Which of the following actions would be useful if you wanted to record a target behavior?

Identification of Target Behaviors

Target behaviors are an important first step in treatment for both behavior analysts and social work practitioners. Current disciplinary myths suggest that social workers and behavior analysts only attend to covert and overt behaviors, respectively. Covert behaviors that occur within the skin (Skinner, 1974) are difficult to directly observe and are easily discarded by behavior analysts (c.f., Skinner, 1953, for inclusionary argument of covert behavior within an analysis of human behavior). Conversely, overt behaviors occur in ways that are easier to directly observe, often increasing the ability to establish reliability between multiple observers. Both Susan and Barry presented with a specific behavior of concern (e.g., gambling and insomnia, respectively), and both had a series of comorbid mental health diagnoses ranging from OCD and adult onset ADD to PTSD and depression. For instance, Susan’s primary behavior of concern was her gambling, even though she reported negative thoughts and depressive symptoms during intake. Barry presented with insomnia, in addition to PTSD, depression, suicidality, and current alcohol use. Given the high rates of mental health comorbidity (e.g., Ledgerwood & Petry, 2004; McCormick, Russo, Ramirez, & Taber, 1984), secondary target behaviors (e.g., self-reported and clinician reported covert symptoms) should be selected to create a balance between covert and overt behaviors.

Target behaviors should be idiosyncratically identified based on the client self-report and clinician observation during intake. Treatment outcomes for each selected target behavior are most effective when developed with client input. Further, the combination of both client self-report and clinician assessment can be useful when targeting covert events as target behaviors. Potential target behaviors for Susan included gambling-related thoughts and urges, gambling episodes, and perceived control over gambling episodes. A range of potential target behaviors for Barry included frequency of insomnia, avoidance of activities that reminded him of his traumas from his military service, survivor guilt mixed with suicidal thoughts, and frequency/intensity of depressive symptoms.

I.C.2. Identify the Target Behavior

Target behaviors can be identified in a number of ways: (1) self-report from either the client or in the case of children, legal guardians or caregivers; (2) paper-and-pencil surveys; or (3) natural observations. Establishing a baseline, which is taken prior to beginning any formal intervention, involves data collection to gather information about frequency, duration, and the topography of the behavior. The baseline data are then used to measure effectiveness of the procedure. It is also useful during the course of treatment. If there is not a positive change in the data or the trend line, modifications can be made to the scenes to incorporate new information and alter the scripts to reflect this new information.

6.01.1.11.2 Measuring and Recording Behaviors

Once the target behavior has been identified, the client and clinician may begin measuring and recording the behaviors. There are several approaches to this which are described in more detail later in this article. In general, measurement includes direct assessment of the target behavior, either through direct observation of the client, or self-report. This may include information on the number of discrete occurrences of the target behavior and the duration and intensity of the behavior (Cooper et al., 2020). In terms of measuring the frequency of a behavior, this can be done for each discrete occurrence in a particular amount of time (event recording), or it can be measured as the number of time intervals during an observation period in which the target behavior occurs (interval recording) (Cooper et al., 2020).

Constructing a functional analysis and functional analytic clinical case diagram

Once the target behaviors, causal variables, and causal relations have been identified, it is important to synthesize the information in a clinically useful manner in order to inform the design of an intervention and communicate the findings to others. The FACCD is one strategy for accomplishing these two goals. The construction of FACCDs is outlined in detail in Haynes et al. (2019). Here we will provide an abbreviated step-by-step strategy that can guide FACCD construction.

Step 1: Obtain informed consent. Informed consent is a collaborative process involving the clinician, the client, and other relevant persons (e.g., family members). The process is designed to help all parties arrive at an agreement about the purposes of the assessment, the relative costs and benefits of the assessment, assessment strategies to be used, and confidentiality.

Step 2: Evaluate the need for referral and the safety. After obtaining consent, a clinician should determine whether the nature of the client’s problems could be more effectively addressed by another professional. Additionally, in cases where risk for harm is detected, assessment and interventions targeting risk reduction are prioritized over the gathering of additional assessment information.

Step 3: Identify the target behaviors and treatment goals. The goal of this step is to identify a wide range of possible target behaviors. Behavioral interviews and self-report inventories are the primary method used to identify behavior problems and treatment goals in this initial stage of the functional analysis. Behavioral observation, such as during a clinical interview, can also help identify behavior problems in some assessment contexts.

Step 4: Specify target behavior response modes. Once the client’s target behaviors have been identified, specific information about response modes is gathered. In this step, the clinician generates a comprehensive operational definition of each target behavior that will include cognitive, emotional, physiological, and over-motor components.

Step 5: Specify target behavior dimensions. The clinician must determine which dimensions are most relevant to the overall functional analysis and FACCD. For some target behaviors, frequency is the most relevant dimension (e.g., number of days absent from work). For other target behaviors, magnitude is most important (e.g., pain intensity). And, in many cases, more than one dimension is relevant.

Step 6: Estimate target behavior importance. FACCDs are designed to help the clinician make decisions about intervention foci. A central aspect of this decision is to prioritize target behaviors. Specifically, the clinician aims to determine which target behaviors are most important for client wellbeing and functioning. Thus, a rating of the relative importance of target behaviors is needed.

Relative importance can be estimated in a number of ways. One important consideration is risk of harm to self or others. A second set of considerations is target behavior dimensions: Target behaviors can be prioritized according severity, frequency, and duration. Third, target behaviors can be ranked according to their impact on quality of life. Finally, the clinician must consider the client’s subjective rating of importance.

Step 7: Identify the target behavior effects. Target behaviors influence other behaviors. For example, the intensity, frequency, or duration of panic attacks can influence interpersonal relationships and social isolation, as noted in our earlier example.

Step 8: Identify the strength and direction relationships among target behaviors. Target behaviors can interact in important ways, such as in behavior chains and functional response classes. They can also have unidirectional or bidirectional causal relations. Identifying the relationships among target behaviors allows the clinician to better estimate the possible impact of an intervention, as illustrated in Fig. 16.1.

Step 9: Identify causal variables. Identifying causal variables is one of the most critical steps in the generation of a FACCD because most interventions involve the modification of causal variables in order to exert an effect on target behaviors. For example, if a client’s depressed mood were mainly caused by marital distress, then marital therapy would be an appropriate intervention. Alternatively, if a client’s depressed mood were mainly caused by social isolation, then social skills training would be an appropriate intervention.

Step 10: Estimate the modifiability of causal variables. Causal variables can differ in terms of modifiability. A causal variable that can be modified (e.g., social behaviors) is more relevant to intervention design than unmodifiable or less modifiable causal variables (e.g., childhood abuse, traumatic brain damage).

Step 11: Estimate the strength and direction of relations among causal variables and target behaviors. After identifying the important and modifiable causal variables, the clinician will then estimate the strength and direction of relationships among these variables.

Step 12: Estimate the strength and direction of causal relations among causal variables. There can be unidirectional and bidirectional causal relations of varying strength among causal variables. As with target behaviors, these causal relations influence the selection of the best intervention focus. Modifying a causal variable that has multiple or stronger functional relations with other causal variables and target behaviors will result in a larger intervention effect than a casual variable that has fewer or weaker functional relations.

Step 13: Identify potential moderator and mediator variables. A moderator variable alters the strength or the direction of a causal relation between two other variables. A moderator variable can increase, decrease, and in some cases reverse the effects of a causal variable. For example, “social support” is a commonly referenced moderator variable that alters the strength of the relations between stressor exposure and health problems. Mediators help “explain” a causal relation.

Step 14: Generate a FACCD. The prior steps involve gathering information about the target behaviors, causal variables, and relations among them. In the final step, the information can be synthesized using vector diagrams that are analogous to structural equation modeling. As illustrated in Fig. 16.3, target behaviors and causal variables are differentiated by the shapes in the vector diagram. Importance ratings of these same variables are captured using line thickness of the shapes (low, medium, high importance). The functional relations among variables are depicted by vector lines that are either unidirectional or bidirectional. The strength of relations is depicted by vector line thickness (small, medium, large).

II.C. Negotiating the Contingency Plan

Once target behaviors and rewards are identified, a contingency plan is negotiated with the mental health professional assuming the role of an arbitrator. In this process, the mental health professional reviews the identified target behaviors and rewards with the patient only and instructs the patient to determine a fair contingency plan. The patient's initial contingency plan is brought to the significant other for approval, or modification. If changes are desired by the significant other, modifications are performed, and the revised plan is presented to the patient for approval, or modification. Contingency plans go back and forth in this manner until a plan is mutually adopted by both parties. At any time in the aforementioned negotiation process, a therapist may bring both parties together to expedite compromise. If the patient refuses to participate in the negotiation process, or if the patient is incapable of participating in the negotiation process, the significant other solely determines the contingency plan.

4.3.1 Procedure

1.

Target behaviors. Start by identifying the skill areas in which the learner has difficulty sustaining attention. Possible target behaviors include morning and evening routines (see lesson in this chapter), homework routine (see lesson in this chapter), practicing sports or musical instruments, puzzles, constructive play activities (Legos, making a model car), and projects of any kind; also see the Planning: Short- and Long-Term Goals lesson in Chapter 6, Problem Solving, Time Management, and Planning.

Baseline. For the target behaviors of interest, collect baseline data on how many minutes the learner is able to stay on task without getting distracted or engaging in challenging behavior. Use the Sustained Attention Data Sheet (Fig. 4.5) to track your data.

Target time length. In an effort to ensure the learner is successful, set the initial target time length just below the baseline time length. To determine your goal time length, keep in mind that a kindergartner and first grader should be expected to stay on task for 10 minutes, a second grader for 20 minutes, and increase by 10 minutes for each grade level. By high school, learners should be able to attend to a task for a couple hours.

Prompting.

If the learner requires prompting to get back on task during the interval, the interval is likely too long. Start with an interval that the learner can successfully carryout on his own, even if only for a few seconds.

As the learner is successful, you will begin to increase the interval across sessions. Once the learner is up to an interval that is a few minutes or longer, if you feel you must provide a prompt to get back on task, I suggest using nonvocal prompts such as a point in the direction of the task.

You might also consider setting a device prompt such as an app to beep on an interval to remind the learner to “check in” with himself to determine if he is on task. See Chapter 2, Principles Behind the Lessons, for more on device prompts.

Reinforcement.

When the learner meets the target time length without getting distracted or engaging in challenging behavior, provide immediate reinforcement.

After two to three consecutive instances of meeting the target time length, increase the time length required to stay on task by small increments initially (e.g., 1–2 minutes or even just a few seconds if needed). This is how reinforcement will be gradually thinned out.

After two to three consecutive instances of failing to meet the target time length, back-step to a previously successful time length, and then increase by a smaller increment when the learner again meets the target time length two to three consecutive times.

You may need to implement different time goals for each type of task, as some tasks will be more difficult to pay attention to than others.

Training Data Collectors

After a target behavior has been identified/defined, and the lead data collector has used decision flowchart of Kahng et al. (2011) to select a measurement system, the next step is to train observers to collect and summarize data. We recommend providing instructions to observers on the necessity of accurate data collection, but keeping the observers blind to as many goals/aspects of the study/intervention as possible (Boykin & Nelson, 1981). Next, the lead data collector should use the TTC program (Ray & Ray, 2008) to individualize observers’ training to the actual behaviors, definitions, and methods relevant to the study/intervention. The TTC program delivers empirically supported training practices to the trainees to quickly and effectively produce adequate data collection. Finally, the lead data collector should use calibration analyses of Mudford, Zeleny, Fisher, Klum, & Owen (2011) within the TTC to ensure that observers’ data are valid and precise before permitting in vivo data collection.

3.2 Self-Management

3.2.2 Procedure

1.

Identify target behavior. The first step is to identify what behavior will be targeted. You can use the Self-Awareness Worksheet (Fig. 3.1) from the Self-Awareness lesson in this chapter to identify behaviors that the learner will self-manage. It is recommended that you start with one target behavior to avoid overwhelming the learner. Once that target behavior is tackled, you can add in more target behaviors.

2.

Take baseline data. Collect data on the target behavior to identify how much it occurs during baseline.

a.

Frequency. In many cases, you will be collecting frequency of the behavior during a set observation period. Note it is important that the observation session length remain constant so that the frequency is not inadvertently inflated or deflated. Let’s say, e.g., you observe 5 instances during a 10-minute observation and 15 instances during a 30-minute observation. You may incorrectly conclude that the behavior was more frequent in the second observation session, when in fact, the behavior occurred at the same rate as the first observation session (i.e., in both observations, the learner engaged in the behavior five times every 10 minutes). If you cannot always keep the duration of the observation the same, then divide the frequency by the number of minutes in the observation, to give you a mean rate per minute.

b.

Duration. In some cases, you may also choose to record the duration of the behavior, but this is only necessary if the behavior does not have an offset that occurs soon after the onset of the behavior (e.g., more continuous behaviors such as hand flapping or exercising).

i.

Decreasing a behavior. If the target behavior is one that you want to decrease, an easy way to take duration data is to start and stop a stopwatch each time the target behavior occurs and ends, respectively, and then convert the total duration to the percentage of observation time the learner engaged in the target behavior. For example, if using a stopwatch to monitor hand flapping, you would start the stopwatch as soon as hand flapping occurs and then stop the stopwatch as soon as it ceases. You would repeat this throughout the observation period. Then, let’s say the behavior occurred for 5 minutes during a 15-minute observation. You would divide 5 by 15 and multiply by 100 to determine that hand flapping occurred during 33% of the observation session.

ii.

Increasing a behavior. Record the length of the time the learner engaged in the behavior (e.g., how long she exercised, practiced piano, etc.).

c.

Take the average. Conduct at least three baseline observation sessions and calculate the average (1) frequency of the target behavior, (2) duration, or (3) percentage of session time engaged in the target behavior.

3.

Set a goal. In an effort to choose a goal that is achievable, you will want to set a goal that is only slightly better than the baseline average. The key to teaching the learner to decrease or increase a target behavior is to set the learner up for success so that she can access reinforcement for meeting her goal. Example goals for decreasing a behavior might be to decrease from a frequency of 10 to a frequency of 8 or to decrease from 90% of the duration of an observation session to 80% of an observation session.

4.

Identify a reinforcer. Give the learner a choice for what she wants to earn when she meets the goal. Don’t assume you already know what will be an effective reinforcer.

5.

Teach self-monitoring. Teach the learner to self-record when she engages in the target behavior. This may begin by using shadowing (see Chapter 2, Principles Behind the Lessons, for more about shadowing), and then fading out your presence as the learner becomes more self-aware of when the behavior is occurring.

6.

Teach self-evaluation. Communicate the goal to the learner and teach the learner to identify if she has met the goal. When the goal has been met, you might also teach the learner to engage in self-reinforcement. This would involve allowing the learner to self-deliver the reinforcer rather than requiring the learner to check in with the interventionist to earn the reinforcer. If you choose to allow the learner to use self-reinforcement, make sure to do honesty checks. Otherwise, there’s the possibility that the system will fail because the learner may give herself reinforcers without actually meeting the goal.

7.

Make new goals and a terminal goal. As the learner meets a goal, you and the learner will need to set a new goal. The idea is to make the goal gradually more difficult than the previous goal so that eventually the learner’s target behavior meets a terminal goal. The terminal goal is the final goal that suggests the learner no longer needs improvement on the target behavior.

Consider mastery of decreasing or increasing a specific target behavior to have occurred once the learner meets the terminal goal for a predetermined time without prompting. An example mastery criterion is meeting the terminal goal for three consecutive weeks. In addition, consider the skill of self-management to be mastered once the learner is able to engage in self-management as outlined in this lesson independently and consistently with new target behaviors. An example mastery criterion is engaging in self-management of three new (untrained) target behaviors independently (without being asked). This mastery criterion ensures that the learner has developed a repertoire of self-awareness and self-management.

The learner can use the Self-Management Data Sheet (Fig. 3.2) to collect data on frequency or duration of the target behavior. He will use the Tracking section of the data sheet to tally frequency or duration and the Data Conversion section to calculate total frequency, duration, or percentages. You should also collect data and compare your data with the learner’s data in order to improve the learner’s accuracy. Graph the frequency, duration, or percentage of the target behavior across observation sessions.

Figure 3.2. Self-Management Data Sheet to be used with the Self-Management lesson.

Behavioral experiments

Once problematic target behaviors and behavioral patterns have been effectively identified, multiple behavioral interventions can be used to address these behaviors. Behavioral experiments, or activities designed to help patients’ test-out maladaptive beliefs, can be an invaluable tool with antagonistic patients. For example, the belief, “If I help someone else out, they will take advantage of me,” could be tested by doing small good deeds for strangers, followed by favors for friends, colleagues, and family members and examining how the results confirm or disconfirm this belief. Ideally, behavioral experiments should be performed in a repeated and systematic manner across multiple contexts to be most effective. Relatedly, in vivo exposure or activities can be planned or performed within or between sessions to aid the patient in confronting avoided situations or responding differently when maladaptive beliefs or schemas are activated. For instance, antagonistic individuals may be more likely to interpret ambiguous events in a hostile manner and thus react aggressively, so the patient can plan to react in a nonaggressive manner to test whether a different outcome than expected occurs. Between sessions, simple homework activities can be assigned, or depending on the nature of the behavior, activity scheduling can be used to help the patient problem-solve when and how often they will engage in planned behaviors. New behavioral patterns are most likely to be maintained if they practiced repeatedly across diverse contexts.