Return to MODULE PAGE
Introduction to the Methods Used to Study Perception
Rob Stufflebeam: Author
Why study methods of study?
Broadly construed, the aim of science is to understand the nature of "the world." We will use the term 'the world' to refer not just to our planet, but to any object, place, idea, event, or individual found in the universe. As there are many aspects of the world that we want to understand, to explain, and to predict, Cognitive science, or more broadly, the cognitive and learning sciences, include the disciplines of computer science, psychology, philosophy, neuroscience, linguistics, artificial intelligence, and robotics (among others). What binds researchers from these diverse fields is the aim to understand how intelligent systems work. An intelligent system is something that processes internal information in order to do something purposeful. Although humans are intelligent systems, other "natural" kinds of intelligent systems are also studied. So too are robots and other "artificial" intelligent systems.
While it is true that perception is one of the many aspects of intelligent systems studied by cognitive scientists, it is also true that there are many aspects to perception. Thus, cognitive scientific research into the nature of perception is driven by many kinds of questions.For instance, many cognitive scientists labor to resolve issues about human perception. Some of the questions motivating their research include:
-
How do our sensory systems work?
-
Why are we susceptible to illusions?
-
Which structures in the brain perform which perceptual functions?
-
Do we owe our ability to recognize faces, places, objects, etc. to special computer programs running on our brains?
-
How do perceptual states acquire meaning?
But not all cognitive scientists labor to resolve questions about human perception. Indeed! The goal of many cognitive scientists is to build mobile robots -- artificial perceivers capable of exploring their environment. As such, a great many cognitive scientists labor to settle issues about machine perception. Some of the questions motivating their research are:
-
Is it possible to build a creature that perceives as we do?
-
How can a mobile robot be built that recognizes danger, then carries out the intelligent action of moving out of harms way?
-
Must robots use computers to do their perceptual processing, or are there other options?
-
How do perceptual states in a machine acquire meaning?
-
Is it possible for an artificial "perceiving" machine to have a mind? If so, how can we tell whether one does?
These lists of questions are not intended to be exhaustive, but to illustrate the variety of questions that lead cognitive scientists to study perception. And because research cannot occur in the absence of a research method, it is a fact of cognitive scientific life that cognitive scientists use a host of methods to study perception (and other phenomena associated with intelligent systems). Some methods are used by researchers in only one cognitive scientific field (e.g., neuroscience, artificial intelligence, etc.) Others are used by researchers from several disciplines. While it is important for you to understand what has been discovered about perception through cognitive scientific research, it is equally important for you to understand something about the research methods through which discoveries are made.
From the outset, it is important to recognize that no single research method can answer all our questions about the nature of perception. After all, every method has limitations. For example, suppose that a team of NASA roboticists succeeded in building a robot that can not only perceive signs of life on some distant planet, it can also "act" to investigate what it perceives. Despite the team's contribution toward resolving a host of questions concerning artificial perception, the observations and data collected through their research will not necessarily be of any help in answering innumerable questions about human perception (e.g., "Why do some people loose their ability to recognize faces?").
A less obvious reason why no one method can answer all our questions is that some foundational questions about the nature of perception are not empirical questions (questions about the way the world is that can be settled by observation). Rather, they are philosophical questions; that is, they are fundamental, open questions about the meaning, truth, or logical relations among our ideas, concepts, theories, etc. Fundamental questions of this type cannot be resolved through empirical research alone. For instance, consider the following question:
Is it possible (with present technology) to build a machine that can think?
This is not merely an engineering issue concerning the limits of existing technology. It is also a question about the concept of "thinking". Two scientists might agree about all issues surrounding the present state of technology, but disagree about whether a sophisticated robot built by MIT is actually doing something that can properly be described as "thinking". Consequently, this question cannot be settled by merely observing the way the world is. Instead, it requires exploring some deep and controversial questions like: What exactly does it mean to say that something is a "machine"? Is genuine "thought" possible only for biological creatures? How can we tell if a thing has thoughts?
While researchers in every scientific field are forced to address philosophical questions, cognitive scientists deal with such questions all the time. This is so not only because the subject of cognitive scientific research is intelligent systems, but because cognitive science is a "young" field and there are several competing theoretical camps vying for position as the proper framework within which to explain how intelligent systems work. Each of these frameworks brings along its own set of assumptions or presuppositions. These assumptions have a significant influence not only on how research will be conducted, but also on how the results of that research will be interpreted.
Clearly, assumptions matter. Since a scientist will interpret what she observes in relation to her assumptions, another researcher using different assumptions may see the world differently. And as there are very few theoretical assumptions that are accepted in all quarters of cognitive science, and as each research method is based upon its own set of assumptions, there are many competing claims about the nature of perception (and other aspects of intelligent systems) coming from cognitive scientists. Where there are competing claims, controversies are sure to follow.
But let's not get ahead of ourselves. Before you can explore the controversies, you need to understand the methods. Let us then turn to an overview of the specific research methods, techniques, and assumptions through which cognitive scientists explore questions about the nature of perception.
Perception research methods
There are many questions about perception that engender cognitive scientific research. No single discipline can answer all of the questions about perception. And since no research can be done in the absence of a research method, instrument, or technique, it should come as no surprise to hear that cognitive scientists use a host of methods to study perception. Just as no single discipline can answer all the questions, no single research method can do so either.
The question at hand is: What methods do cognitive scientists use to study perception? As this is primarily a gentle introduction to the specific research methods you will encounter in the curriculum that follows, do not feel slighted should you come away from this section not knowing everything you need or want to know about a specific method. Rest assured, much more will be said when the circumstances warrant doing so.
Arguments
An argument is what we offer through language as a means of proving, explaining, persuading, convincing, or otherwise showing that the truth of something follows from the truth of something else. Every argument consists of two parts. One is the claim, a statement asserting that such-and-such is the case. The other is the evidence, the statement(s) offered to show that the claim is true. Because asserting claims and defending them with evidence occurs throughout cognitive science, arguments are advanced everywhere in cognitive science. Sometimes arguments are advanced in support of answers to empirical questions (questions about the way the world was, is, or will be). Sometimes arguments are advanced in support of answers to philosophical questions (fundamental or open questions about the meaning, truth, or logical relations among our ideas, concepts, theories, etc.). And sometimes they try to answer both types of questions simultaneously -- a very tricky affair. Regardless, progress in cognitive science requires both empirical arguments and philosophical ones. In the curriculum to follow, you will be introduced to a wide range of both kinds of arguments.
Introspection
How do your perceptions compare with those of other people? What does a red apple look like? Do we see the same color of red? What does chicken taste like? What does giving birth feel like? If we both place our hands on a hot stove, will our pains be similar? How do you feel when you recognize the sound of gunfire (tornado sirens, or loud music at 3:00 AM)?
These are but a few of the host a questions about the nature of subjective perceptual experience for which introspection is the method of choice. All scientific methods are "inspections" of a sort -- ways of observing some subject in the world. Introspection is the method whereby you "look" within yourself to report what is going on in your mind, how you feel, or what it is like to be you.
For instance, suppose you and a friend were to visit the top of the Empire State Building. As your friend approaches the edge, you notice that he becomes flushed, anxious, and nervous. You infer on the basis of this observation that he has a fear of heights. While there is a sense in which your "outward" inspection answers the question how your friend feels, there is another sense in which it does not. In this other sense, your friend needs to report the quality of his own experience. To do that, he must look within himself and report how he feels: "I feel scared." "I'm afraid of heights." "My pulse is racing and I feel lightheaded." That is introspection. Researchers have been using it (or evoking it in their subjects) to gather evidence about minds for as long as humans have been interested in the how minds work.
Of course, there is no way for a researcher to tell for sure that a subject is being truthful or accurate in her reports of subjective experience. Further, since a researcher can never share another person's experiences, it is not possible to tell if a word used to describe an experience (e.g., 'red' or 'lightheaded') is actually being used to refer to the same kind of experience in the reports of two different people. For these reasons, and others, the reliability of introspection (and data collected through it) is sometimes called into question. Nevertheless, introspection is a common method used to study perception (and other aspects of cognition) even today. And we shall use it ourselves in the curriculum to follow.
Experimental psychology
Psychology is the science of the mind. While there are many subfields of psychology, three of them -- cognitive psychology, animal psychology, and neuropsychology -- have contributed a great deal to the study of perception through experimentation. Hence, each of these subfields qualifies as experimental psychology too. As the name implies, experimental psychology is the type of psychology whereby a researcher forms a hypothesis, tests the hypothesis by requiring that a subject perform a relevant task, observes the subject's behavior, then evaluates the hypothesis in relation to the positive or negative data collected. Unsurprisingly, there are a number of methods, instruments, and techniques used by experimental psychologists to test hypotheses about perception.
Some of these methods require merely presenting the subject with a sensory stimulus and having the subject report what she sees. Experiments designed to "evoke reports" are used to study many types of perceptual phenomena:
- "pop-out" effects
- visual illusions
- naming (object recognition)
- matching abilities
- discrimination abilities
- motion detection
You will see such experiments at several places in the curriculum.
Other methods require the use of instruments to record a subject's behavior as he performs a perceptual task. This is the case in studies of eye-movements and most other neuropsychological research into the brain. But because neuroscience is the science of the brain, let us turn to an overview of neuroscientific research methods.
Neuroscientific methods
Just as there are subfields of psychology, there are subfields of neuroscience. And cognitive neuroscience, computational neuroscience, neurology, and neurobiology are among the main subfields that have contributed to our understanding of perception. This should hardly be surprising. After all, the intelligent system we want to understand most is us -- humans. Human perception occurs as a result of information processing in several kinds of systems: sensory systems (visual, auditory, somatosensory, olfactory, and gustatory), attentional systems, memory systems (for both storage and retrieval), motor systems, etc. Not only do neuroscientists study all of these systems, they do so at every structural level of organization:
Because no one research method can answer every question at every level of processing for every system neuroscientists use a variety of methods, instruments, and techniques to study perception. If we ignore the role of the computer as a research tool in modeling how the brain works (a topic we shall deal with below), neuroscientific methods fall into two classes, invasive ones and noninvasive ones.
Invasive methods are research techniques that require the introduction of an "instrument" into a subject's brain -- a scalpel, a probe, an electrode, a finger, whatever. There are several methods of this sort. Surgery is the oldest. And an enormous amount of knowledge about the functional organization of brains has been gained through modern neurosurgery upon conscious patients. Lesion studies are another classic invasive neuroscientific method. A lesion is a "damaged" area of the brain resulting from trauma ("insult") or disease. Through observing the deficit that a lesion causes, then observing the brain (after autopsy) to localize the lesion, lesion studies have contributed a great deal to our present understanding of structure-function relations in the brain. Other invasive techniques require implanting electrodes to record the electrical activity of a neuron or population of neurons. Methods requiring this procedure include stimulation studies, electroencephalogram, evoked response potentials, and single cell recordings.
Noninvasive methods are research techniques that do NOT require the introduction of an instrument into a subject's brain. All neuroimaging methods are noninvasive ones. Of the several neuroimaging methods used to study perception, some are used to identify brain structures: conventional radiographs (X-rays), computerized tomography (CT), and magnetic resonance imaging (MRI). Others are used to identify functional areas of the brain: functional autoradiography, positron emission tomography (PET), and functional magnetic resonance imaging (fMRI). Should this alphabet soup be confusing, do not fret. In the neuroscience portion of the curriculum we shall explore these methods and their underlying assumptions in greater detail.
Computer modeling
Many people consider 'cognitive science' to be equivalent to 'computational science'. The main reason for this is the nearly universal belief among cognitive scientists that cognition requires computation. Moreover, almost everyone within the cognitive science community considers the following belief to be the fundamental assumption upon which cognitive science itself rests: Minds are to brains what programs are to computers. Some cognitive scientists treat this view only as a metaphor, which is why it is called the computer metaphor. Many others take it to be true literally. As a result, computers play a significant role in cognitive science research. This occurs in two ways. First, computers themselves are the subject of intense (sometimes controversial) research into the nature of computation. Second, in every field of cognitive science, computers are used as a research tool to model how information processing occurs in complex systems. Our focus here is on the latter.
So, what is a computer model? Well, a computer model is a computer generated simulation of something. Computers can be used to simulate many kinds of processes -- tidal waves, weather patterns, economies, traffic jams, . . ., you name it. But while it is correct to say that what you see on the computer is an imitation of some natural, social, economic, or other process, it would NOT be correct to say that what you see is aduplicationof that natural, social, economic, or other process. To simulate is to imitate, not to duplicate. Such is the reason why you would not run for high ground upon seeing a simulation of a tidal wave.
With duplication, things are different. After all, a simulation of a tidal wave on your computer does not duplicate a tidal wave. To duplicate a wave is to create a "real" wave -- even if it is only a "mini" or scale one created under controlled conditions. While a "mini" tidal wave may not make you run for high ground either, it can at least get you wet. Similarly, to clone a frog is to duplicate it -- to create a "real" frog. To duplicate something is to reproduce its essential properties. Whenever we do this, an emulation of something has been created, not a simulation. To emulate is to duplicate, not to imitate.
One of the major controversies in cognitive science today is whether any existing computer models of human cognition go beyond mere simulation to something closer to emulation. Regardless of the correct answer is to this question, computer models, whether as simulations or emulations, are valuable research tools.
But what about perception (cognition, intelligence, or some other aspect of intelligent systems under study by cognitive scientists)? Does a computer model of a perceptual process (say vision) qualify as a simulation or an emulation? For several reasons, it is not obvious what the "correct" answer to this question is. First, if the computer metaphor is literally true, then a computer program that produces the relevant perceptual output would be an emulation of the perceptual process. Second, if the computer metaphor is merely a useful research strategy, then a computer program that produces the relevant perceptual output would only simulate the perceptual process, not emulate it. Third, to complicate matters further, whether the computer metaphor is true depends on the answer to these and other questions about the nature of computation:
-
What does it mean to be a computer?
-
Is your brain literally a computer or something fundamentally different?
-
If your brain is a computer, is it a digital computer (like a Mac or a PC), or is it some other (nonclassical) type of computer?
Now is not the time to explore these issues. We shall deal with them in due course. They have been raised here to make you sensitive to some of the controversies surrounding the widespread use of computers to model and to explain perception.
Robotics
Robotics is the branch of artificial intelligence whose aim is to build machines (or "artificial creatures") capable of intelligent actions. Building a mobile robot capable of perception is not so simple to do. In the pages that follow, you will explore some of the complex methods that are presently being employed to give a machine the capacity to recognize individual objects, to distinguish shadows from objects, to determine the size and distance of objects, etc. This is a very difficult thing to do even if your goal is merely to simulate human vision (i.e., to perform roughly the same tasks) rather than to emulate it (i.e., to perform the same tasks in the same way as a biological creature).
Robotics is important for many reasons. It plays an important role in the testing of certain theories. If one can construct a robot that successfully performs a particular cognitive function, using a particular type of computer program, that is the best (some would say the only) way to test certain claims about the practical application of computational theories. When someone says that a particular theory will "work", often the only appropriate defense is to build a machine that will "prove" that it works.
In the future, as we become more adept at building machines that do not merely simlulate but that emulate our own cognitive functions, robots will play an increasingly important role in testing theories about human cognition. In this curriciulum we will explore different strategies in the construction of robots as we reflect on the nature of our own cognitive abilities.
Enough about methods, let's go explore the mysteries of the human mind!!