Essential Behaviour Analysis

Julian Leslie

• Publisher: Routledge
• Year: 2015
• ISBN:
  9781134665143
• Edition: 1
• Topic: Psicología
• Subtopic: Historia y teoría en psicología

Since the so-called 'cognitive revolution' in psychology in the 1960s, it has often been said that 'behaviourism is dead'. This book demonstrates why this is not the case and how the behavioural approach has continued to flourish. Leslie begins by summarising the behavioural approach to psychology and shows how it differs from other contemporary and cognitive approaches. The basic principles of the discipline are outlined and linked to major areas of interest and importance, such as behavioural neuroscience, resolution of human behavioural problems, and human language and cognition. Behaviour analysis is thus shown to contribute to our developing understanding of the relationship between brain systems and psychological problems, to provide an effective and scientifically based approach to human behavioural problems and to deal with topics central to modern psychology.

1 Opening Statement

∎	The message
∎	Is this a credible theory of human psychology?
∎	Is that all there is to it?
∎	Summary

A man who had only ever read two books was asked which he liked best. 'Well,' he said, 'the story was better in Robinson Crusoe, but the dictionary explained all the words as it went along.'

Textbook writers always face the difficulty of interesting the reader in the story, or general message that they wish to convey, while using most of the space, and the reader's time, providing detailed explanations. In this book, the strategy used is to set out the main points of the story, or message, first and then go through more details in the remainder of the book.

The most important part of the story is a simple message, or explanatory principle, which will be stated and developed in this brief opening chapter. The rest of the book will provide the following: some historical background (because young sciences cannot ignore their history); evidence of the capacity of the explanatory principle of behaviour analysis to predict behaviour in experimental procedures; demonstrations of how phenomena of human psychology can be interpreted in its terms; examples of issues to which it can be applied; and discussion of how this message contrasts with those given, on apparently similar issues, by other contemporary approaches to psychology. A certain amount of space will be devoted to making clear what is not part of the message, and which implications it does not have. It is necessary to deal with these issues because, although the explanatory principle of behaviour analysis is simple, it has sometimes proved very hard for people, including some of those who are expert in other approaches to psychology, to grasp its key features.

1.1 The Message

The message is that human psychology is a biological science in which the key process is one of selection, as is also the case in evolutionary biology. What is selected in the realm of psychology, however, are behavioural characteristics of the individual person (rather than characteristics within a population of a species). In a lifelong process, there is continual interaction between the current behavioural characteristics (usually called simply behaviours) of the individual and the prevailing environment of that person. At every stage, those behaviours that have particular consequences are selected and tend to become more frequent, or persistent, in the current environment, at the expense of alternative behaviours that become less frequent. This process of selection by consequences is usually called operant conditioning.

Because psychology is a biological science, we would expect to find a high degree of similarity between humans and other species in the ways in which operant conditioning works, and this turns out to be the case. While there are also important differences between humans and other species, the most striking finding has been the similarity of the effects of operant conditioning across a huge range of species of animals. This suggests that operant conditioning is an important adaptive mechanism that has been favoured by natural selection.

The behaviours that are affected by operant conditioning are functionally defined. This means that those actions of an individual that count as examples of one type (or class) of behaviour are those that have the same function for that individual. Let us consider two examples of classes of behaviour. In simple terms, 'shutting the door' is a class of behaviour. Because this class is functionally defined, different members of the class will not look the same as each other; we say that they have different topographies (this term is explained below). After all, I can shut a door with many different movements on different occasions, and sometimes different movements are required because of the shape and position of the door, or I can shut it by asking someone standing next to it to do that for me. Despite this variation in form of behaviour-class members, it is also true that the class members could be the same for many different individuals, if we were shutting the same doors. For a second class of behaviour, 'meeting a friend', the situation is more complicated again. One person's friends are different from those of another and thus the behaviour class will have different members for each individual. Given this complexity, even with apparently simple examples, how can we be justified in treating functionally defined classes of behaviour as our basic unit of analysis? This is a very important question, and later on I will show how this claim is supported by experimental findings. At this stage, where we are engaged in preliminary discussion and interpretation of behavioural and psychological phenomena, we can only give a verbal argument in its favour.

The key verbal argument is that classes of behaviour are defined by the consequences that they produce in the environment of the individual. This principle is embedded in a lot of the language we use to describe human behaviour and actions. Reverting to the two examples given above, door-shutting behaviour has occurred if the door is now clearly closed (half-hearted shoving, which leaves it ajar, is not an instance of that class of behaviour), and we have only succeeded in meeting a friend (no matter how long we waited at the railway station) if we are now in the presence of one of our favourite people. In both examples, then, we examine possible instances of the behaviour class to see whether they produce a particular consequence, and only include those examples which do produce that consequence or effect. Thus, we are clearly using functionally defined classes of behaviour. These examples are not unusual: we talk mostly of action-outcomes rather than patterns of bodily movements in describing behaviour. We are thus wedded to describing behaviour in terms of its consequences in the environment, and generally do this at the expense of describing bodily movements. Descriptions of bodily movements, which we will refer to as topographically defined classes of behaviour, are resorted to for particular reasons. These might include accounts of physical exercises (e.g., 'Lie flat on your back, and then raise both knees to your chest'), or for literary effect (e.g., 'He swung round, and his outstretched fist met his assailant's face'), but they are rare in everyday accounts of human interaction. As we talk to each other in our routine discourse, functionally defined behavioural classes are usually employed.

Our account of psychology is necessarily interactive; an account of the person, from a psychological perspective, can never be given without also providing an account of their environment. Along with other approaches to psychology, behaviour analysis uses the term 'environment' to mean those aspects of the physical world that are perceived by the individual and may have an influence on their behaviour (this is a different and more restricted meaning than that used, for example, in the physical sciences). Whether the members of a class of behaviour are the same or different for different individuals, class members can never be defined except with reference to the environment. We must specify stimulus classes as part of the process of specifying the behaviour class.

As with classes of behaviour, stimulus classes are also functionally defined. To take our two examples, what are doors and who are my friends? Well, doors are those objects that through being opened provide access to enclosed spaces, and my friends are those people with whom I have a previous history of positive social interactions or, in less formal language, those with whom I have had a good time in the past. These provisional definitions are functional ones, because they refer to the functions that objects or other people, respectively, have in relation to the behaviour class that is being defined. Note also that there are two stimulus classes that generally participate in the definition of each functionally defined class of behaviour. The first of these is the situation, or context, in which the behaviour occurs; the second of these is the outcome, or consequence, that the behaviour has. In the first example, the context is 'being at a door' and the important consequence, as previously noted, is the successful closure of the door. In the second example, of meeting a friend, possible contexts might be highly diverse but it is clear that they will be functionally rather than topographically defined. That is, there is nothing about the physical location or physical features of a place (these would be topographic features) which make it likely that friends may be met there. Instead, meeting places for friends are either socially and culturally defined (they may meet in places where such meetings are common, perhaps a public space such as a railway station or a restaurant), or they are defined by the history of interaction of those two people (they may meet in places with which they are both familiar). In each case, the context is defined by the function it has for the participants.

The classes of stimuli that act as effective consequences that maintain or change behaviour are also functionally defined. In the case of the door-closing example, the possible functions are familiar but not fixed. That is, we know why we want doors to be closed, but the reason (function) is not always the same. Closing the door may shut out unwanted noise or cold air, it may keep in heat, or it may make the enclosed space more useful for conversation or other social interaction. When we meet a friend, this may have consequences with many different functions. We may receive friendly smiles and gestures or verbal approval, anxiety that arises from loneliness may be reduced, or many subtle but personally important other events may ensue. It is important to recognize that the consequences that are effective in such cases are not only functionally defined, but are also specific to that particular interaction between people. What 'works' in that encounter depends on the history of interaction between the people involved. We can say that they have learnt to treat each other in particular ways (although they may not be entirely conscious of how they behave towards each other) and these patterns of behaviour now contribute substantially to maintaining the relationship between them. An important point to note here is that the behaviour of each of them supplies the context and the consequences for the behaviour of the other. This is an important fact about human psychology; many of the important stimuli or environmental events for one person arise from the behaviour of other people. What behaviour analysis adds to this insight is the notion that the particular histories of interactions between people modify the stimulus functions they each have for each other. In other words, what people do for each other, and the reasons why they spend time with each other, depend a great deal on what has happened when they have been together previously.

From this perspective, the psychology of an individual consists primarily of an account of those functionally defined behavioural characteristics that occur in the environments typically encountered by that individual. A person, if you like, is primarily to be understood as 'what he or she does' and that account of their behaviour cannot, as we have already seen, be described without also describing the location or occasion of those behaviours and the important consequences of those behaviours. There are thus three components to our basic unit of analysis – the context, the behaviour and the consequence – and each of these is functionally defined with respect to the other two. This triumvirate is often called the three-term relationship of operant conditioning.

From the perspective of behaviour analysis, the key process of psychological change is operant conditioning, whereby those behaviours that are functionally effective for the individual become more frequent (in the corresponding environment) while other behaviours decline in frequency. To return to our two examples, we would speculate (if we had no direct evidence) that we have learnt to shut doors in the following sense. We have varied the action used on many occasions when shutting familiar, and sometimes less familiar, doors, and those actions that have had the effect (or consequence) of actually shutting the doors in question have become more frequent when similar situations have subsequently been encountered. At the same time, failing efforts – actions that were not effective in shutting doors – have become less frequent and are now seldom seen. Thus, our history of interaction with the environment (in this case, an aspect of the physical environment) has selected a functional class of behaviour. In the case of meeting friends, there may have been a history of the following kind. We have, over a period of time, made various sorts of arrangements to meet friends; some of these have been followed by successful encounters, while this has not happened in other cases. When meetings have occurred, we have interacted with the friends involved in a variety of ways. Across this more complex set of contexts, behaviours and consequences, those choices that have led in the past to successful social encounters are likely to determine how we behave currently. That is, we will propose meeting places that have been successful in the past, and when the meetings occur we will behave in the ways that have led to desirable consequences on previous occasions. Again, it is our history of interaction with the environment (in this case, aspects of the social environment) that has selected a functional class of behaviour.

It is important to note that behaviour (either shutting a door or meeting a friend, in the two examples used) is not said to be caused by either the individual or the environment. Rather it is the history of interaction between the behavioural repertoire of the individual (that is, the whole range of behaviours shown by the person) and the environment that selects, and in a sense causes, the behaviour. Another important point is that as the history of interaction changes, so the behaviour will change. If the next few doors encountered are of a different design than those encountered earlier, then the class of door-closing behaviours will change. Similarly, if well-tried strategies for meeting friends become ineffective, then that behaviour class will also change. That is likely to happen, for example, if the person concerned goes to live in a different culture.

1.2 Is This a Credible Theory of Human Psychology?

Because the behaviour analysis approach is rooted in biological and experimental science, our first steps in providing empirical support for it, beginning in Chapter 3, will be to show how experimental techniques have been developed with non-human animal species, and to present data that show how operant conditioning effectively changes behaviour in those species. Only findings of empirical research – as opposed to the interpretations of the hypothetical examples given above – should be used as evidence in support of a theoretical position in science. As will be explained in Chapter 2, early researchers in psychology were keen to see whether they could devise ways of studying learning in animals under controlled laboratory conditions that might provide pointers to important processes in human psychology. In this aspiration they were following explicit suggestions from Charles Darwin, published in the 1870s.

Darwin's theory of evolution through natural selection revolutionized the science of biology in the nineteenth century, but it also provided a new impetus to the faltering start that had been made in developing a scientific approach to psychology. Darwin asserted that there was continuity between all species, in that all current species were historically linked to common ancestors, and he suggested that similar psychological processes might be thus be found across animal species.

It was in the development of operant conditioning techniques in the mid-twentieth century that this prediction became a scientific and practical reality. These experimental methods have revealed that the interaction between the behavioural repertoire and the environment proceeds in a similar fashion, and thus throws up similar behavioural phenomena, across a remarkably wide range of animal species, including human beings.

Having established that operant conditioning has a role to play in the analysis of behaviour of many species, researchers in behaviour analysis have also addressed issues that are germane to humane psychology per se. These include:

• demonstrations that behavioural phenomena seen under the more highly controlled experimental conditions possible with other species can also be replicated in human psychology
• use of experimental techniques, usually with non-human animal species, to investigate behavioural neuroscience, and thus begin to understand the links between behavioural processes and brain processes
• development of intervention techniques, based on behavioural principles established through laboratory experiments, to alleviate human and social problems
• analysis of those aspects of human psychology, namely language and cognition, that seem not to be common to other species.

All these issues will be reviewed in later chapters of this book, to give the reader a flavour of the behavio...