Visual Perception

An Introduction, 3rd Edition

Nicholas Wade, Mike Swanston

• Publisher: Psychology Press
• Year: 2013
• ISBN:
  9781136178290
• Edition: 1
• Topic: Psicología
• Subtopic: Historia y teoría en psicología

Does the world appear the same to everyone? Does what we know determine what we see? Why do we see the world as we do?

Vision is our most dominant sense. From the light that enters our eyes to the complex cognitive processes that follow, we derive most of our information about what things are, where they are, and how they move from our vision.

Visual Perception takes a refreshingly different approach to this enigmatic sense. From the function that vision serves for an active observer, to the history of visual perception itself the third edition has been extensively revised, updated and expanded, while still preserving the essential features of historical context, neurophysiology and independent thought that made the earlier editions so engaging.

Covering the perception of location, motion, object recognition and with up-to-date information on the workings of the visual brain, the 3rd edition looks at how our ideas have been shaped, not just by psychology, but by art, optics, biology and philosophy. The emphasis on understanding vision as a basis for action in the real world has also been expanded to cover seeing representations of all sorts, whether they are pictures or computer-generated displays.

The 3rd Edition of Visual Perception is a readable, accessible and truly relevant introduction to the world of perception and will be welcomed by students of visual perception as well as anyone with a general interest in the mysteries and wonder of vision.

1

Understanding visual perception

All living things actively respond to the environment in which they exist; indeed, this may be close to a definition of what it means to be alive. We, descendants of the first land-living vertebrates, have a number of specialised senses that have evolved to sustain survival – to make and maintain adaptive contact with the environment. Receptors for sources of energy in the terrestrial environment that have proved beneficial for survival have evolved and become specialised for the needs of each species. Through the action of their senses, living things seek shelter from the elements and other organisms as well as sustenance and sexual partners in order to survive and reproduce. Thus, environmental features that afford protection can be identified and exploited. Sensitivity to the visual, auditory, olfactory, and tactile characteristics of other living things permits identification of foods as well as potential mates or rivals.

By emphasising the visual dimension of our environmental interactions in this book, we do not wish to suggest that vision operates in isolation from other senses. It is clearly insufficient to study vision alone, as all our senses are intimately integrated, and we have a unified perception of external objects. Perception engages all the senses, but we possess the unique ability to express our perceptual experiences through language, which tends to reflect the operation of particular senses. Thus, it might seem from the way we speak about perception as though the senses work in isolation rather than in concert. The vocabulary devoted to each of the senses is not evenly distributed either. Vision has the lion’s share of words as well as the work associated with our perceptual experiences.

Human senses are linked to an intricately organised brain, which has evolved to do more than simply identify and act on the necessities for survival. Humans not only use their senses to survive; they also reflect on their experience through language and thought. Through language, we know that our own and other people’s perception of the world is generally stable: objects have positions, shapes, and colours that we seem to be conscious of instantly, and we can reach for them or move to where they are, without any apparent effort. We can recognise small differences between objects and we can categorise or name them appropriately. Some objects, like food, will be particularly significant for our survival, and we learn how to recognise them from all sorts of positions. Perceiving the location of objects and recognising them is achieved regardless of whether we, or the objects themselves, are stationary or moving. Clearly, there is some process within us that gives rise to our experience of visual perception and the behaviour that it guides. It is not surprising that throughout history students of the senses have found it fascinating to speculate what this might be.

What we call ‘perception’is an experience that normally results from stimulation of the senses by the environment. Perceptions can also arise through hallucination through dreaming, abnormal brain function or the use of drugs. However, our aim is to describe the characteristics of perception in its normal mode of operation, which has evolved to support interaction with the world around us. Perception can be examined in a variety of ways, such as describing our experiences verbally or measuring them by psychophysical experiment; it can also be related to the processes in the nervous system that accompany such experiences. Nerve impulses in the sensory nerves are relayed to the brain and behaviour, including speech, usually results from this sequence of events. Traditionally the only records of action of the senses were provided by behaviour but now a wide range of indirect physiological measures can augment them. In humans the range of behaviours is broad and includes describing the experiences initiated by sensory stimulation and the links it might have with previous stimulations. We refer to these as observations and we associate them with verbal descriptions. Observations provide the bedrock of perception and of other actions of the brain. Historical records of observation precede records of their verbal descriptions; that is, the products of art precede those of writing, by many thousands of years. Relatively little is known about the origins of visual art: examples of marks made on tools and cave walls have been dated to tens of thousands of years ago, but we do not know when such activities began. We assume that the cave-dwellers saw their world in much the same way as we see ours. It is not perception that has changed, but the world in which it takes place.

While we perceive the world around us with alacrity and ease, we have no direct knowledge of how this experience comes about. In fact, it can often be hard to believe that there is any mechanism involved in perception at all; for most people, most of the time, perceptions are simply ‘given’ as facts about the world that are obviously correct. Perception is not only a basic psychological process, but also a very remarkable one. Its success in providing us with accurate information about the characteristics of the world around us is an index of its power, because there are relatively few situations in which it is seriously in error. A perceptual process that gave rise to subjective experiences grossly different from physical reality would make survival virtually impossible. The function of perception is not to furnish us with subjective impressions of our surroundings and the significant objects in the environment. Rather, it is to provide an effective platform for action. Perceptions and actions take place in a three-dimensional environment and so it is necessary for spatial perception to share the three-dimensional coordinate system in which behaviour occurs.

Visual perception can be studied in many ways. We have chosen a particular approach to understanding our interaction with the world through perception, by emphasising the means by which we achieve correspondence between the external environment and its internal representation. This approach places special emphasis on the perception of space and spatial relationships (within which we include the perception of motion) as a platform for effective action. This is not to underestimate the significance of the perception of other characteristics of the world such as colour, but it is clear that spatial perception is a consistent requirement for action, whereas colour perception may or may not be. A crucial element of this approach to spatial vision is defining – rather than ignoring or assuming – the frame of reference within which sensory information is represented. When discussing any aspect of space or spatial perception it is necessary to define the frame of reference relative to which measures are made. This is commonplace for physicists, who are accustomed to defining the framework within which any measurement is expressed. Surprisingly, it is less common among perceptual psychologists, even though they are dealing with a system that has detectors in a moving eye, two eyes that can move with respect to one another, and eyes that can be displaced when the head or body moves. Information about where the image of an object falls on the retina could alter because the object moves, because the eyes move in the head, because the head moves with respect to the body or because the body moves in space. Any of these changes, alone or in combination, will result in the displacement of an image on the retina. Generally we perceive the world while there are complex patterns of eye, head and body motion, but our perception of objects is not disturbed. To understand how this might be achieved, we need to understand how frames of reference can be specified, and how they can apply to visual perception. It is this functional path that we will pursue and it is this aspect that we believe is neglected in many books on vision.

There are many excellent texts describing the eye, how it passes signals to the brain and how its many areas refine the physiological processing further. Moreover, these books describe, and often illustrate beautifully, phenomena that can be perceived. These tend to be presented in chapters with titles like brightness, colour, size, shape, depth, motion, and illusions, and they tend to give the impression that these are isolated aspects of visual perception. We have tried to avoid such categories by focusing on the functions of vision in a three- dimensional environment – determining where objects are, whether they are moving and what their identities are. We believe that our emphasis on function rather than phenomenology provides a stable basis for understanding visual perception and for appreciating the importance of historical achievements.

This task is in some ways rendered paradoxical by the fact that the medium we are using to present these ideas lacks some of the properties we wish to accentuate. It is tempting to assume that perception can be understood by studying pictures of solid objects rather than solid objects themselves, but we believe that this is one temptation that should be resolutely eschewed. The pages of a book are flat, as are the pictorial images on them. On the other hand, the book itself is an object in space and we shall refer to it as a means of illustrating issues about frames of reference in spatial perception. In this context, the term ‘book’ could refer to a printed text, an e-reader or any other surface on which text is presented.

A further potential impediment to our purpose is that we are using words to describe the processes of perception. This can encourage the impression that perception itself is like some internal dialogue about the world, and that perception is therefore idiosyncratic. Since people describe the world slightly differently we might think that they perceive it differently, too. We know that this can happen in the case of colour, but less so for the perception of objects. We have tried to discourage such thoughts by employing the assistance of a metaphorical guide dog for a blind person. Clearly such dogs can be trained to sustain the survival of blind people by avoiding obstacles, seeking selected sites, indicating impending danger and even crossing roads. The achievements of such guide dogs are remarkable. They can discriminate between objects in a similar way to us, but they cannot name them. No one would attribute their perceptual abilities to linguistic ability and yet they perform tasks that are equivalent to those of humans. Do we need language to understand human perception?

Another unusual aspect of the book is that there are relatively few references given in the text and, indeed, relatively few experiments that are described in detail. The reason for this is that we have sought to describe aspects of perception on which there is general consensus, like the perceptual constancies, but nonetheless aspects that still pose considerable theoretical challenges. It is in such contexts that we have found the notion of frames of references particularly useful.

This chapter provides an overview of central issues in the study of visual perception, many of which will be discussed in more detail in later chapters. It is important to understand the functions that any visual system must perform if there is to be coordinated, effective action, and the problems of devising explanations for how this comes about. If perception is to be explained, appropriate measurements of its characteristics must be obtained, and related to the information potentially available from the physical environment. Each of these issues contributes to the general framework of ideas that guides the investigation of vision.

Functions of visual perception

We all enjoy contemplating the experiences provided by our senses, and much of our language is associated with describing them. In human cultures considerable effort is devoted to enhancing perceptual experiences by decorating our bodies and our surroundings and by producing artefacts (like pictures) to stimulate the senses and to channel our contemplations. With so much emphasis on extending our perceptual experiences it is tempting to think of their function as enabling us to enjoy and describe them. In evolutionary terms the function of perception is much more mundane – it is to enable us to interact with the objects in the world surrounding us. More specifically, we use our perceptions to guide our behaviour. We use vision to determine the location of objects with respect to us, so that we can approach them, grasp them, cast them aside, or avoid them as appropriate for our survival. Some objects, like food, will be particularly significant for our sustenance, and we learn how to recognise them from all sorts of positions. Perceiving the location of objects and recognising them is achieved when we are still or moving, or if the objects themselves move. Accordingly, we need to be able to distinguish between static and moving objects whether we ourselves are static or moving.

Action and recognition

As we have emphasised above, vision does not occur in isolation, although we often describe it as though it does. It is integrated with the other senses and also with the movements we make. The actions we perform are based on vision and vision, in turn, is influenced by the actions we perform. This can be illustrated in terms of what you are doing while reading this text, whether from a traditional book, an e-reader tablet or a computer monitor. The surface you are reading from is likely to be in a fixed position and appear stationary. This is the case even though the reading surface and the text on it are moving over your eyes as a consequence of eye movements. We now know quite a lot about the ways the eyes move when reading text and also when viewing scenes: they tend to remain relatively still for a few tenths of a second (called fixation) and then flick rapidly (or saccade) to a new location, with this sequence repeated about three times every second. Despite these jerky movements of the text over your retinas the experience is of a stable surface with stable text. For this to happen there must be an intricate internal integration between the pattern of stimulation on the retinas and the signals for contracting the eye muscles. The situation becomes even more complex when we take gross movements of the body into account; the head can move with respect to the trunk and the whole body can locomote through space.

Thus, at this basic level of perception, action is a fundamental determinant of vision. All observer movements may displace the patterns of light projected on the retinas, irrespective of the location and motion of the objects in the environment that give rise to them. This issue will be discussed in greater detail in later chapters, where we will identify the frames of reference with respect to which object characteristics are successively represented. For example, the retina itself sets the frame of reference for any information deriving from its light detectors and is subject to perturbations from eye and body movements. To be useful for guiding action by a human or a guide dog, this potentially ambiguous information must ultimately be represented in the same frame of reference as that within which action takes place; that is, the three-dimensional environment in which we live.

The complexities of this seemingly straightforward process can be illustrated with an example of a task we perform successfully every day – crossing the road. We will also enlist the cooperation of our metaphorical guide dog in order to contrast its performance and perception with ours. When we wish to cross a road we walk up to the kerbside, look in both directions and cross if it looks safe to do so. But what does this apparently simple sequence entail, and how can it be explained? Crossing the road itself requires coordinated activities in many muscle groups that control balance and locomotion. However, before the walking is initiated a whole series of decisions has to be made on the basis of perceptual information. How far away is the kerb and in what direction is it? How wide is the road and how much time will be required to cross it? Is there a vehicle approaching? What size is it? How far away is it? How fast is it travelling? How long will it take to reach here? What type of vehicle is it? Each of these questions is directed to some aspect of the three-dimensional environment and our relation to it. The consequences of making wrong decisions for any one of these questions can be life-threatening.

How far away is the kerb and in what direction is it? To answer these questions it is necessary to determine where the pavement stops and where the road starts. This could be done on the basis of some markings (e.g., white lines) which we have learned to associate with roads, or on a difference in the perceived level of the two surfaces. Some theorists consider that we can derive information regarding the orientations of surfaces in the environment without recourse to knowledge about the nature and purpose of objects. That is, the information available in the pattern of stimulation is sufficient to specify the layout of space with respect to the viewpoint of the observer, and assumptions about the nature of the world are not necessary. On the other hand, perception can be likened to thinking and other high level cognitive processes like reasoning and problem-solving. Such cognitive theorists argue that we use our knowledge of roads and their features to recognise the kerbside, and that we make use of a ‘knowledge-base’ to find and identify objects in the visual scene. Thus what we see is a logical, rational interpretation of sensory information on the basis of stored knowledge, in terms of possible real events. It is certainly not necessary to have verbal labels like ‘pavement’ and ‘road’ to perceive the positions and sizes of the actual objects, as guide dogs for blind people carry out such discriminations with remarkable skill. In the discussion that follows, we will contrast the capabilities of a human and a guide dog in crossing a road, as this helps to indicate how far the perceptual processes involved depend on verbal knowledge.

How wide is the road and how much time will be required to cross it? When the edges of the road have been isolated, by whatever means, some estimate of the distances of both from the observer is required. Distance information can be with respect to the observer or between objects in the visual scene. For example, the near kerb could be perceived as two metres away and the far kerb twenty metres distant, then the width of the road could be determined from the difference between the two estimates. Alternatively, the distance between the two kerbs could be perceived in some relative way, such as ‘three cars wide’. The time required to cross the road could then be computed as long as there is some value of walking speed available.

Is there a vehicle approaching? Before this question can be addressed it is necessary to define what a vehicle is. An abstract definition in terms of, say, a metal body with windows supported by four wheels, would not suffice visually, as we rarely see all the wheels on a vehicle at the same time; often we do not see any wheels at all. Moreover, some vehicles have two wheels, oth...

Table of contents

• Front Cover
• Visual Perception
• Title
• Copyright
• Dedication
• Contents
• Preface to the first edition
• Preface to the second edition
• Preface to the third edition
• 1 Understanding visual perception
• 2 Fundamentals of vision
• 3 Theoretical approaches to vision
• 4 Optics and the eye
• 5 The visual brain
• 6 Spatial location
• 7 Spatial motion
• 8 Spatial representation
• 9 Summary and conclusions
• References
• Name index
• Subject index