Functional Behavioral Assessment

Functional behavioral assessment (FBA) refers to a range of assessment strategies to identify variables that influence behavior in the natural environment. The purposes of FBA are to (1) define and describe problem behaviors, (2) predict when problem behaviors are likely and unlikely to occur, and (3) identify consequences that maintain the behavior (i.e., the functions of the behavior). Learning theory has provided the foundation for the principles of FBA. A variety of processes are used in conducting an FBA, including direct (e.g., behavioral observations) and indirect methods (e.g., interviews) and functional (or experimental) analysis. FBA data are used to create individualized behavioral interventions that account for the identified predictive variables and behavioral functions.

Theoretical Principles

Functional behavioral assessment is rooted in the philosophical perspectives of empiricism, contextual-ism, and determinism. These positions assume that many problem behaviors are learned, that learning occurs through particular interactions, and that these interactions can be altered to produce desired changes in behavior. Learned behaviors refer to behaviors that occur (or fail to occur) because of how they are responded to by significant others or events in the individual’s environment. Principles of operant conditioning contribute heavily to the learning theory principles underlying functional behavioral assessment.

The earliest formulation of operant principles is known as the law of effect. This law is based on the observation that voluntary behavior is influenced by its effects, namely, its consequences. According to the early version of the law of effect, behavior that produces satisfying consequences tends to become more frequent over time; behavior that produces discomfort tends to become less frequent. Subsequent revisions of the law of effect recognize the importance of relevant situational cues. That is, a behavior may have positive effects in one situation but negative effects in another situation. As a result, individuals become sensitive to situational cues, especially to antecedent cues that precede their behavior and allow them to determine whether a behavior is likely to produce positive or negative effects. Behavior is influenced not only by the effects that follow it but also by the situational cues that precede it.

During operant conditioning, the consequences of a behavior influence the frequency of the behavior in the future. Any operant behavior can be strengthened or weakened depending on the type of consequences that follow the behavior. Reinforcers are types of consequences that strengthen a behavior. Punishers are types of consequences that cause a behavior to become less frequent. Reinforcers and punishers can be either positive or negative, depending on whether the consequence involves the addition or obtaining of something or the removal or escape of something. Positive reinforcement, then, strengthens behavior through the addition of something desirable (e.g., money, smiles). Negative reinforcement strengthens behavior through the removal of something undesirable (e.g., chores). Positive punishment weakens behavior through the addition of something undesirable (e.g., social disapproval), and negative punishment weakens behavior through the removal of something desirable (e.g., access to a favorite toy or activity). According to behavioral theory, behavior is not random—those behaviors with strong reinforcement histories continue, and those that have been more strongly punished than reinforced will discontinue.

It should be noted that there are often competing consequences for behaviors. A behavior may be followed by multiple consequences, some of which are reinforcing and some that are punishing. Whether a behavior is strengthened or weakened by those consequences depends on whether the overall pattern of consequences is perceived as more reinforcing or more punishing. For example, adolescents who miss curfew may be reinforced by having increased opportunity to socialize with peers, but they may be punished at home. Some adolescents may perceive the reinforcement to outweigh the punishment and so continue to miss curfew. Others may perceive the punishment as strong enough to deter similar behavior in the future. The strength of the consequence, the delay between the behavior and the consequence, and the frequency of the consequences all play a part in their impact.

There are essentially six functions of behavior. Three of these involve positive reinforcement and three involve negative reinforcement:

Obtain internal stimulation: A behavior may continue because it provides desirable stimulation to the individual. For example, a child might rock back and forth because it is soothing to do so.
Obtain attention: A behavior may continue because it draws attention from others (attention that is perceived as reinforcing). For example, a female adolescent might dress provocatively because it results in attention from boys at school.
Obtain activities or objects: A behavior may continue because it results in activities or the acquisition of objects. For example, stealing is reinforced by the objects acquired.
Escape/avoid internal stimulation: A behavior might be strengthened because doing the behavior allows the individual to avoid an unpleasant sensation. For example, children with ear infections often pull on their ears because it lessens the pain of the earache.
Escape/avoid tasks or activities: A behavior might be strengthened because it allows the individual to escape or avoid an unpleasant task or activity. For example, children who play with objects in their desk may be doing so because it keeps them from having to do work they do not like.
Escape/avoid attention: A behavior might be strengthened by helping an individual avoid unwanted attention. For example, an individual might avoid his or her job because at work, the boss continually degrades and belittles the individual.

Antecedents are also important to consider in an FBA. Antecedents are stimuli that precede behaviors and often provide clues about which behaviors will pay off. They do not, however, cause behavior. For example, a red stoplight sets the stage for the behavior of hitting the brakes, because an individual’s learning history suggests that doing so will be reinforced (i.e., it is negatively reinforced by the avoidance of getting a ticket or getting in an accident). Antecedent cues that preceded behaviors that were reinforced in the past tend to set up a pattern for repeating those behaviors when the antecedents appear in the future (so most individuals are likely to brake when they see a red stoplight in the future). Antecedent cues that preceded behaviors that were punished in the past signal an individual not to repeat those behaviors in the future.

Not all antecedents immediately precede behaviors. Some antecedents occur very close to the behavior. For example, a teacher asking students to get out their books serves as an antecedent for a child getting out his or her book. Other antecedents are more temporally distant but still impact behavior. For example, setting events refer to conditions that impact the occurrence of the behavior, and they often are more distant. For example, a person might be more likely to do a poor job at work (behavior) if he or she did not get a good night’s sleep the night before (setting event). All antecedents need to be considered when doing an FBA.

Historical Progression

By the late 1960s, the methods used to record behavior in laboratories became known as the experimental operant paradigm, whereas the consequences that were programmed and their effects on responding became known as the principles of operant conditioning. In the 1970s, researchers began applying the principles of operant conditioning to clinically relevant and socially significant problems. For such applications to occur, methods that had been used to program consequences in the laboratory (e.g., schedules of food reinforcement) needed to be replaced with procedures that could be used in clinical and social settings. Similarly, in the absence of mechanical recording devices, systematic observation became the standard means of reliably detecting changes in an individual’s behavior. During the 1970s and early 1980s, the term behavior modification was coined. Behavior modification is the systematic and scientific use of behavior principles to improve individuals’ thoughts, feelings, and actions. It was during this period that many strategies involved in conducting a functional behavioral assessment were developed.

Processes and Methods

Behavioral analysts incorporate multiple sources of information when making clinical judgments. Each method employed during the course of a functional assessment involves thoroughly describing the problem behavior(s) and gaining an understanding of the events leading to and following the behavior. The clinician gathers information pertaining to the individual client and the function(s) of his or her behavior primarily through the use of interviews, direct observations, and functional analysis.

Indirect Assessment

Information gathered via interviews is considered to be the launching pad for each of the other FBA procedures. Naturally, conversing with the client about his or her behaviors is often the first step taken by professionals. Clinicians perform interviews with the client and those individuals interacting with the client on a frequent basis (e.g., family members, teachers) to glean rich information pertaining to the problem behavior. Questions typically are designed to elicit information regarding the events prior to and following the client’s behavior as well as details about the behavior itself. The interview format may vary in terms of formality based on the professional’s preferences and the needs of the individual client.

The behavioral interviewer attempts to gain specific information in order to develop an individualized treatment plan. During each interview, however, the clinician is also mindful of building good rapport with the client. It is essential that the questions posed by clinicians comprise an interview schedule that sufficiently elicits specific and accurate information concerning high-risk behaviors, such as violence, suicide, and abuse.

Other indirect FBA methods include rating scales and behavior checklists. These tools are used to identify problem behaviors as well as to ascertain relevant setting events, antecedents, and functions (consequences) for those behaviors. Like interviews, other indirect measures may be completed with the client or relevant individuals in his or her environment.

Subsequent to interviews and other indirect measures, hypotheses are often formulated regarding variables that predict behavior occurrence as well as the consequences maintaining problem behavior. In some circumstances, however, indirect information does not elucidate a clear pattern that can be used to make predictions about behavior. Alternatively, the clinician may want to gather additional information before proceeding with an intervention plan. Behavioral observations can provide additional information with which to form hypotheses or to support or disconfirm those made following interviews. Direct observation is frequently informed by the information gained through indirect methods. For example, a clinician may use the specific definitions of a behavior arrived at through interviews as the basis for observing that behavior in the individual’s environment.

Behavioral Observation

Systematic observation procedures are an integral component of this empirical approach to understanding behavior. Indirect assessment relies on the client, or another informant knowledgeable about the behavior in question, to recall and accurately report information. In contrast, direct behavioral observation allows the recorder to witness and immediately record information about the target behavior, antecedents, and consequences. The observer may be the professional conducting the FBA or an individual who works with the client frequently (e.g., group home employee, teacher, parent). Alternatively, the client may perform this form of data collection pertaining to his or her own behavior. For example, the clinician may ask the client to record each time he or she smokes a cigarette. Additionally, the client would record the events occurring before and after the smoking behavior.

The observer should receive training on how to accurately record antecedent-behavior-consequence (ABC) observations.

Clinicians are able to choose from several approaches to observe samples of the client’s behavior: event sampling, duration recording, and interval recording. Event sampling or frequency recording involves counting the number of times that the behavior is seen occurring during a specified time period. Duration recording accounts for how long an instance of the behavior lasts. Interval recording methods provide an indication of the percentage of prespecified intervals during which the behavior is observed to occur. The observer then extrapolates the sample data to help explain the entire behavior. To help ensure that the observer witnesses the target behavior, it is important to schedule the observation period at times when the behavior is most likely to occur. It is also important that observations are scheduled at times and in places where the behavior is not typically seen. Recording observational data in a variety of settings and times throughout the day will allow the clinician to understand when the behavior does and does not occur and can be useful in developing interventions.

The amount of information gained through behavioral observation techniques is vast and can seem overwhelming during the initial stages of investigation. With time, common themes emerge from clients’ data. Ideally, behavioral observation should occur until themes emerge from the data concerning the individual client.

Functional Analysis

Clinicians employ functional analysis procedures to better understand the purpose, or function, of the client’s behavior. Although the goals of interviews, behavioral observation, and functional analysis are similar, functional analysis procedures require the clinician to directly manipulate the environment. During interview and behavioral observation sessions, the clinician attempts to understand the client’s behavior within the context of his or her everyday life. Interviews and behavioral observation allow the clinician to develop hypotheses about the antecedents and consequences that maintain the client’s target behavior. The functional analysis process allows the clinician to test the validity of these hypotheses through experimentation. Identified antecedents and consequences are isolated and altered while the target behavior is monitored. Often, functional analysis includes four conditions—one in which attention is presented contingent upon the problem behavior being exhibited, one in which escape is allowed contingent upon the behavior, a third in which the client is alone, and a fourth in which no task demands are made and reinforcement is available. The clinician utilizes data kept on the target behavior to determine which environmental modifications have the greatest impact. The results of functional analyses allow clinicians to develop individualized, data-based interventions for clients.

Developing Interventions Based on a Functional Behavior Assessment

Once relevant antecedents and functions of behavior have been identified, they are used in the development of a plan for remediating the problem behavior. With regard to antecedents, the intervention should decrease or minimize antecedents that lead to the problem behavior as well as increase those that are not shown to lead to the behavior. For example, if it is determined that overeating at dinner is more likely when a person skips breakfast and when he or she is dining alone, the intervention likely would include encouraging the individual to eat breakfast every day and to make plans to dine with others as often as possible. Alternatively, it is important to consider those antecedents that precede appropriate behavior and to increase their frequency. For example, if the individual in the example above tends to eat moderately when seated at the dining room table with the television off, the plan might include encouraging the individual to set up those conditions when eating dinner. Both temporally distant setting events and immediate precursors to the behavior should be considered when addressing antecedents.

Building interventions based on functions is a bit more complicated, because there are several factors to consider. According to behavioral theory, all problem behavior is being reinforced (i.e., only reinforced behaviors continue). One part of the intervention, then, is to find a way to decrease the problem behavior, either by punishing it or by lessening its reinforcement value. However, simply decreasing a problem behavior is unlikely to be effective unless an appropriate alternative behavior is put in its place. If, for example, a child learns that he or she can get access to toys by grabbing them from other children, a plan might involve making sure that an adult prevents the child from playing with the toy after it is snatched from someone else. The plan might also include punishing the behavior of grabbing toys from others (e.g., by having the child play separately from other children for a brief period). In this case, the plan is to ensure that the behavior of grabbing toys does not lead to the reinforcer of getting to play with toys that have been grabbed. If the behavior of grabbing decreases and there is no behavior to put in its place, it is likely that the behavior may return or another inappropriate behavior (e.g., hitting) might appear. It is important, therefore, to include in interventions a way to promote more appropriate behaviors that might serve the same function as the problem behavior. In the above example, the child might be taught how to appropriately ask for toys he or she wants to play with. The idea is that the child learns to get access to toys (reinforcement) for a more appropriate behavior than grabbing at them.

In considering appropriate behaviors to replace inappropriate ones, it is important to recognize that some individuals will need to be taught those behaviors. In the examples used above, an individual might need to be taught how to eat slowly and mindfully to avoid overeating. Children might need to be taught how to ask others to give them a toy.

Overall, comprehensive interventions based on FBA procedures should include multiple strategies, each of which is individually tailored to the client and based on the specific findings from the FBA. Following the FBA process helps ensure that interventions are based less on the topography (or name) of a problem behavior and more on the circumstances surrounding the behavior for a particular individual. Research has shown that such interventions are more likely to result in reduction of problem behaviors and/or increases in appropriate replacement behaviors.

References:

Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice Hall.
Bellack, A. S., & Hersen, M. (1998). Behavioral assessment: A practical handbook. Boston: Allyn & Bacon.
Cooper, J. O., Heron, T. E., & Heward, W. L. (2007). Applied behavior analysis(2nd ed.). Upper Saddle River, NJ: Pearson.
Haynes, S. N., & O’Brien, W. H. (2000). Principles and practice of behavioral assessment. New York: Kluwer.
Horner, R. H. (1994). Functional assessment: Contributions and future directions. Journal of Applied Behavior Analysis, 28, 401—404.
Luiselli, J. K., & Cameron, M. J. (1998). Antecedent control: Innovative approaches to behavioral support. Baltimore: Brooks.
O’Neill, R. E., Horner, R. H., Albin, R. W., Sprague, J. R., Storey, K., & Newton, J. S. (1997). Functional assessment and program development for problem behavior: A practical handbook (2nd ed.). Pacific Grove, CA: Brooks/Cole.
Plaud, J. J., & Gaither, G. A. (1996). Behavioral momentum: Implications and development from reinforcement theories. Behavior Modification, 20, 183-201.
Repp, A. C., & Horner, R. H. (1999). Functional analysis of problem behavior. Belmont, CA: Wadsworth.
Skinner, B. F. (1953). Science and human behavior. New York: Macmillan.