Applied Behavior Analysis
Applied behavior analysis is the application of behavioral science to address socially important problems. It is one element of the larger discipline of behavior analysis, which consists of the experimental analysis of behavior, radical behaviorism, and applied behavior analysis. The field of applied behavior analysis emerged in the 1950s and 1960s from early theory (Skinner, 1938, 1953; Terrace, 1966) and animal research identifying basic principles of behavior (Honig, 1966). Behavioral principles, such as positive reinforcement, generalization, and extinction, began to be applied to socially important human behavior with startling and encouraging results. Sidney Bijou and Donald Baer (1961), for example, demonstrated that children with intellectual disabilities and very limited communication skills could learn to interact effectively when provided with instruction based on behavioral principles. Children with self-injury and severe aggression learned alternative skills and reduced their dangerous behavior. These early examples of behavior change emphasized that behavior analysis was both a scientific approach for studying behavior and a technology that could be harnessed for positive social change. This distinction between applied behavior analysis as a field of science, and applied behavior analysis as a technology for intervention, is important. The core features of applied behavior analysis exist in both, but are emphasized differently when the goal is to advance the science of human behavior rather than address a personal intervention need.
Applied behavior analysis became a distinct field of study in 1968 with the launching of the Journal of Applied Behavior Analysis and the classic, inaugural article by Donald Baer, Montrose Wolf, and Todd Risley (1968). These authors used this article to describe eight defining dimensions of applied behavior analysis: (a) applied (b) behavioral, (c) analytic, (d) technological, (e) conceptually systematic, (f) effective, (g) generalizable, and (h) durable. These dimensions warrant review for anyone interested in understanding and contributing to the field.
Applied behavior analysis is a science guided by values. While behavior analysis as a field focuses on variables affecting behavior (and that can be any behavior by any organism) applied behavior analysis is the study of variables that affect socially valued human behavior. The very first feature of an applied behavior analysis research study is description of the social issue or concern under study. What is the behavior of study, and what criterion would make that behavior appropriate and acceptable for that individual in his context? Applied behavior analysis is a science/technology with an overt goal of improving society, of assisting people to achieve identified goals, and of applying the science of human behavior toward those ends.
It is worth noting that the initial dimension selected by Baer, Wolf, and Risley to define applied behavior analysis was that the analysis must focus on problems in which society shows a clear interest and concern. The applied element of applied behavior analysis is defined not in technology, procedure, or science, but in the social value of the issues under study. As such, research and practice in applied behavior analysis typically will not only focus on highly valued behavior (e.g., reading, speaking, social interaction, play) but will examine that behavior in the actual context where it typically occurs. One of the underlying messages from behavior analysis is the central role that contextual variables (prompts, social opportunities, consequences) have on behavior. Thus an applied understanding of human behavior requires study of that behavior in natural contexts. For example, an applied study of reinforcement would be more likely to occur with a valued behavior, such as reading, in a natural context, such as a school or home, rather than with a behavior such as lever pressing studied in a laboratory setting. Some applied behavior analysis research may occur in atypical contexts such as clinics when the goal is to isolate variables that are causing the behavior, but even then the ultimate goal is to apply what is learned to more natural settings. For example, a child who exhibits self-injury may receive assessment in a clinic in which conditions can be managed with precision and safety to determine what triggers and maintains the behavior. The goal of the resulting intervention, however, would be to achieve behavior change in the child's typical home and school. The applied dimension of applied behavior analysis emphasizes valued behavior in typical contexts.
Applied behavior analysis focuses on observable human behavior. There are increasingly sophisticated definitions of behavior (Johnston & Pennypacker, 1980), but the basic message is that the focus of an applied behavior analysis is human behavior that can be observed and counted. The emphasis on observable behavior is part of the precision that makes applied behavior analysis a science. Importantly, observed does not necessarily imply a focus only on behavior that is observable by others— simply that the behavior be observable at least by the person exhibiting that behavior (Skinner, 1953). Thus, thinking is considered a behavior but inferred, internal states such as anger would not be a focus for applied behavior analysis, but hitting, kicking, screaming, throwing would be observable and countable behaviors suitable for study. Because applied behavior analysis focuses on changing behavior via altering events around the behavior, the focus most often is on behaviors that are observable to others as well as to the person exhibiting the behavior.
The goal in applied behavior analysis is the direct study of behavioral phenomena. If eating were the focus of the analysis, then the study would likely involve observing and counting bites or calories ingested. The study would not, for example, focus on verbal descriptions of what was eaten. Verbal descriptions would be an indirect measure of what was eaten (though a direct measure of talking about eating). Similarly, if a study were examining a child's screaming, the analysis would not focus on fear or anxiety (which are not observable or countable) but on the behavior of screaming, and the conditions when screaming was most and least likely.
The emphasis on direct study of observable behavior reflects concern about the potential for confusion and miss-calculation that can occur when inferred emotions, intentions, and motivations are used as the heart of an analysis. A hallmark of applied behavior analysis is excruciating precision in the definition of, and direct observation of, the behavior under study.
Both as an approach to studying behavior and as a technology for behavior change, applied behavior analysis is analytic. As a science, applied behavior analysis incorporates a systematic process characterized by (a) valid, reliable measurement of behavior, (b) operational description of intervention procedures, (c) utilization of research designs that allow demonstration of experimental control, and (d) replication of findings. In most cases, research examples of applied behavior analysis involve a process in which individuals are observed over time and the researcher manipulates a specific feature of the context or setting to determine if this feature (e.g. the consequence following a behavior) will affect the frequency, duration or form of the behavior under study. The presentation of a consequence, such as praise, for example, would be manipulated within a formal experimental design that allowed the researchers to determine (a) if the behavior changed when praise was provided as a consequence and (b) if change in the behavior was functional related (causally related) to manipulation of the consequence (presentation of contingent praise).
As a technology of behavior change, applied behavior analysis is analytic in that repeated and precise measurement of behavior continues to be essential. When the technology of applied behavior analysis is used in home, school, work, and community settings it is not always expected that a formal research study will be performed. Rather, the expectation is that the procedures used in the behavior change effort will be drawn from previous studies, that the behavior of the focus individual will still be measured directly to determine if desired behavior change is achieved, and implementation of the intervention or practice will be done with precision (and most often with measurement of fidelity).
Applied behavior analysis is a technology as well as a science. As with any technology it is essential that the specific elements of an intervention or practice are described with sufficient clarity and precision that someone reading the description can replicate what was done. This is often easier to say than to do. It would be one thing for an individual to describe her morning routine for getting out of bed and preparing for the day. It would be another thing entirely to describe her routine with sufficient precision that it could be replicated by someone reading her account. The same is true for describing how to teach reading or how to respond to a behavioral tantrum. Both as a science and as an approach to behavior change, applied behavior analysis is characterized by careful, complete, and specific description of procedures.
Applied behavior analysis is the application of behavioral principles. Behavioral principles have been defined from research in the experimental analysis of behavior and have been shown to be useful for explaining the conditions under which a behavior will or will not occur. Principles commonly applied in applied behavior analysis include reinforcement, punishment, extinction, stimulus control, generalization, and maintenance. There are countless ways these principles can be applied, and many examples of applied behavior analysis offer innovative applications. The use of positive reinforcement, for example, may include the delivery of praise and/or toys to develop toilet training skills with young children. But from a conceptual perspective it is not the praise or toys per se that are important, but the systematic application of positive reinforcement as a conceptual principle of behavior. Thus, applied behavior analysis should not be perceived as a particular procedure or set of intervention tricks (e.g., prompts to perform desired behavior; praise following correct behavior; correction, reprimand, or redirection following incorrect behavior), but the application of behavioral principles. It is this conceptual foundation that links the science and technology of applied behavior analysis.
Applied behavior analysis is a pragmatic enterprise. The focus is on using principles of behavior to achieve not just improved behavior, but behavior that is improved to a level that it is socially important. It is useful, but unimpressive, to reduce bouts of screaming from 20 events in a day to 15 events in a day. This is clearly an improvement, but 15 bouts of screaming in a day will likely remain socially unacceptable. When screaming has reduced to once per week, the family may define the intervention as successful. Both the science and technology of applied behavior analysis are held to a standard of change defined by the values of the people in the applied setting. The goal is to change behavior to a socially important level. For example, in determining how many screams are acceptable those concerned with the child's screaming (e.g., parents, teachers, the child) need to say what is and is not acceptable. The task of someone using applied behavior analysis is to apply behavioral principles to create a context that produces acceptable and valued levels of behavior.
Generalized Effects. In their seminal article Baer, Wolf, and Risley (1968) established the expectation that applied behavior analysis should focus not just on isolated behavior change, but generalized behavior change. Socially important change seldom is contained in a single context. Working successfully with one teacher in one room is an excellent accomplishment for a young child who has autism and a history of severe self-injury. But for this accomplishment to meet the standards of applied behavior analysis one would look for an approach that produces student success across locations, across instructional contents, and across instructors. Most important behaviors must be performed across many contexts or settings to be practical, useful, and functional for the individual. The focus on applied outcomes led the founders of applied behavior analysis to see generalization of behavior as a particularly valued achievement.
Maintenance of Effects. The strategies for producing a socially important change in behavior are not always the same ones used to maintain the effect across time. Although often discussed together, maintenance is different from generalization. Maintenance is based on a different outcome measure (durable responding across time) and also is affected by different variables (typically the type, level, and consistency of consequences). But like generalization, maintenance is important for applied behavior analysis due to its applied relevance: behavior change typically becomes socially important only if it endures for socially significant periods of time. It is wonderful to demonstrate that a child who previously refused to eat has learned to eat new foods and has been eating a healthy diet for a week. But the social importance of that demonstration lies in maintenance of the effect for months and years. Applied behavior analysis is a science and technology focused on practical, socially important behavior change. As such applied behavior analysis includes careful attention to the variables that affect maintenance.
Between the 1950s and the early 2000s applied behavior analysis demonstrated increasing value both as a science for understanding human behavior and as a technology for helping people achieve desired behavior change. Early applications of applied behavior analysis were most common with children and adults with severe disabilities, for whom other intervention approaches had proven less effective. As successful demonstrations of behavioral and lifestyle change became more common, the application of applied behavior analysis to typical work, school, and community contexts has increased. Descriptions of applied behavior analysis addressing extreme aggression by individuals with severe intellectual disabilities are now matched by examples of applied behavior analysis being used to teach social play skills, to decrease classroom talking out by children without disabilities, and to affect the quality and quantity of organizational outcomes.
An overarching message from this work is that the setting or context matters. Human behavior is more than personal willfulness. The physical conditions, social interactions, activities, and consequences within a setting do, over time, affect how people behave. Understanding why, when, and how these effects occur is at the heart of applied behavior analysis, and this is the promise that applied behavior analysis brings to help people understand both their current patterns of behavior and how to change dangerous and undesirable patterns of behavior.
Amidst the scientific legacy of applied behavior analysis two messages warrant emphasis. The first is that the consequences that follow a behavior will, over time, affect the likelihood of that behavior in the future. If a behavior (e.g., crying) is followed by a positive consequence (e.g., attention), then it is likely that under similar conditions in the future crying will become more likely. If a behavior (e.g., hitting) is followed by negative consequences (e.g., reprimand), then it is likely that under similar conditions in the future hitting will become less likely. Importantly, positive or negative cannot be determined a priori; what is negative for some people may actually be very reinforcing for others. This “law of effect” (Herrnstein, 1970) has an extensive research foundation and has been characterized in applied behavior analysis as the “function of the behavior.” Research indicates that an individual need not overtly define that he is engaging a behavior to achieve a particular function (i.e., to get or avoid something) for that function to occur. A person's behavior is being changed continually by the consequences in his world, whether he is aware or not of that process.
The importance of understanding the function of a behavior has become especially important for those using applied behavior analysis as a behavior change technology. Building on the basic research and philosophical foundations of B. F. Skinner and others, Sidney Bijou is recognized as among the first to encourage the careful recording and manipulation of antecedents and consequences in his early research on child communication (Bijou, Peterson & Ault, 1968). Then in 1977 Edward Carr wrote a compelling analysis of how access to positive and negative consequences affects the self-injurious behavior of children with severe disabilities. The model proposed by Carr created a taxonomy of behavioral functions that has direct relevance for both assessment of behavior problems and development of behavioral interventions. Carr's paper was followed in 1982 by a research analysis conducted by Brian Iwata and his colleagues (Iwata, Dorsey, Slifer, Bauman & Richman, 1982) that has become foundational reading for all students of applied behavior analysis. Iwata et al. (1982) documented effects of different behavioral consequences on the self-injury of adolescents and adults with severe intellectual disabilities. Their results were consistent with Carr's model and provided the foundation for both future research and clinical intervention. Iwata et al. found that when designing interventions for an individual with problem behavior, it is as (if not more) important to understand the behavioral function of the problem behavior (the specific consequence maintaining the behavior) as it is to define the type of disability or clinical diagnosis. An intervention that may be effective for a person who engaged in kicking maintained by attention may be completely ineffective for a different person who also engages in kicking but does so to avoid unpleasant tasks.
The impact of the vision, theory, and research that Bijou, Carr, and Iwata provided resulted in a major transformation in applied behavior analysis. Any current research or clinical intervention employing applied behavior analysis is in the early 2000s likely to include a functional analysis (the systematic analysis of behavioral function) or a functional behavioral assessment (the use of interviews, direction observation, and/or systematic analysis to define behavioral function). Because the consequences following a behavior are so important, it has become a professional expectation (and in many states a legal requirement) that care will be taken to assess the behavioral function of a behavior prior to designing behavioral interventions and supports.
The second major message in applied behavior analysis research in the early 2000s has been emphasis on investing in prevention of problem behavior. This means understanding and changing the events and conditions that occur before problem behaviors are performed. Since H. S. Terrace (1966) first summarized the scientific understanding of how stimuli (events, actions, and objects perceptible to the senses) can control behavior, there has been on-going study of how stimuli come to influence when and how behavior patterns develop. This research is important for understanding practical issues such as how the stimulus b comes to control the sound /b/ for a child learning to read, and how the stimulus “please help wash the dishes” from a parent comes to control the response “whine and cry” from a child who does not find dish washing reinforcing.
Applied behavior analysis emphasizes the important role of consequences, but research results also demonstrate the key role of events that precede target behaviors. Events that reliably precede behavior are important if specific consequences differentially occur in their presence or absence. For example, if a student's talking out is ignored by substitute teachers but reprimanded by the regular teacher, then the presence of who is teaching comes to control the behavior: The student is more likely to talk out when the substitute is there. Clinical applications of applied behavior analysis now regularly involve (a) manipulating the antecedent stimuli in a setting, and (b) investing in teaching new skills that produce functional outcomes for an individual, as ways to prevent problem behaviors. If a child with autism, for example, finds the background hum of florescent lights highly aversive, she may engage in screaming, throwing, and hitting as behaviors that result in her removal from the room with the aversive noise. Attention to prevention would suggest (a) remove the aversive noise by using lights that do not produce the negative hum, and (b) teach the child a communication skill that she can use to tell adults when she is in distress (without engaging in aversive histrionics). Changing the lights removes the aversiveness of the room and hence the function of screaming, throwing, and hitting—removal from the situation—no longer is relevant. Teaching her an alternative communication skill that produces the same effect (removal from aversive noise) gives her a socially appropriate (and more efficient) strategy for achieving the maintaining function.
The message from this example is that applied behavior analysis has matured beyond just the manipulation of positive and negative consequences. Both the research being done in the early 2000s, and the clinical applications of the technology, focus extensively on (a) the events that set the occasion (or prompt) problem behavior, and (b) alternative skills that can be taught to make problem behaviors unnecessary. In essence applied behavior analysis is being used to apply the principles of human behavior to the design of effective school, work, play, and home environments. This is an exciting development in that applied behavior analysis is being used as a technology to create situations that prevent problems as well as a technology to address problems when they develop.
The field of applied behavior analysis remains promising, but under-utilized in U.S. society. The contributions that basic principles of behavior can make to improve living and learning opportunities far outstrip current applications. The early decades of the twenty-first century are anticipated to show elaboration and scaling of these contributions. For the first years of the 2000s, however, (a) research in applied behavior analysis can be expected to improve the on-going understanding of how the environment affects human behavior, and (b) any clinical application of applied behavior analysis can be expected to (1) be based in application of basic behavioral principles, (2) include an initial functional behavioral assessment or functional analysis to identify the consequences maintaining the target behavior(s), (3) employ behavioral interventions that combine manipulation of prevention variables (e.g. antecedent stimuli and instruction on new skills) in addition to consequences, and (4) include measurement of behavior over time to assess effects.
See also:Classroom Management
Baer, D., Wolf, M., & Risley, T. (1968). Some current dimensions of applied behavior analysis. Journal of Applied Behavior Analysis. 1, 91–97.
Bijou, S., & Baer, D. (1961). Child development I: A systematic and empirical theory. Upper Saddle River, NJ: Prentice Hall.
Bijou, S., Peterson, R., & Ault, M. (1968). A method to integrate descriptive and experimental field studies at the level of data and empirical concepts. Journal of Applied Behavior Analysis, 1, 175–191.
Carr, E. G. (1977). The motivation of self-injurious behavior: A review of some hypotheses. Psychological Bulletin, 84, 800–816.
Herrnstein, R. J. (1970). On the law of effect. Journal of the Experimental Analysis of Behavior, 13, 243–266.
Honig, W. (1966). Operant behavior: areas of research and application. Upper Saddle River, NJ: Prentice-Hall.
Iwata, B., Dorsey, M., Slifer, K., Bauman, K., & Richman, G. (1982). Toward a functional analysis of self-injury. Analysis and Intervention in Developmental Disabilities, 2, 3–20.
Johnston, J., & Pennypacker, H. (1980). Strategies and tactics of human behavioral research. Mahwah, NJ: Lawrence Erlbaum.
Sidman, M. (1960). Tactics of scientific research: evaluating experimental data in psychology. Boston: Authors Cooperative.
Skinner, B. F. (1938). The behavior of organisms. New York: Appleton-Century Crofts.
Skinner, B. F. (1953). Science and human behavior. New York: Macmillan.
Terrace, H. S. (1966). Stimulus control. In W. Honig (Ed.), Operant Behavior: Areas of Research and Application (pp.271–344). Upper Saddle River, NJ: Prentice Hall.