Applied Behavior Analysis

Applied behavior analysis is a methodology for systematically applying the principles of learning theory to develop interventions that will improve socially significant behaviors to a meaningful degree, as well as demonstrate that the interventions employed are responsible for the improvement in behavior. Applied behavior analysis has been repeatedly demonstrated to be a highly effective approach across a wide range of problems and environments, including education, mental health and mental retardation, parent training, environmental management, and organizational management. Applied behavior analysis is a specialty used by various professions. It is not regulated by most states, except as part of psychology or other established professions, although there have been a few attempts to recognize trained and qualified behavior analysts over the years. Most recently the nonprofit Behavior Analyst Certification Board has promoted a national certification program to identify and credential qualified practitioners and trainers.

Principles and Techniques

Several terms besides applied behavior analysis have been used to describe intervention methods based on behavioral learning theories, including behavior modification, behavior therapy, and others. Although sometimes used interchangeably, there are possible distinctions made. Applied behavior analysis is usedmost often for the orientation that derives predominantly from Skinnerian operant conditioning and follows a radical behavioral philosophy. Although other behavioral orientations often utilize operant principles to differing degrees, they typically place a greater emphasis on classical conditioning processes (the neobehavioristic mediational model) or cognitions and perceptions as targets for change (social learning theory and cognitive behavior modification) than does applied behavior analysis.

Operant conditioning eschews hypothetical mental constructs as explanatory contracts. Behaviors are viewed as being selected by environmental consequences, much like adaptive changes are in Darwinian evolution, rather than being emitted to serve some future purpose. That is, a child does not cry in order to attract attention but cries because crying has resulted in reinforcing consequences in similar situations in the past (unless, of course, the crying is in response to actual physical discomfort).

The central precept of applied behavior analysis is that behaviors are under the control of environmental stimuli. Functional relationships are described by a three-term contingency that consists of antecedents, responses (behaviors), and consequences. At times these are referred to as the ABCs of behavior. Early uses of behavior analysis focused primarily on changing behaviors through manipulating consequences. All consequences are seen as directly influencing whether behaviors will recur in the future. Consequences can be grouped into three main types: reinforcing, punishing, or neutral stimuli.

The first class of consequences, reinforcers, consists of events that increase the future probability of a behavior they immediately follow. These include events that strengthen behaviors when they are presented following the behavior, such as food, attention, or social praise. This operation is referred to as positive reinforcement. For example, a child may learn to apologize because the apology consistently is followed by parental praise. Behaviors also can be strengthened through the removal of an aversive (negative) stimulus following a behavior. This operation is termed negative reinforcement. This would be the case if a child learns to apologize if the apology terminates (or avoids) being scolded by his or her parents.

Reinforcers can either be biologically preestablished (primary reinforcers), such as food or water, or can acquire reinforcing properties through careful pairing with primary reinforcers (conditioned reinforcers). Most reinforcers are differentially effective with different people rather than being universal. This is particularly true of conditioned reinforcers such as praise or tokens. The schedule of reinforcement used to deliver reinforcers also is important. Early in the process of strengthening a behavior, reinforcers are typically delivered on a continuous schedule, where a reinforcer is given each time the response occurs. Later in the process, various schedules are used that deliver reinforcers more intermittently and help increase resistance to extinction and/or produce specific types of responding patterns.

Punishment consists of two operations, but these weaken the likelihood of behaviors recurring. In positive punishment, the presentation of the consequence following the behavior results in weakening future occurrences of a behavior. An example might be a brief swat on the bottom when a young child chases a ball into the street. Removing a reinforcing stimulus contingent upon a behavior also can weaken it. This isnegative punishment and includes, for example, taking away television-watching privileges for a short time to weaken a child's lying behavior. Applied behavior analysts typically advocate negative punishment as a more appropriate reductive method in most cases.

Behavior also may be decreased through the use of extinction, where the connection between a response and its maintaining consequences is discontinued, leading to a progressive decline in the rate of a previously reinforced response.

Antecedents, which are the front end of the three-term contingency, influence behaviors largely through a prior history of differential association with reinforcing or punishing consequences. When a child's behavior has been reinforced previously in a situation, the likelihood of that behavior is heightened under similar circumstances. It is lowered in situations in which reinforcement has been consistently withheld or punished. For example, a child whose father consistently buys him candy in a store when he whines is likely to exhibit similar behaviors in the future when shopping with his father. With his mother, who does not ‘give in,’ the child will learn not to whine.

Another type of antecedent stimulus used programmatically includes visual, verbal, or physical promptsgiven to increase the likelihood a child will respond appropriately to the given situation. For example, pictures of objects may be placed with letters to assist a child in learning letter sounds. These can then beremoved either abruptly or through a gradual process called fading. Modeling can be seen as a special form of prompting in which someone demonstrates a desired behavior to increase its likelihood.

When a behavior is not present in the individual's repertoire, the procedure of shaping or successive approximations may be used. Shaping involves reinforcing progressively closer approximations of the desired behavior. For example, in teaching new words to a young child, the child is reinforced for vocalizations that are increasingly more like the desired word. As the sequence progresses, the word must be more and more like the target word for the child to receive reinforcement.

Particularly important behavioral concepts for instructional applications include discrimination, generalization, and concept formation. Behaviorally, concept formation occurs when the same response occurs to a group of discriminably different objects that have some aspect in common (as well as responding differently to other classes of stimuli). Concept formation involves both generalization and discrimination: generalization within classes and discrimination between classes. Thus, a child who responds ‘dog’ to different examples of dog but not cats or other animals is exhibiting concept formation.

Many other effective techniques have evolved over time from this relatively small set of basic principles. In addition to those just mentioned, these include procedures such as overcorrection, token economies, time-out, response cost, self-monitoring, and task analysis.

Methodological Practices

Although numerous step-by-step models have been proposed for developing, implementing, and evaluating applied behavior analytic interventions, each generally incorporates at least the following components:

1. Selecting a behavioral excess or deficit for change
2. Establishing a method for measuring behaviors
3. Measuring the baseline (current) level of performance
4. Specifying goals and objectives
5. Designing and implementing interventions to teach or strengthen behaviors/skills and/or to reduce excessive behaviors
6. Continuously measuring the behaviors to determine the effects of the intervention
7. Modifying the intervention based on ongoing measurements

In measuring behavior, applied behavior analysts focus on observable behaviors recorded using techniques such as direct observational recording (event, duration, or latency recording using continuous, time sampling or interval recording) and analysis of permanent products. In recent years, the practice offunctional behavioral assessment —where the functions or maintaining factors of an existing behavior are determined experimentally prior to designing an intervention for implementation—has become increasingly more important.

A particularly important aspect of applied behavior analysis is the demonstration that the intervention used, rather than some unspecified extraneous variable, caused the behavior change. Applied behavior analysts accomplish this primarily through within-subject experimental designs that focus on the functional relationships between intervention changes and changes in target behavior. Typical experimental designs include reversal (ABAB) designs, multiple baseline designs, multiple-treatment designs, and changing criterion designs. The generality of effects is established through replication across subjects over time.

Historical and Current Applications

The field of operant conditioning, or the experimental analysis of behavior, began to grow exponentially during the 1930s. At that time B. F. Skinner built upon the basic concepts of radical behaviorism espoused by John Watson. Within a few years early attempts were made to apply Skinner's model of operant conditioning to human behavior and development. This was accompanied by a number of research programs, mostly in institutional settings. That work used arbitrary responses, as in animal operant research, to investigate the applicability of operant conditioning to humans. Concurrent with this, a few psychologists began working with the mentally retarded and mentally ill in clinical settings. The often-cited first human application, published in 1949, demonstrated that squirts of milk could be used to establish arm raising in a profoundly retarded 18-year-old man. Examples of other early applications included use of psychiatric nurses to change behaviors in psychotic patients and early attempts to change problem behaviors in preschool children.

During the late 1960s the field became increasingly focused on the application of operant principles to socially important problems rather than artificial laboratory research. The Journal of Applied Behavior Analysis was founded in 1968 as a primary outlet for this research. In the initial volume, Donald Baer, Montrose Wolf, and Todd Risley proposed seven defining dimensions of applied behavior analysis. When the same authors reviewed the status of these dimensions almost 20 years later, they concluded that the dimensions continued to be relevant. First, the behaviors or stimuli studied must be selected for applied (practical) rather than theoretical significance. They must need improvement and be measurable (behavioral). It is essential that the factors responsible for the occurrence or nonoccurrence of the behavior be established (analytic). The behavior change procedure used needs to be described in terms of the relevant principles (conceptual systems) and must be completely identified and described to permit replication of the interventions (technological). Behavioral techniques used must produce significant practical effects (effective). Lastly, those effects must be stable over time and situations, or they must extend to untrained responses (generality).

Since the 1950s applied behavior analysis has become an increasingly important applied field of psychology that has been used to modify behaviors across numerous areas of human functioning, including mental health and mental retardation, general and special education, organizational performance management, environmental management, and behavioral medicine.

Applied behavior analysis has been extended across many areas of education over the past few decades. Besides the obvious clinical applications for individuals with disabilities, a number of instructional methods used across regular education, special education, or adult and higher education have been grounded in applied behavior analytic principles. These include direct instruction, precision teaching, and personalized instruction. Although there have been numerous studies demonstrating the effectiveness of these approaches, they have not been widely adopted. In fact, numerous appeals have been made to have education be more accepting of applied behavior analysis as a basis for instructional innovations.

Entry Citation:

"Applied Behavior Analysis." Encyclopedia of Educational Psychology. 2008. SAGE Publications. 22 Sep. 2009. .