance of general background knowledge and more-or-less logical thought processes. How far these apply to perception is controversial. Generally, visual perception has been thought of as quite separate from cognitive problem solving, but this can be questioned. However, Hermann von Helmholtz did think of visual perceptions as unconscious inferences, and so related perception to thinking.
The Cambridge psychologist Kenneth Craik (1914–45) put forward the notion that the brain works with physiologically existing functional 'internal models' of perceived and imagined objects and situations. This is now generally modified to a more symbolic account; but the notion of representing by the brain is accepted as central to cognitive approaches.
The intelligent eye
This philosophy, or paradigm, is largely derived from Helmholtz. It is, that visual and other perception is intelligent decision-taking, from limited sensory evidence. The essential point is that sensory signals are not adequate for direct or certain perceptions; so intelligent guessing is needed for seeing objects. The view taken here is that perceptions are predictive, never entirely certain, hypotheses of what may be out there.
It was, perhaps, the active intelligence of perception that was the evolutionary start of conceptual problem-solving intelligence. When, a generation before Freud, Helmholtz called perceptions unconscious inferences he was much criticized—for how could blame or praise be applied to unconscious perceptions, and actions? We are still puzzled by these issues.
There are many traps along the way of exploring Eye and Brain. It is important to avoid the temptation of thinking that eyes produce pictures in the brain which are perceptions of objects. The pictures-in-the-brain notion suggests an internal eye to see them. But this would need a further eye to see its picture—another picture, another eye—and so on forever, without getting anywhere. Early this century, the Gestalt psychologists held that perceptions were pictures inside the brain: supposed electrical brain fields copying forms of objects. So a circular object would produce a circular brain field. Presumably a house would have a house-shaped electrical brain-picture, though this is far less plausible. A green object having a green brain-trace is ridiculous. This notion, known as isomorphism, led to supposing that properties of brain fields produce visual distortions (like bubbles tending to be spherical), visual phenomena being explained by their supposed mechanical or electrical properties. There is no evidence for isomorphic brain traces.
We now think of the brain as representing, rather as the symbols of language represent characteristics of things, although the shapes and sounds of language are quite different from whatever is being represented. Language requires rules of grammar (syntax), and meanings of symbols (semantics). Both seem necessary for processes of vision; though its syntax and semantics are implicit, to be discovered by experiment.
Some puzzles of vision disappear with a little thought. It is no special problem that the eyes' images are upside down and optically right–left reversed—for they are not seen, as pictures, by an inner eye. As the image is not an object of perception, it does not matter that it is inverted. The brain's task is not to see retinal images, but to relate signals from the eyes to objects of the external world, as essentially known by touch. Exploratory touch is very important for vision. It matters that touch-vision relations remain unchanged. When changed experimentally (with optically reversing prisms, or lenses or mirrors) then a problem is set up, and special learning is required. No special learning is needed for a baby to see the world the right way up.
The Gestalt psychologists made some excellent suggestions, realizing that the visual system has to solve some very difficult problems, which arise right at the start. How does the mosaic of retinal stimulation give perception of individual objects? (This also applies to hearing, especially of speech. We can only distinguish separate words in familiar languages). The visual separations of objects are not given simply by borders of light on the retinas. Separations into objects is given by various rules, and by knowledge. Sharp borders are rather rare, except for line drawings, which are not typical and present their own problems.
The tendency to group elements into wholes was investigated by the Gestalt psychologists, with patterns of dots. Their experiments suggested various rules of organization for object creation. This was a much more useful idea than isomophism, and it has turned out to be important for programming visual computers, though this development is in its infancy. We can see something of dynamic grouping in an array of dots (see Figure 1.1). The dots are equally spaced, but there is a tendency to 'organize' them into columns and rows. We can see in ourselves active try-outs for organizing visual data into objects. With more complexity, groupings and re-groupings can become more dramatic (Figure 1.2).
If the brain were not continually trying out organizations of data, for searching for objects, such as faces, the cartoonist would have a hard time. In fact all he or she has to do is to present a few well-chosen lines and we see a face, complete with an expression. This essential process of vision can, however, go over the top to make us see faces in the fire, galleons in the clouds, or the Man in the Moon. Vision is certainly not infallible. This is largely because knowledge and assumptions add so much that vision is not directly related to the eyes' images or limited by them—so quite often it produces fictions. This can be useful, as images are inherently inadequate, but visual fictions, and other illusions, worry philosophers seeking certainty from sight.
How did such complex processes for representing things start? What is their evolutionary benefit? For simple organisms, the eyes' signals do initiate behaviour quite directly. Thus tropisms toward or away from light may serve to find protection or food in typical conditions, without the creature being capable of making decisions between alternative courses of action. We might say that primitive organisms are almost entirely controlled—tyrannized—by tropisms and reflexes. Many reflexes still protect us (such as blinking to a puff of air on the eye, or to a sudden loud sound) and reflexes are essential for the maintenance of body functions such as breathing and digestion. But gradually, through evolution, direct control from outside objects has been largely replaced by more and more indirect representations of objects and situations. This has the huge advantage that behaviour can be appropriate to properties of objects that are not and often cannot be signalled by the senses. Thus we pick up a glass to drink not simply from stimuli, but from knowledge of glasses, and what they may contain. By contrast, a frog surrounded by dead flies will starve to death, for though they are edible it does not see them, as they do not move.
These brain representations are far more than pictures. They include information of what various kinds of objects may do, or be used for. For behaviour to be appropriate in a wide variety of situations requires a great deal of knowledge of the world. Knowledge must be selected and accessed within a fraction of a second to be useful for perception, or the moment for action (or survival) will pass. So the intelligence of vision works much faster than other problem solving. This may be why perceptions are quite surprisingly separate from generally more abstract conceptions, and may disagree. Thus, one experiences an illusion, though one knows it is an illusion and even what causes it. Illusions tell us a great deal—sometimes, as I shall show, more than we would wish to know!
(Continues...)
Excerpted from Eye and Brain by Richard L. Gregory. Copyright © 1997 Richard L. Gregory. Excerpted by permission of PRINCETON UNIVERSITY PRESS.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Table of Contents
Pretext
1 Visions of vision 1
2 Light 14
3 Eye 24
4 Brain 67
5 Seeing brightness 84
6 Seeing movement 98
7 Seeing colours 121
8 Learning how to see 136
9 Realities of art 170
10 Illusions 194
11 Speculations 244
Bibliography and notes 256
Index 269