Basic and Applied Research on Extinction Bursts

Overview

Although highly effective, treatments like FCT include extinction, which can have adverse side effects. The extinction burst, an increase in the frequency or intensity of destructive behavior at the start of treatment, is the most common side effect of extinction, and can increase the risk of harm to the patient and others. The goal of the current study is to evaluate the prevalence of extinction bursts when various parameters of reinforcement (i.e., rate, magnitude, quality) are manipulated.

PRELIMINARY PROCEDURES Paired-Stimulus Preference Assessment As part of our standard clinical practice, the investigators will use a paired-stimulus preference assessment to evaluate each participant's preference for various leisure items. During this assessment, therapists will select an array of items informed by the Reinforcer Assessment for Individuals with Severe Disability (RAISD), which is a structured interview between the behavior analyst and the participant's caregiver to identify stimuli that may serve as reinforcers. Then, on each trial of the assessment, the therapist will place two of the items (e.g., iPad, Nintendo Switch) in front of the participant. When the participant reaches for a given item, the therapist will remove the unselected item while the data collector records which of these two items the participant approached and the participant's consumption of the selected item. After approximately 30 s, the therapist will present another pair of stimuli to the participant, with processes in place to randomize the positioning (left, right) of the stimuli and the order in which pairs are presented. This will continue until the therapist presents all potential pairs of stimuli to the participant. The data collector will then compute and graph the number of times each stimulus was selected to derive a preference hierarchy (e.g., low, moderate, high). The most selected item that the participant reliably consumes will serve as the stimulus programmed in Experiment 3's high-quality condition and the tangible reinforcer for all conditions in Experiments 1 and 2 should the participant's destructive behavior be maintained by tangible reinforcement. The least selected item with which the participant interacts will serve as the stimulus programmed in Experiment 3's low-quality condition. Competing-Stimulus Assessment As part of our standard clinical practice, the investigators will use a competing-stimulus assessment to evaluate each participant's preference for various food and leisure items relative to the functional reinforcer for destructive behavior. During this assessment, therapists will select an array of items informed by the RAISD. Then, on each 2-min trial of the competing-stimulus assessment, the therapist will place one of the stimuli (e.g., M\&Ms) in front of the participant and allow the participant to interact with or consume that item. If at any point during the trial, the participant displays destructive behavior, the therapist will deliver the functional reinforcer (e.g., attention) for 20 s. During each trial, trained observers record the duration of engagement with each competing stimulus and each occurrence of destructive behavior. Each stimulus is assessed three times (i.e., three 2-min trials) in a quasirandom order. Then, the results are tabulated and summarized in a graph showing the duration of stimulus engagement and the rate of destructive behavior for each stimulus. The highest competing stimulus is the one with the highest duration of stimulus engagement and the lowest rates of destructive behavior. Researchers will use this selected stimulus in Experiment 4 in the rate-drop/quality increase condition to determine whether delivering a higher quality reinforcer will counteract the effects of a drop in the rate of reinforcement and prevent or mitigate an extinction burst at the start of FCT. Functional Analysis As part of our standard clinical practice, the investigators will conduct a functional analysis of each participant's destructive behavior. Functional analyses help identify what consequences (e.g., access to attention) maintain destructive behavior. Prior to conducting a functional analysis, researchers routinely conduct a risk assessment to ensure that it is safe to conduct a functional analysis with each patient using the procedures developed in our program. Researchers also will conduct preference assessments with each participant to determine a preference hierarchy (e.g., of toys, foods). The investigators will use this information to individualize each condition of the functional analysis for each participant. Researchers will extend session duration when indicated (e.g., if it appears that destructive behavior begins to occur near the end of a 5-min session). The functional analysis will include at least three test conditions (social attention, demand, and monitored alone/ignore) and one control condition (play) that researchers conduct within a multielement design. Researchers will interview the caregivers prior to the functional analysis to determine the relevant stimuli (e.g., types of attention, preferred items, demands) to program within each condition. In accordance with "best clinical practice," researchers will include an additional test condition (tangible) if the caregiver reports providing, or is observed to provide, preferred tangible items following destructive behavior. For some participants who do not display destructive behavior during standard test conditions, researchers may evaluate other test conditions relevant to their case to determine idiosyncratic sources of reinforcement (e.g., social control, where adult compliance with child requests functions as reinforcement for destructive behavior). Researchers will program a uniquely colored surgical smock, worn by the therapist, for each condition to facilitate discrimination between test and control conditions. In the attention condition, the therapist will provide the participant with high-quality attention for 1 min prior to the session. Then, the therapist will withdraw attention and pretend to read a magazine while the participant has an opportunity to play with a moderately preferred toy. If the participant emits destructive behavior, the therapist will deliver 20 s of vocal (e.g., "Stop that, you'll hurt yourself") and physical (e.g., rubbing the participant's back) attention according to a fixed-ratio 1 (FR 1) schedule. In the demand condition, the therapist will deliver non-preferred demands (e.g., "Write your name") using sequential verbal, modeled, and physical prompts every 5 s. Compliance will produce praise (e.g., "Nice job writing your name!"), noncompliance will result in physical guidance (e.g., hand- over-hand prompting the participant to write his or her name) and no praise, and destructive behavior will produce a 20-s break from demands on an FR 1 schedule. In the monitored alone condition, the participant will be alone in a treatment room without any toys or materials, but a therapist will monitor the participant from behind a one-way observation window. If the participant displays aggression toward others, researchers will conduct a monitored ignore condition instead of an alone condition, during which a therapist will monitor the participant from inside the therapy room but will not interact with the participant or respond to the participant's destructive behavior. In the tangible condition, the therapist will provide the participant access to a highly preferred toy for 1 min prior to the start of the session. The therapist will then withdraw the toy at the beginning of the session and return it to the participant for 20 s following destructive behavior according to an FR 1 schedule. In the control condition (play), the therapist will provide continuous access to the participant's highly preferred toy from the tangible condition and will deliver attention every 20 s for the absence of destructive behavior (e.g., "Nice job playing with your blocks!"). Researchers will conduct at least three sessions in each condition or until researchers verify that destructive behavior is maintained by social reinforcement (e.g., access to attention or tangible items) using the ongoing visual inspection criteria developed and validated by our research team. IN-HOME BASELINE GENERALIZATION SESSIONS Following the functional analysis described above, the treatment team determines whether it is safe to conduct baseline sessions in the home prior to initiating treatment. If the participant caused physical harm to self, others, or the environment or if they required specialized equipment (e.g., protective equipment; a padded treatment room; emergency restraint) to prevent such harm during the functional analysis, then researchers do not conduct baseline sessions in the home, but still collect post-treatment generalization sessions in the home (described below). If it is safe to conduct baseline sessions in the home, researchers use the following procedures. Baseline Sessions Researchers collect a series of three 5-min baseline sessions in the test condition from the functional analysis with the highest rates of destructive behavior. Prior to the first session, the therapist inspects the room in the home where the sessions will take place and removes any objects that could be dangerous to the participant or the caregivers (e.g., an object that could be used as a weapon). Next, researchers use behavioral skills training to teach the caregiver how to implement the session procedures using verbal instruction, modeling, behavioral rehearsal (i.e., role-play with a therapist), and performance feedback. Researchers train each caregiver to a mastery criterion (i.e., at least 90% accuracy across all components when implementing practice sessions with a therapist) prior to conducting baseline sessions with the participant. Additionally, prior to conducting session in the home, researchers train caregivers in our managing-challenging-behavior techniques (e.g., blocking hits and kicks, releasing bites) until the caregivers implement all techniques with 90% accuracy during role-play. If at any point the participant displays destructive behavior that poses an imminent danger to self, others, or the environment that cannot be safely blocked (e.g., eye poking, biting), the therapist terminates the session. Data Collection During generalization sessions, our research team will monitor the sessions inconspicuously (e.g., collecting directo bservation data from a nearby hallway using specialized software; routinely checking inter-observer agreement). General Instructions to Caregiver "We are asking for your assistance in helping us to understand and observe your child's destructive behavior. We would appreciate you help in working to replicate the challenging situations that you discussed with your primary therapist. Our goal in conducting these sessions is to help us to identify things and may trigger and reinforce your child's problem behavior in the home. The session that we would like you to conduct will be 5 minutes in length. For each 5- minute session, we would like you to follow the procedures described below as closely as possible. After you read the instructions, we will answer any questions that you have before we start the session. If at any point, you feel that you or your child are unsafe, please let a therapist know, and we will immediately assist in keeping you and your child safe. It is our goal to maintain safety throughout the assessment and treatment of your child's destructive behavior." Condition-Specific Instructions to Caregiver The therapist will set up the materials and arrange the room and select the specific instructions to give to the caregiver from the options below based on the results of the functional analysis (e.g., for escape-reinforced destructive behavior, they will set up for a demand condition and use the instructions below labeled "Demand"). The therapist will give the caregiver a copy of the selected instructions described below, have them read it, and then answer any questions they have. Demand. "In this session, we want you to instruct your child to complete a variety of nonpreferred tasks from the list of such tasks that you and your child's primary therapist specified during the parent interview. The session will last 5 minutes. You may use any materials in the room that you would like to complete the instructions during this session. Please do exactly what you normally do at home to get your child to complete such tasks. We are interested in observing how you give instructions and how your child responds to those instructions. When the 5-minute session is over, a therapist will give you a brief break. Again, if at any point you feel that you or your child are not safe, please let a therapist know, and we will immediately assist in keeping you and your child safe." Attention. "In this session, we want to observe you and your child in a situation in which you must manage your child's behavior while you also complete a task. In this case, the task will be to complete this questionnaire using the pencil and clipboard we have provided to you. While you complete the task, please interact with your child as you normally would. After one minute, you will be signaled by a therapist to tell your child that you have some work to do and that they should play quietly. After you tell your child this, please move away from your child and work on the task. Please react to your child's behaviors as you normally would but do your best to complete the task we have placed on the clipboard throughout the 5-minute session. If you complete the task, please review it, and continue to "work on it" until the session is complete. When the 5-minute session is over, a therapist will give you a brief break. Again, if at any point you feel that you or your child are not safe, please let a therapist know, and we will immediately assist in keeping you and your child safe." Tangible. "In this session, please give your child this toy that we have identified as one of your child's most preferred items. After one minute, you will be signaled by a therapist to take the toy away from your child. You should play with the toy/item for the remainder of the session. If your child displays destructive behavior, please react to it as you normally would. When the 5-minute session is over, a therapist will give you a brief break. Again, if at any point you feel that you or your child are not safe, please let a therapist know, and we will immediately assist in keeping you and your child safe." GENERAL EXPERIMENTAL PROCEDURES Selecting the Functional Communication Response (FCR) During functional communication training (FCT), researchers often teach the child an FCR modality that therapists can physically guide, such as a card exchange or card touch, because it is critically important to minimize exposure to the establishing operation for destructive behavior during the early stages of FCT (i.e., limiting the time the child does not have access to the functional reinforcer). Typically, researchers create a laminated index card with a photograph of the child consuming the reinforcer (e.g., the child playing with an iPad) and then teach the child to either exchange or touch the FCR card to gain access to the reinforcer maintaining destructive behavior. Caregiver input (e.g., preference for a vocal response) and participant-specific factors (e.g., low vision) are also considered when selecting the FCR for each patient. Experimental Design Researchers will evaluate the effects of functional communication training on destructive behavior using an ABAB (baseline, FCT, baseline, FCT) reversal design. Researchers will conduct within-participant comparisons of FCT conditions using an embedded alternating-treatments design. All sessions will last 5 min. Researchers will conduct Experiment 3 with participants whose destructive behavior is reinforced by access to tangible items so that researchers can vary reinforcement quality based on the paired-stimulus preference assessment. Data Collection and Computation of Interobserver Agreement and Procedural Integrity Trained observers will collect data on the frequency of participant destructive behavior and therapist (or caregiver) implementation of the assessment and treatment protocols to assess procedural integrity using BDataPro® software developed in our lab. A second observer will score at least one third of sessions independently to assess data accuracy (reliability). The second data collector will be blind to the project's research questions and hypotheses for at least 1/6th (17%) of sessions. For at least one third of sessions, observers will collect procedural-integrity data to ensure that the assessment and treatment protocols are implemented as planned. That is, researchers will collect data on whether therapists correctly implemented the planned antecedents, prompts, and consequences for each target response. Researchers will then transform the data into a percentage-correct measure by dividing the number of correct therapist responses by the number of opportunities for a correct response. Researchers will retrain therapists who show less than 90% implementation accuracy for two consecutive sessions. The reliability of direct-observation measures is typically established through measurement of interobserver agreement. To calculate inter-observer agreement, sessions will be partitioned into successive, 10-s intervals (e.g., Seconds 0-9, 10-19, 20-29). In each 10-s interval, researchers will determine whether the observers agreed or disagreed on the frequency of each target behavior. An exact agreement will be defined as both observers recording the same frequency of a target behavior in each 10-s interval. Researchers will then calculate the percentage of exact agreements per session. Interobserver agreement in our program averages above 90%, and observers undergo retraining if agreement levels fall below 80% for two consecutive sessions. EXPERIMENT-SPECIFIC PROCEDURES Experiment 1: Effect of Reinforcement Rate Drop Baseline. Researchers will conduct three identical baselines each in separate and distinct contexts differing in the color of the room and the therapists' clothing (e.g., blue versus yellow versus brown rooms and therapist clothing). Each baseline will be identical to the condition from the functional analysis with the highest rates of destructive behavior, except for the following modifications. Researchers will program an independent, dense variable-interval (VI) 1.5-s schedule of reinforcement for destructive behavior in each baseline and deliver the reinforcer for 20 s each time to make the baselines highly like typical clinical baselines. Researchers will use independent VI schedules rather than an FR 1 schedule to better control and equate the obtained rate of reinforcement in each condition. The investigators have used VI schedules for this purpose to good effect in multiple prior studies on resurgence of destructive behavior, the investigators have used both VI schedules and FR 1 schedules and produced comparable results. Researchers will equate reinforcement magnitude (i.e., each reinforcer delivery lasting 20 s) and quality (i.e., the functional reinforcer identified during the functional analysis) across the baseline conditions. Stability criteria. Researchers will conduct at least five baseline sessions in each context until the standard deviation of responding in the last five sessions of each condition is no more than 50% of the mean of those sessions. Treatment. The reinforcement magnitude and quality will be identical to baseline. Researchers will randomly assign the three treatments (extinction-only, rate-drop, and rate-hold conditions) to the color-correlated baseline contexts. In the extinction-only condition, researchers will discontinue reinforcement for destructive behavior, but not deliver alternative reinforcement for the FCR. In the rate-drop condition, researchers will place destructive behavior on extinction and deliver the functional reinforcer for the FCR according to a VI 15-s reinforcement schedule (i.e., a robust reduction in reinforcement rate relative to baseline). In the rate-hold condition, researchers will place destructive behavior on extinction and deliver the functional reinforcer for the FCR according to a yoked VI 1.5-s schedule that matches the exact rate and timing of reinforcer deliveries for destructive behavior in baseline. To teach the participant the FCR in the rate-drop and rate-hold conditions, researchers will embed a constant prompt delay in each FCT condition. Each time the programmed VI schedule elapses, researchers will implement a 5-s prompt delay wherein researchers allow up to 5 s for the participant to emit the FCR independently prior to prompting the response (e.g., physically guiding the participant to exchange the FCR card). To minimize adventitious reinforcement of destructive behavior in the FCT conditions, researchers will incorporate a 3-s changeover delay by withholding reinforcement for FCRs until the response occurs without destructive behavior occurring within 3 s of the FCR. Researchers will continue treatment sessions until researchers observe an 85% reduction in destructive behavior relative to baseline across all three conditions for two consecutive sessions. Experiment 2: Effect of Reinforcement Magnitude Drop Baseline. Baseline procedures will be identical to Ex 1 except researchers will program three times the magnitude of reinforcement. That is, researchers will program 60 s of reinforcement, instead of 20 s, when researchers deliver the functional reinforcer in each baseline context. Researchers will equate reinforcement quality (i.e., the functional reinforcer identified during the functional analysis) and rate (i.e., VI 1.5-s schedule) across the baseline conditions. Researchers will use the stability criteria described above for Ex 1a to determine when to end baseline and initiate treatment. Treatment. The reinforcement rate and quality will be identical to baseline. Researchers will randomly assign the three treatments (extinction-only, magnitude-drop, and magnitude-hold conditions) to the color-correlated baseline contexts. In the extinction-only condition, researchers will discontinue reinforcement for destructive behavior, but not deliver alternative reinforcement for the FCR. In the magnitude-drop condition, researchers will place destructive behavior on extinction and deliver the functional reinforcer for the FCR for 6 s (i.e., a large reduction in reinforcement magnitude relative to baseline). In the magnitude-hold condition, researchers will place destructive behavior on extinction and deliver the functional reinforcer for the FCR for the same duration as in baseline (i.e., 60 s). To teach the participant the FCR, researchers will incorporate the prompt-delay and changeover-delay procedures as described in Ex 1a's FCT conditions. Researchers will continue treatment sessions until researchers observe an 85% reduction in destructive behavior relative to baseline across all three conditions for two consecutive sessions. Experiment 3: Effect of Reinforcement Quality Drop Participant Selection. Researchers will conduct Ex 3 with participants who display destructive behavior reinforced by access to tangible items so that researchers can vary reinforcement quality using the results of a paired-stimulus preference assessment. Baseline. Baseline procedures will be identical to Ex 1, except that researchers will conduct two identical baselines in separate and distinct contexts (rather than three). Researchers will use the stability criteria described above for Ex 1a to determine when to end baseline and initiate treatment. Treatment. The reinforcement rate and magnitude will be identical to baseline. Researchers will randomly assign the quality-drop and quality-hold conditions to the color-correlated baseline contexts. In the quality-drop condition, researchers will place destructive behavior on extinction and deliver the reinforcer from the paired-choice assessment that the participant chooses approximately 1/12th as often as the highest preferred stimulus from that assessment contingent on the FCR (i.e., which will produce a large reduction in reinforcement quality relative to baseline). In the quality-hold condition, researchers will place destructive behavior on extinction and deliver the same reinforcer from baseline (i.e., the highest quality reinforcer from the paired-stimulus preference assessment) for the FCR. To teach the participant the FCR, researchers will incorporate the prompt-delay and changeover-delay procedures as described in Ex 1's FCT condition. Researchers will continue treatment sessions until researchers observe an 85% reduction in destructive behavior relative to baseline across both conditions for two consecutive sessions. Experiment 4: Clinical Study on Counteracting Reinforcement-Rate Drop with Quality Increase. Baseline. Baseline procedures will be identical to Ex 1, except that researchers will conduct two identical baselines in separate and distinct contexts (rather than three). Researchers will use the stability criteria described above for Ex 1 to determine when to end baseline and initiate treatment. Treatment. The reinforcement magnitude will be identical to baseline. The investigators will randomly assign the rate-drop-only and rate-drop/quality-increase conditions to the color-correlated baseline contexts. In the rate-drop-only condition, researchers will place destructive behavior on extinction and deliver the functional reinforcer for the FCR according to a VI 15-s reinforcement schedule (i.e., a robust reduction in reinforcement rate relative to baseline). In the rate-drop/quality-increase condition, researchers will program the same large drop in reinforcement rate by delivering reinforcement on a VI 15-s schedule for the FCR, but also will increase reinforcement quality by simultaneously delivering the highest quality reinforcer identified during the competing-stimulus assessment described above. To teach the participant the FCR, researchers will incorporate the prompt-delay and changeover-delay procedures as described in Ex 1's FCT condition. Researchers will continue treatment sessions until an 85% decrease in destructive behavior is observed relative to baseline across both conditions for two consecutive sessions. Post-Study Procedures Researchers will continue to provide relevant treatment services following the conclusion of the patient's participation in the experiment. After initial ABAB designs with dense reinforcement schedules for FCRs (like those in Experiments 1, 2, 3, and 4), researchers progress treatment with systematic evaluations of reinforcement-schedule thinning. Over the course of several weeks, researchers incorporate discriminative stimuli (e.g., multiple or chained schedules) and progressively thin reinforcement to a practical reinforcement schedule selected by the caregiver. During the last two weeks of treatment, researchers transfer treatment to relevant stakeholders (e.g., caregivers, teachers) and contexts (e.g., home, school, community). IN-HOME POST-TREATMENT GENERALIZATION SESSIONS After the post-study procedures described above, caregivers will implement treatment sessions in their home to assess for generalization of treatment effects. Researchers will use behavioral skills training to teach the caregiver how to implement the treatment procedures using verbal instruction, modeling, behavioral rehearsal (i.e., role-play with a therapist), and performance feedback. Researchers will train each caregiver to a mastery criterion (i.e., at least 90% accuracy across all components when implementing practice sessions with a therapist) prior to conducting treatment sessions with the participant. In addition, researchers have the caregiver implement treatment sessions under close supervision in our clinic setting before having them implement the treatment in the home setting. During generalization sessions in the home, our research team will monitor the sessions inconspicuously (e.g., collecting data from a nearby hallway) and provide the caregiver with performance feedback as needed. After the participants meet their behavioral goals in the clinic, caregivers will implement the terminal mult-FCT procedures within their homes during three 5-min sessions under conditions that parallel the in-home baseline generalization condition described above (e.g., If the patient displayed the highest rates of destructive behavior in the demand condition of the functional analysis, the treatment generalization sessions will also occur in a demand context). Researchers will use data from the generalization sessions to evaluate how our final FCT treatment works for caregivers in a natural setting. Researchers will terminate any session if destructive behavior increases to unsafe levels for the caregivers or participants. Researchers will provide additional caregiver training, if necessary, until participants exhibit low levels of destructive behavior with their caregivers, as is typical of our routine clinical practice.