Functional Assessment in Applied Behavior Analysis

Abstract

Functional assessment checklists and experimental functional analysis have become standard practice in applied behavior analysis. This chapter incudes an overview of existing methods and how they are utilized in various applied contexts. We discuss the standard functional analysis model first described by Iwata et al. (1982/1994) and significant variations on the standard model, such as trial-based functional analysis, interview-informed synthesized contingency analysis, among others. We also comment on the strengths and weaknesses of these methods and how they can complement one another during clinical decision-making. The article also discusses methods of analyzing functional assessments and assessment informed treatment. Finally, we discuss barriers and criticism of functional assessment and how these can be addressed by behavior analysts.

Appendix 1: Standard Functional Analysis Model Protocol

According to the methods by Iwata et al. (1982/1994a). Based on material developed by Dr. Brian A. Iwata, reproduced with permission of the author.

General

Rationale

The assessment presents the client with conditions that could result in problem behavior (PB). These conditions assume that either positive and negative reinforcement (either social or automatic) are consequences that maintain PB. By identifying which source of reinforcement accounts for a client’s problem behavior, individualized treatment programs can be developed.

Risk management

Although client protection is of paramount importance, it is best to use as little protective equipment as possible during the assessment. If physical risk is a concern, implement session termination criteria agreed with medical staff. If the behavior is so severe that all instances must be blocked, that person is not suitable to participate in a functional analysis. Refer such persons for consultation or participation in modified assessment procedures.

Location of the session

Select a room away from the activity in progress; a room with a one-way window is preferable. If a one-way window is not available, observers should sit away from the client and avoid eye contact or any type of interaction.

Medication

Ideally, functional testing should be performed in the absence of psychotropic medication. If this is not possible, the level of medication should be held constant until the evaluation is completed (for further details see Cox & Virues-Ortega, 2016).

Informed consent

It is important to ensure that parents, advocates, and members of the interdisciplinary team understand the purpose of the evaluation and approve of its use.

Session duration

Sessions usually last 10 min, unless completion criteria are met, and are timed with a stopwatch. In the event of time constrains, it is possible to reduce session duration to 5 min.

Target behavior

The behavior that is scored by observers and that receives consequences from a therapist in the demand and attention conditions. In most circumstances, each instance of the target behavior will be recorded.

Therapist

This is the individual who interacts with the client. Ideally, a different therapist is associated with each of the conditions listed below (e.g., one therapist leads all attention sessions, another leads all demand sessions, etc.). If a single therapist leads all sessions, it may be helpful to wear a different color shirt when running each type of session.

Alone/Ignore/No Interaction

Objective

This is a test condition for automatic reinforcement. If the behavior occurs at a high rate in the absence of social interaction, the behavior is likely to produce its own reinforcers (i.e., automatic reinforcement).

Antecedents

No therapist is necessary for this condition or, if present, no social interaction should take place (including eye contact or behavior-contingent movement or proximity). The room should contain no toys, leisure materials or reinforcers.

Consequence

No social consequences for the behavior (e.g., no comments or changes in facial expression).

Attention

Objective

This is a test condition for positive social reinforcement. If the rate of a target behavior is higher in this condition than in other conditions, the behavior is likely to be maintained by attention as a consequence.

Antecedents

The room should contain some toys or leisure materials, which are freely available during the session. Sessions begin with the therapist saying that he or she needs to “do some work” or something to that effect (i.e., not attending to the client). From this point on, the therapist does not interact with the client unless the target behavior occurs.

Consequence

After each target behavior, the therapist approaches the client and makes a statement of concern, which may be accompanied by brief physical contact. For example, while saying “Don’t do that, you might hurt yourself,” the therapist may also gently hold the client’s arm, stroke the client’s back in a “reassuring” motion, or even look at and point to the location on the client’s body that is suffering the injury. These interactions should last between 5 and 10 s. If the client continues the behavior during the interaction, the interaction should also continue.

Demand

Objective

This is a test condition for negative social reinforcement. If a target behavior occurs more frequently in this condition, the behavior is likely to be maintained by escape from the demands of the task.

Antecedents

The therapist begins the session by presenting a relevant task (e.g., an unpreferred and age-appropriate academic activity) to the client, e.g., “Peter, let’s do some homework” If the client does not comply after 5 s, the therapist demonstrates the correct response or provides a tactile prompt. If the client does not comply after 5 s of the prompt, the therapist physically guides the client through the task. This form of instructional trials continues until the end of the session. It is usually best to present a variety of tasks. Consider including tasks that the client is suspected of disliking. There should be no unscripted forms of interaction between the therapist and client.

Consequence

After each occurrence of a target behavior, the demands end immediately without comment from the therapist, and the next trial is delayed by about 30 s.

Play

Objective

This condition functions as a control for the other three test conditions. Specifically, the client is not alone, attention is available, and no tasks are presented. As a result, the target behavior should occur less frequently in this condition. If the target behavior occurs at a high rate in this condition, it is possible that the behavior produces its own reinforcers and that alternative activities do not compete with the problem behavior.

Antecedents

The room must contain known toys, leisure materials and reinforcers. The client should have free access to the objects. At 30-s intervals, the therapist should approach the client and engage in conversation for 5–10 s. In addition, the therapist should respond to any appropriate social behavior initiated by the client.

Consequence

There are no consequences for problem behavior, except that care should be delayed if the problem behavior occurs just as care is about to be delivered.

Other Considerations

Sequence of conditions

Assessment conditions are usually presented in an alternating sequence in the following order: alone, attention, play, and demand. Whenever possible, it is advisable to keep the therapist, environment, and materials within each condition constant (e.g., the same therapist, environment, and materials are used in all play sessions). These steps are taken to facilitate discrimination between assessment conditions. Sometimes, clients do not discriminate very well between the different conditions; this can produce unclear results. Therefore, two alternative arrangements are possible. In the reversal design, one continues with a single condition (e.g., gambling) until the data are stable; then the next condition (e.g., demand) is introduced, and so on. The other alternative is pairwise evaluation, in which a test condition (e.g., attention) is alternated with the control (game), followed by another test-control pair (demand vs. game), and so on.

Other conditions

The above conditions represent a general set of evaluation “probes” that simulate the natural environment. Some clients have unusual histories that may require modifications of the above conditions or the addition of new conditions. For example, an individual whose problem behavior is maintained by attention might exhibit problem behavior only if another person (a peer) is receiving attention. Another may exhibit problem behavior for escape from some tasks (e.g., self-care), but not others. If the initial assessment data are unclear, application consistency has been verified, and conditions have been attempted using the reversal design, consideration should be given to systematically altering the assessment conditions.