Printer Friendly

Effects of Function-Based Crisis Intervention on the Severe Challenging Behavior of Students with Autism.


A percentage of individuals with autism and other developmental disabilities engage in severe, dangerous challenging behaviors. Two common approaches for intervening for these behaviors are function-based interventions (FBI), which have been effective at reducing such behaviors, and crisis intervention, which has been shown to increase staff skills for managing dangerous situations. These two methods have different theoretical foundations, and as such the two approaches often recommend competing strategies for a given challenging behavior. Therefore, practitioners may feel they need to choose between the two approaches, selecting either FBI or crisis intervention. This study examined a way to blend the strengths of two approaches into a synthesized model referred to as function-based crisis intervention (FBCI). Using a delayed multiple-probe design, results showed that FBCI reduced the severe challenging behavior of three students with autism. Implications for future research and practice are discussed.

Keywords: functional behavior assessment, function-based intervention, crisis intervention


Estimates suggest between 10 to 15% of students with disabilities engage in challenging behavior (Emerson et al., 2001). A subset of this group engages in challenging behavior that can be dangerous to themselves and others. This includes 7% who engage in aggression, 4% who engage in self-injurious behavior (SIB), and 4-5% who engage in destructive behavior (Emerson et al., 2001). Students with autism and intellectual disabilities are at even greater risk of exhibiting severe challenging behavior such as aggression and SIB (Blacher & McIntyre, 2006).

The clearest risk these behaviors pose is the threat of injury to the students themselves and those around them. In addition to the safety threat, these behaviors pose a number of other risks. This includes increased likelihood of experiencing intrusive and ineffective strategies (e.g., punishment, exclusion; Scott et al., 2005); receiving medications with severe side effects (Powers, 2005); and increased academic failure, disproportionate rates of suspension, and poor social supports (Chandler & Dahlquist, 2006). The negative effects also extend to caregivers, who may experience a lack of confidence in their ability to work with these individuals (Allen & Tynan, 2000), increased levels of stress (Jenkins, Rose, & Lovell, 1997), and increased negative interactions (Lawson & O'Brien, 1994).

Two commonly used approaches to intervene with these behaviors are function-based intervention (FBI) and crisis intervention, sometimes referred to as reactive management. FBI consists of a functional behavior assessment (FBA) and ensuing interventions based on the results of that assessment (Carr, Langon, & Yarbrough, 1999). FBIs are generally attempted to decrease the challenging behavior by (a) minimizing evocative and maintaining variables, and (b) increasing functionally equivalent behavior(s) to replace the challenging behavior with a more socially appropriate behavior. FBI is often viewed as the preferred approach for addressing severe challenging behavior for individuals with autism and developmental disabilities for two primary reasons. First, decades of empirical research support FBI's effectiveness at reducing severe challenging behaviors for this population (Beavers, Iwata, & Lerman, 2013). Second, FBI has been shown to be more effective than non-function-based interventions at reducing challenging behaviors (Ingram, Lewis-Palmer, & Sugai, 2005).

With that being said, there are limitations to the FBI literature that warrant consideration. For instance, some function-based strategies may be difficult to apply for dangerous behaviors. For example, it may not be possible to implement extinction for attention maintained self-injury because response blocking is necessary to maintain safety. This can be problematic because research has shown that at times extinction is necessary for behavior reduction (Hagopian, Fisher, Sullivan, Acquisto, & LeBlanc, 1998).

Second, the complexity of the FBI research, which focuses on a host of different types of intervention (e.g., antecedent intervention, consequence interventions, motivating operation interventions, interventions targeting additional behaviors such as replacement and precursor behaviors), can present a barrier to implementation for caregivers not trained in behavior analysis. This is supported by research on practitioner fidelity, which shows caregivers struggle to properly conduct FBA and/or implement FBI (Blood & Neel, 2007; Van Acker, Boreson, Gable, & Potterton, 2004). This was the case for Bethune and Wood (2013), when three out of four teachers were unable to properly implement a basic FBI after a full-day workshop. The teachers ultimately required in-vivo coaching to implement the strategy.

Another reason for the poor procedural fidelity could be a lack of cultural fit. In discussing the failure to properly implement FBI, Scott, Nelson, and Zabala (2003) argue the process may simply be unfamiliar and uncomfortable for some practitioners. This suggests that a more resonant model may facilitate broader use of FBI.

Last, Hanley, Jin, Vanselow, and Hanratty (2014) noted the literature on FBI for severe challenging behavior is almost exclusively focused on the effects of isolated components of assessment and/or intervention, with relatively little research on the kinds of comprehensive treatment models necessary to be "recognized as important by colleagues and constituents who are not behavior analysts" (Hanley et al., 2014, p. 31). Consequently, there is a paucity of resources on the comprehensive, synthesized models for intervention that are required for FBI to be broadly adopted in applied settings. Therefore, despite the effectiveness of FBI, it may be unrealistic to expect practitioners without extensive training in behavior analysis (e.g., special education teachers, paraprofessionals) to be able to implement these strategies with fidelity, even though in practice they are regularly tasked with serving these individuals.

The second approach used to intervene with students with disabilities exhibiting dangerous challenging behavior is crisis intervention. Crisis intervention strategies organize their interventions according to the intensity of an individual's challenging behavior, dividing a student's escalation to and recovery from dangerous behaviors into distinct phases and developing interventions for each phase. Interventions range from attempts to de-escalate the situation using verbal and non-verbal methods, to physical intervention such as self-defense, seclusion, and restraint (Allen, 2000). Many prevalent intervention frameworks are organized in this fashion including nonviolent crisis intervention (Ryan, Peterson, Tetreault, & Hagen, 2007) and the Cycle of Acting-Out Behavior (Colvin & Scott, 2015). Their prevalence, with the Crisis Prevention Institute boasting having trained over 10 million human service professionals in nonviolent crisis intervention (Crisis Prevention Institute, n.d.) and the Cycle of Acting-Out being a model distributed through prominent formats including a bestselling textbook and Vanderbilt's Iris Modules, indicates the models resonate with the general public.

Similar to FBI, crisis intervention has strengths and limitations when applied to students with disabilities who exhibit severe challenging behavior. One of the strengths is that training in crisis intervention has led to an increase in safety for those involved (Williams, 2009). Additionally, training has also led to increased confidence among caregivers working with these students (Baker & Bissmire, 2000; Dawson, 2003; Soenen, Goethals, Spriet, D'Oosterlinck, & Broekaert, 2009). Last, crisis intervention models meet the suggestions that experts in the field of behavior analysis have been making to facilitate the adoption of behavior analytic strategies such as being a "cohesive, easily understandable framework" (Smith, 2016, p. 79) and using non-technical descriptions for intervention methods (Becirevic, Critchfield, & Reed, 2016).

However, the limitations of the research on crisis intervention strategies are significant. Primary among these limitations is that few studies have examined the effectiveness of crisis intervention techniques on the rates of challenging behavior for individuals with developmental disabilities. Although those few studies have reported some general improvement, effects were not found to be statistically significant (Allen, McDonald, Dunn, & Doyle, 1997; Baker & Bissmire, 2000). Also, these studies had substantial methodological limitations such as lack of a control group and indirect measures of behavior. Thus, the effectiveness of crisis intervention strategies leading to meaningful reduction in challenging behaviors is unclear.

In addition to the lack of research on effectiveness, there is reason to believe these models would be ineffective for individuals with developmental disabilities. For instance, crisis intervention techniques often rely heavily on language, requiring activities such as debriefing or discussing alternative options (Colvin & Scott, 2015). Students with developmental disabilities may have difficulty with these strategies given the higher rates of language delays among this population. Additionally, some of the most common strategies are not conceptually systematic with behavior analysis. For instance, a common intervention is to provide calming activities when a person is agitated, which could reinforce a behavior chain that leads to the dangerous challenging behaviors. Another example is the recommendation to focus purely on safety when dangerous behaviors are exhibited, with no discussion of function. This may be effective for some, but as stated previously, it has been shown that extinction is necessary in some cases (Hagopian et al., 1998). Therefore, to not consider function at all when responding to dangerous behaviors may serve to differentially reinforce intense behaviors.

With these two models of intervention existing concurrently and recommending conflicting interventions at times (e.g., in response to dangerous behaviors), practitioners may feel a need to choose between the two. However, this may be a false choice as it may be possible to combine the two by embedding FBI throughout a crisis intervention framework to create a comprehensive, function-based treatment package in a resonant, easy to understand framework (i.e., the intensity of challenging behaviors). The Cycle of Acting-Out Behavior (Colvin & Scott, 2015) presents a promising framework for this task as it claims to be rooted in applied behavior analysis. Colvin and Scott describe seven phases to the cycle, with recommended interventions for each phase. The phases are (a) calm, (b) triggers, (c) agitation, (d) acceleration, (e) peak, (f) de-escalation, and (g) recovery. For a description of the phases and intervention foci as described by Colvin and Scott see Table 1.

However, Colvin and Scott's (2015) model has many of the limitations for application to individuals with developmental disabilities described above (e.g., a lack of research, the effectiveness of the model, language intensive strategies, all interventions not conceptually systematic). These issues could be addressed by using the overall framework of the phases, but prescribing appropriate FBI for every phase (e.g., motivating operation interventions during the calm phase, precursor interventions during acceleration, consequence interventions during peak). Therefore, the purpose of this study was to investigate the effects of a treatment package that embeds FBI throughout Cycle of Acting-Out, called function-based crisis intervention (FBCI), on the rates of severe challenging behavior for students with autism.



Inclusion criteria for the study were that participants must: (a) be students in a public school; (b) be diagnosed with a developmental disability; and (c) exhibit behavior(s) that required "immediate attention to protect the physical and/or psychological safety of that student, the teacher, or others" (Gilliam, 1993, p. 224).

This led to three students (pseudonyms used throughout) being selected. Kyon was an 18-year-old African-American male with autism who engaged in multiple forms of aggressive behavior. Pedro was a 16-year-old Hispanic male who was diagnosed with autism, severe intellectual disability, hearing impairment, visual impairment, and a seizure disorder. Pedro engaged in SIB, aggression, and property destruction. Daniel was a 16-year-old African-American male with autism. He engaged in property destruction and SIB.


This study took place in a separate public school for students with severe disabilities located in a large, urban district in the southeastern United States. For two of the three students (i.e., Kyon, Pedro) sessions took place in a 10-foot by 20-foot conference room. Due to concerns about his property destruction, sessions for Daniel took place in a dedicated section of his classroom that had cinder block walls. The section was approximately 20 feet by 20 feet, and was separated from the rest of the classroom by four-foot by six-foot rectangular tables to ensure other students did not interfere in the sessions.

Experimenter and Interventionists

The experimenter and primary interventionist (first author) was a doctoral student in special education, a Board Certified Behavior Analyst, certified in nonviolent crisis intervention by the Crisis Prevention Institute, and had 11 years of experience working with students with developmental disabilities and challenging behavior. Employees of the school district served as secondary interventionists. They included a district-level support professional, the students' teacher, and teacher assistants. At least one secondary interventionist was present for each session to support the interventionist when necessary to maintain safety (e.g., implementing a two-person restraint).


This study used a delayed multiple-probe across participants design adapted from the multiple-probe design and the delayed multiple-baseline design (Cooper, Heron, & Heward, 2007). The delay was added due to ethical and safety concerns as it minimized the number of sessions in which students and interventionists participated during baseline when effective intervention was absent.

The order students entered baseline was determined based on recommendation by the teacher for which student had the greatest need. The second and third student entered baseline prior to the preceding student beginning intervention. This ensured a degree of internal validity as the components of baseline logic (i.e., prediction, verification, replication) were able to be demonstrated (Cooper et al., 2007). Namely, the prediction from Kyon's baseline would be verified by the overlap with Pedro's baseline, the prediction from Pedro's baseline would be verified via the overlap with Daniel's baseline, and replication of the effect could be seen with the delivery of the intervention.

Dependent Variable and Data Collection

The dependent variable was challenging behavior. Each student had an individualized definition to reflect their unique topographies. Kyon's challenging behavior was defined as anytime he hit, scratched, bit, pushed, pulled, ran away from, or actively resisted the prompting of another person. Pedro's challenging behavior was defined as anytime he (a) bit himself or pulled his hair; (b) hit, kicked, head butted, bit, or pushed another person; (c) threw, knocked over, shook, or hit property; or (d) whined, yelled, grunted, or spit. Daniel's challenging behavior was anytime he (a) hit himself; (b) threw, knocked over, kicked, shook, or hit an inanimate object; or (c) urinated.

Data were taken using partial interval recording. Sessions were 10 min in duration, divided into 10 s intervals. This resulted in 60 intervals per session. All sessions were video recorded and scored at a later time (Cooper et al., 2007).


The study consisted of the following phases: (a) FBA, (b) baseline, (c) FBCI, (d) maintenance, and (e) generalization. If the student exhibited challenging behaviors at the end of the 10 min session, the intervention continued until the student had not exhibited challenging behavior for 30 s.

FBA. First, the experimenter conducted an FBA for each student to create operational definitions for challenging behavior, identify precursor behaviors, and identify potentially controlling variables for the students' challenging behaviors. This included initial functional assessment interviews conducted with the teacher using the template provided by Hanley (2012). Next, functional assessment observations were conducted during activities reported as likely to evoke challenging behaviors. These two methods did not produce conclusive functions, so brief functional analyses were conducted for each participant as described by Northup et al. (1991).

Functional analysis. Brief functional analyses consisted of four conditions: contingent escape, contingent attention, contingent access to tangible, and control. The alone condition was not conducted because the setting did not have a location where a student could be observed without seeing the experimenter, and it was not permitted by the participating school district to leave a student unattended. Results indicated that Kyon's challenging behavior was maintained by access to tangibles. Pedro and Daniel's challenging behaviors were multiply controlled, maintained by access to specific tangibles (primary) and access to staff attention (secondary).

Baseline. Baseline sessions created an establishing operation for the challenging behaviors by replicating naturally occurring situations that matched the identified function for a student's challenging behavior. For Kyon this meant transitioning him from coloring to a work activity (e.g., selecting the correct coin when the name or value was verbally stated). For Pedro this involved having him wait at a table with a neutral activity while the experimenter and secondary interventionist worked on their computers. For Daniel, baseline involved prompting him to engage in an independent task (i.e., table wiping) that was incompatible with playing with his toys.

Interventions matched the strategies used by the teacher, a licensed special educator who obtained her board certification in behavior analysis during the study. She reported she attempted to use FBI including environmental enrichment, reinforcing alternative behaviors using a token economy, demand fading, and extinction. However, when the behaviors became dangerous she would abandon the FBI to maintain safety. Therefore, when students began exhibiting precursor and low intensity challenging behavior, they were maintained at the current activity and given prompts for how they could earn the putative reinforcers. However, when behavior became severe students were allowed to cease the current activity, have free access to the room, and physical intervention was used as needed to maintain safety.

FBCI. FBCI sessions replicated baseline sessions, except the students' FBCI plans were implemented. To create FBCI plans, each phase of the Cycle of Acting-Out was given an operational definition. General definitions for each phase were created (e.g., peak refers to when a student exhibits dangerous behaviors), and then student specific definitions were created for each phase (e.g., for Kyon peak was anytime he hit, scratched, etc.). Next, interventions were created for each phase. Again, first general interventions were created for each phase (e.g., for acceleration prompt replacement behavior using a controlling prompt), and then student specific applications were designed (e.g., when Kyon is agitated, physically prompt him to exchange his communication card). Last, schedule thinning protocols for replacement behaviors were designed for each student including reinforcement schedule steps (e.g., fixed interval 5 s, fixed interval 10 s, etc.) and criteria to move between the schedule thinning steps (e.g., move to the next step after two sessions with no challenging behavior). For a list of the modified definitions and specified interventions for each phase used in FBCI see Table 2.

Replacement behavior and schedule thinning. Functional communication training was targeted for Kyon. Functional communication was defined as anytime Kyon placed the "toys" picture card in the experimenter's hand. Beginning with CRF the schedule was thinned to 45 s using delay to reinforcement (Hagopian, Boelter, & Jarmolowicz, 2011). During the delay, Kyon was asked to perform tasks ranging from initially simple (e.g., clap hands) to more difficult (e.g., IEP goals).

Leisure was targeted for Pedro. Communication was not targeted because he was able to communicate his wants and needs, and his teacher reported that the challenging times involved Pedro having to wait because his wants were unavailable. Three preferred leisure activities were selected to serve as alternative reinforcement (Rooker, Jessel, Kurtz, & Hagopian, 2013). Because Pedro's challenging behavior was maintained by access to tangibles and attention, schedule thinning occurred in two phases. First, continuous attention was provided as he engaged in the leisure activity. Upon engaging in the activity without challenging behavior for the targeted interval, he was allowed to access his desired items. Once he engaged in leisure activities for 1 min without engaging in challenging behavior, the reinforcement schedule for attention was thinned.

The replacement behavior targeted for Daniel was functional communication. He was provided with an iPad using the GoTalk app. Because Daniel's challenging behavior was controlled by access to tangibles and attention, there were two buttons available: one to request his toys and one to request attention from the experimenter. A chained schedule of reinforcement was used (Hagopian et al., 2011) with a timer serving as a stimulus delta for when reinforcement for functional communication was unavailable.

Calm. Calm referred to when the student was not exhibiting any precursor or severe challenging behaviors. This phase included strategies to create abolishing operations (i.e., reduce the effectiveness) for the reinforcer. For all three students this entailed ensuring they had access to attention and/or preferred items for a minimum of 1 min prior to placing any demands.

Triggers. Triggers referred to environmental variables known to evoke, or trigger, the severe challenging behavior(s). For Kyon, the identified trigger was denying access to his toys; for Pedro and Daniel the triggers were denying access to preferred toys while limiting staff attention. Strategies in this phase were designed to prevent challenging behavior by making triggering variables less aversive. All three participants were provided with 1 min and 30 s warnings prior to delivering the trigger, and their replacement behavior was verbally reviewed (e.g., saying "Remember, if you want your toys, you can hand me your card"). Pedro was also provided with a 10 s countdown prior to removing his preferred items, given a visual schedule (i.e., first demand, then preferred activity), allowed to choose the reinforcer, and shown a timer for when reinforcement was available. Daniel also received a timer during schedule thinning which served as stimulus delta, signifying when reinforcement for functional communication was not available.

Agitation. In this phase students engaged in behaviors indicating an establishing operation was in effect. Kyon's agitation was defined as anytime he attempted to access items after he had been denied access to them. Pedro's agitation was anytime he gestured towards an item, approached a person not working with him, or resisted engaging in the leisure activity. Daniel's agitation was anytime he engaged in low intensity versions of his precursor behavior (i.e., tapping himself any place other than his head, moving his toys back and forth lightly).

Strategies in the agitation phase focused on calming the students by indicating when reinforcement was available, prompting the replacement behavior using least-to-most prompting, and assuming a safe body position. For Kyon this meant (a) standing just behind his shoulder; (b) blocking access to the desired item; and (c) escalating through verbal, gesture, and physical prompts to hand the experimenter the communication card. During CRF, toys were immediately provided. During delay-to-reinforcement, Kyon was provided with a timer and directed to wait.

When Pedro exhibited agitation, the experimenter stood behind his shoulder, paused the timer, and provided a verbal prompt to be "calm" to earn his reinforcer. Upon ceasing agitated behavior for 5 s, verbal praise was provided and the timer resumed.

When Daniel exhibited agitated behavior the experimenter stood sideways to Daniel and between Daniel and the windows. The experimenter also blocked Daniel's access to preferred items. During CRF least-to-most prompting was used to ensure he requested the reinforcer. When reinforcement was unavailabe due to schedule thinning, Daniel was redirected to his timer with a reminder that he could have his reinforcer after the timer expired.

Acceleration. Acceleration referred to the final phase before the severe challenging behaviors were exhibited. It was defined as students exhibiting precursor behaviors, indicating the severe challenging behavior was imminent (Fritz, Iwata, Hammond, & Bloom, 2013). Strategies during this phase were intended to prevent the severe challenging behavior by ensuring the student emited the replacement behavior when appropriate (e.g., physical prompts during CRF), clarifying contingencies if it was not appropriate to prompt the replacement behavior (e.g., stating "wait calmly, then toys"), and delaying access to reinforcement until the student was absent of precursor behavior.

If Kyon exhibited precursors, the experimenter remained behind Kyon's shoulder and blocked access to preferred items. The experimenter then either (a) physically prompted Kyon to use his communication card, or, if he had already requested toys and was required to wait, (b) paused his timer and redirected him to it stating "you need to wait." Upon ceasing to exhibit precursor behaviors for 5 s, the timer was resumed and verbal praise was provided.

Upon Pedro exhibiting precursor behaviors the experimenter's body position behind Pedro's shoulder was maintained. If Pedro was engaging in the target leisure activity, his timer was paused and he was told "you need to be calm." If he had access to his preferred items, the experimenter blocked access by placing his hand in front of the items and provided the same verbal reminder. Upon ceasing to exhibit precursor behavior for 5 s, the experimenter provided verbal praise and resumed the timer or allowed Pedro to resume playing with the items.

During Daniel's acceleration the experimenter maintained body position between Daniel and the window, blocked access to toys, and averted his gaze. If reinforcement was available, Daniel was physically prompted to request reinforcement. If reinforcement was not available (i.e., extinction component of his chained schedule), Daniel's timer was paused and he was redirected to his timer with a verbal reminder that he needs to work first. Upon ceasing precursor behavior for 5 s, Daniel's timer resumed and verbal praise was provided.

Peak. Peak referred to the phase when the student exhibited severe challenging behaivor (e.g., self-injury, aggression). Strategies were designed to keep students safe while still implementing extinction for the challenging behavior. If extinction was not possible, then attempts were made to diminish the quality of reinforcement provided. Nonviolent physical intervention was necessary to maintain safety at times. Prior to implementation, all interventions were approved by the research team, school staff, and parents/guardians of the students.

Upon Kyon exhibiting peak behaviors he was escorted to a table, using a CPI two-person escort if necessary, and seated in a chair with arms on the side. Once at the table, the experimenter stood sideways with his hip against the back of the chair to prevent Kyon from standing up, but also ensured space remained between Kyon and the table so no pressure was exerted on him. The experimenter then leaned away out of Kyon's reach. If Kyon was waiting, the timer was stopped. This continued until Kyon did not exhibit aggression for 30 s.

When Pedro was in the peak phase, he was escorted to his seat in the same manner as Kyon. The experimenter and secondary interventionist stood on either side of Pedro and shadowed his wrists to block self-injury and aggression. A verbal prompt was provided every 30 s to "be calm." These continued until Pedro did not exhibit peak behavior for 30 s.

Upon Daniel entering the peak phase the experimenter moved to or remained between Daniel and windows and removed items that could be thrown from his vicinity. Access to preferred items was blocked. Once every 30 s the experimenter told Daniel, "you need to be calm to earn toys." This continued until Daniel was absent of peak behavior for 30 s.

De-escalation. De-escalation referred to a phase after the student ceased to exhibit the peak behavior, but was still exhibiting precursor behaviors or other inappropriate behaviors. Strategies focused on further calming students while also maintaining extinction (e.g., praising for appropriate behaviors, reminders of upcoming preferred activities, maintaining the demand).

The de-escalation phase for Kyon and Pedro was defined as being absent of peak behavior for 30 s, but continuing to exhibit precursor behavior, resist prompting, and/or crying. Daniel's definition of de-escalation was simply absence of peak behavior for 30 s, but continuing to exhibit precursor behavior. If a demand was in place (e.g., Pedro's leisure, Daniel's work), least-to-most prompting was used to ensure students engaged in it. If there was no activity or task, students were presented with a neutral, mastered task to complete. Verbal praise was provided when students engaged in the task. Students were prompted to continue with the task until they were absent of peak behavior, precursor behavior, crying, and/or active resistance for 30 s.

Recovery. Recovery refered to the phase after the peak behavior had been exhibited and the participant was calm (i.e., 30 s absent of all challenging behavior). Strategies involved reviewing the replacement behavior (e.g., you need to ask, first puzzle) and then implementing a correction trial. This involved returning to and recreating the evocative event with additional prompting to ensure the replacement behavior occurred appropriately. If the student was waiting (i.e., Kyon), engaging in the leisure task (i.e., Pedro), or it was during an extinction part of a chained schedule (i.e., Daniel), the interval was reset and began again.

Maintenance. Maintenance data were collected on a minimum of three sessions per student. Conditions for maintenance matched those described for FBCI. Due to scheduled vacations and school testing, maintenance lasted between 6 and 12 weeks for the students.

Generalization. Generalization sessions were planned to be conducted in the students' classroom during naturally occurring opportunities. The teacher was provided with a video camera and asked to record times that matched the baseline or FBCI conditions. Unfortunately the teacher did not complete the generalization sessions, and thus no data are available.

Social Validity

Social validity data were collected from the students' teacher and teacher assistants. They were asked to rate items related to whether FBCI was (a) appropriate, (b) safe, (c) effective, (d) led to an acceptable reduction in challenging behavior, (e) could be used in the classroom, and (f) should be used in the classroom. A 5-point Likert scale was used with 5 meaning "strongly agree" and 1 referring to "strongly disagree." The teacher and teacher's assistants were unanimous in their strong agreement that FBCI was appropriate, effective, led to significant reductions in severe challenging behavior, could be used by teachers and paraprofessionals, and should be used by teachers and paraprofessionals. Regarding safety, the teacher and one paraprofessional strongly agreed that safety was adequately maintained, while the second paraprofessional rated that he simply agreed (i.e., a score of 4) that safety was maintained.

Interobserver Agreement

Interobserver agreement was calculated for 37% of all sessions including 35% of baseline sessions, 39% of FBCI sessions, and 30% of maintenance sessions. A second observer viewed recordings of the sessions and recorded data on the dependent variables. Interval-by-interval agreement was used to determine percent agreement (Cooper et al., 2007). There was 94% agreement.

Procedural Fidelity

Procedural fidelity was assessed for 35% of sessions. A second observer viewed the recordings and scored whether each step of the plan was implemented as written. Data showed 98.5% of steps were implemented with fidelity.


The FBA showed that all students' challenging behavior were maintained by access to tangibles, with Pedro's and Daniel's behaviors having secondary functions of attention. All students' challenging behavior showed high or increasing trends during baseline. Introduction of the intervention resulted in immediate improvements in trend and level, with Kyon's behavior demonstrating initial variability that eventually stabilized at low levels. Low levels of behavior were maintained for all students throughout the maintenance phase, except for the final session for Daniel. Results are displayed in Figure 1.

Kyon. During baseline Kyon's behavior showed an increasing, variable trend. He exhibited challenging behavior for a mean of 41.3% of intervals, ranging from 10% to 82%. His data changed to a lower level with a decreasing trend during the intervention phase as he exhibited challenging behavior for a mean of 13.7% of intervals, ranging from 0% to 70%. Maintenance reflected low, stable rates with challenging behavior occurring for a mean of 0.3% of intervals, ranging from 0% to 2%.

Pedro. During baseline, Pedro's challenging behavior data showed a stable, increasing trend with a mean of 83% of intervals and a range from 62% to 92%. The intervention resulted in an immediate change to a decreasing trend and low level. During this phase, Pedro exhibited challenging behavior for a mean of 23.3% of intervals and a range from 5% to 98%. Pedro's decreased responding remained at low, stable rates throughout the maintenance phase with a mean of 9.3% of intervals ranging from 3% to 15%.

Daniel. Daniel exhibited challenging behavior for a mean of 29% of intervals during baseline, ranging from 0% to 87%, which reflected an increasing trend. The intervention resulted in an immediate change in the data to a low, stable level. During this phase, Daniel's challenging behavior occurred for a mean of 2.2% of intervals, ranging from 0% to 7%. Daniel's low, stable trend continued until the last day of maintenance. During this phase he exhibited challenging behavior for a mean of 12.3% of intervals and a range from 0% to 37%.


The purpose of this study was to evaluate whether organizing FBI according to a crisis intervention framework, namely the Cycle of Acting-Out, would effectively reduce the challenging behavior of three students with autism. As can be seen by the data, all students had improvement in trend and level upon receiving the intervention. These results suggest crisis intervention frameworks can provide an effective vehicle for organizing and delivering FBI, indicating that practitioners may not need to choose one model or the other, but can implement both models simultaneously. Last, and as stated previously, crisis management models meet expert recommendations to improve the adoption of behavior analytic strategies, such as embedding interventions in an easily understood framework. As such, this method appears to have some promise for increasing caregiver use of behavior analytic methods when working with individuals with severe challenging behavior.

A second contribution this study makes to the literature is that it reports the effects of a comprehensive treatment model for intervening with severe challenging behavior. As Hanley et al. (2014) mentioned, the vast majority of research using FBI has focused on the effects of isolated intervention components on severe challenging behavior as opposed to the effects of comprehensive treatment models. It is research on these types of comprehensive models that Hanley et al. argue are necessary for broader adoption of FBI. The model described here addresses that gap by providing an effective framework to organize a number of behavior analytic interventions. This includes, but is not limited to, creating abolishing operations, functional communication training using behavioral indication, targeting precursor behaviors, extinction, change-over delays, schedule thinning, and behavioral shaping.

Third, this study provides an effective method for using crisis intervention frameworks. As stated earlier, the effectiveness of crisis intervention interventions on actually reducing the challenging behavior of individuals with developmental disabilities is lacking. Organizing FBIs according to a crisis intervention framework suggests a way to use crisis intervention frameworks to effectively reduce challenging behaviors.


Given the fact this was an initial study, the results should be understood within the context of the following limitations. First, little can be said about generalization because the teacher did not complete the generalization sessions as directed. Thus there are no data on students' performance outside of the experimental sessions.

A second factor that limits generalization is the study occurred in a contrived setting. All sessions took place in a conference room, or isolated portion of the classroom. Although conditions were designed to replicate those encountered in the classroom, the experimenter designed and implemented the intervention, not the natural caregivers. Also, the intervention was only implemented for 10 min sessions; it was not implemented through the entire day. It was decided to conduct the study in this way because the intervention did not have any empirical support to this point. Therefore, if any negative effects occurred the chances of these negative effects generalizing back to the natural setting were minimized. However, this limits the conclusions that can be drawn about the success of intervention as it still has not been shown to be effective in natural settings implemented by natural caregivers throughout the day.

Finally, indication of acceptability by natural caregivers is limited to the social validity questionnaire. The ultimate hope for this model of intervention is that by organizing intervention in this way, it will increase use of FBI. This preliminary study was not designed to explore this question. Instead, it was designed to determine if the model was effective as a prerequisite to training caregivers and evaluating their perceptions of it. However, it should still be noted this study did not include a formal investigation into whether natural caregivers prefer this over traditional behavior analytic frameworks, or whether they can implement it with fidelity.

Suggestions for Future Research

There are three major areas for future research. First, future research should extend to natural settings. This could be done in several ways. Future studies could be implemented in the classroom or other settings during naturally occurring opportunities. Additionally, natural caregivers (e.g., teachers) could assume increasing responsibility for the assessment and treatment process. For instance, initial extensions could have natural caregivers simply be trained as the interventionists. If these studies prove effective, studies could be conducted to determine if caregivers can be trained to conduct the planning and/or assessment processes. Another line of research that needs to be conducted is to formally examine caregiver perceptions of using a crisis intervention framework for behavior analytic interventions. There are indications, as described in the introduction, that caregivers would prefer this model (i.e., a comprehensive, resonant model in conversational language). Social validity in this study supports this notion. However, the acceptability of this model in particular has not been demonstrated. Therefore, additional research should focus on whether caregivers prefer these models over traditional behavior analytic plans.

Implications for Practice

The results of this study suggest that behavior analytic interventions can be effective when organized according to an individual's escalating behavior. Therefore, when designing plans for people not trained in behavior analysis, practitioners should consider organizing their interventions in this way. This is because having a planned response for the different phases a person escalates through can be very resonant for caregivers. Additionally, it can provide a comprehensive model that incorporates several strategies (e.g., creating abolishing operations, targeting precursor behaviors, targeting behavioral indication, extinction, shaping appropriate behavior, change-over delay) in a model that is easily understood.

Bradley S. Stevenson

Charles L. Wood

University of North Carolina at Charlotte

Alana C. Iannello

Autism Society of North Carolina

Address correspondence to: Bradley S. Stevenson, Ph.D., Melmark Carolinas, P.O. Box 77591, Charlotte, NC 28271. E-mail:


Allen, D. (2000). Recent research on physical aggression in persons with intellectual disability: An overview. Journal of Intellectual and Developmental Disability, 25, 41-57.

Allen, D., McDonald, L., Dunn, C., & Doyle, T. (1997). Changing care staff approaches to the prevention and management of aggressive behaviour in a residential treatment unit for persons with mental retardation and challenging behaviour. Research in Developmental Disabilities, 18,101-112.

Allen, D., & Tynan, H. (2000). Responding to aggressive behavior: impact of training on staff members' knowledge and confidence. Mental Retardation, 38, 97-104. doi:10.1352/0047-6765(2000)038 <0097:RTABIO>2,0.CO;2

Baker, P. A., & Bissmire, D. (2000). A pilot study of the use of physical intervention in the crisis management of people with intellectual disabilities who present challenging behaviour. Journal of Applied Research in Intellectual Disabilities, 13, 38-45.

Beavers, G. A., Iwata, B. A., & Lerman, D. C. (2013). Thirty years of research on the functional analysis of problem behavior. Journal of Applied Behavior Analysis, 46(1), 1-21.

Becirevic, A., Critchfield, T. S., & Reed, D. D. (2016). On the social acceptability of behavior-analytic terms: Crowdsourced comparisons of lay and technical language. The Behavior Analyst, 39, 305-317.

Bethune, K. S., & Wood, C. L. (2013). Effects of coaching on teachers' use of function-based interventions for students with severe disabilities. Teacher Education and Special Education, 36(2), 97-114.

Blacher, J., & Mclntyre, L. L. (2006). Syndrome specificity and behavioural disorders in young adults with intellectual disability: Cultural differences in family impact. Journal of Intellectual Disability Research, 50(3), 184-198.

Blood, E., & Neel, R. S. (2007). From FBA to implementation: A look at what is actually being delivered. Education and Treatment of Children, 30(4), 67-80.

Carr, E. G., Langdon, N. A., & Yarbrough, S. C. (1999). Hypothesis-based intervention for severe problem behavior. Functional analysis of problem behavior: From effective assessment to effective support, 9-31.

Chandler, L. K., & Dahlquist, C. M. (2006). Functional assessment: Strategies to prevent and remediate challenging behavior in school settings (2nd ed.). Upper Saddle River, NJ: Pearson.

Colvin, G., & Scott, T. M. (2015). Managing the cycle of acting-out behavior in the classroom (2nd ed.). Thousand Oaks, CA: Corwin.

Cooper, J. O., Heron, T. E., & Heward, W. L. (2007). Applied Behavior Analysis (2nd ed.). Upper Saddle River, NJ: Merrill/Prentice Hall!

Crisis Prevention Institute (n.d.). About CPI. Retrieved from: https://

Dawson, C. A. (2003). A study on the effectiveness of Life Space Crisis Intervention for students identified with emotional disturbances. Reclaiming Children and Youth, 11,223-230.

Emerson, E., Kiernan, C., Alborz, A., Reeves, D., Mason, H., Swarbrick, R., ... & Hatton, C. (2001). Predicting the persistence of severe self-injurious behavior. Research in Developmental Disabilities, 22(1), 67-75.

Fritz, J. N, Iwata, B. A., Hammond, J. L., & Bloom, S. E. (2013). Experimental analysis of precursors to severe problem behavior. Journal of Applied Behavior Analysis, 46, 101-129.

Gilliam, J. E. (1993). Crisis management for students with emotional/ behavioral problems. Intervention in School and Clinic, 28, 224-230.

Hagopian, L. P., Boelter, E. W., & Jarmolowicz, D. P. (2011). Reinforcement schedule thinning following functional communication training: Review and recommendations. Behavior Analysis in Practice, 4(1), 4-16.

Hagopian, L. P., Fisher, W. W., Sullivan, M. T., Acquisto, J., & LeBlanc, L. A. (1998). Effectiveness of functional communication training with and without extinction and punishment: A summary of 21 inpatient cases. Journal of Applied Behavior Analysis, 31, 211-235.

Hanley, G. P. (2012). Functional assessment of problem behavior: Dispelling myths, overcoming implementation obstacles, and developing new lore. Behavior Analysis in Practice, 5(1), 54-72.

Hanley, G. P., Jin, C. S., Vanselow, N. R., & Hanratty, L. A. (2014). Producing meaningful improvements in problem behavior of children with autism via synthesized analyses and treatments. journal of Applied Behavior Analysis, 47,16-36.

Ingram, K., Lewis-Palmer, T., & Sugai, G. (2005). Function-based intervention planning: Comparing the effectiveness of FBA function-based and non-function-based intervention plans. Journal of Positive Behavior Interventions, 7,224-236.

Jenkins, R., Rose, J., & Lovell, C. (1997). Psychological well-being of staff working with people who have challenging behaviour. Journal of Intellectual Disability Research, 41, 502-511.

Lawson, D. A., & O'Brien, R. (1994). Behavioral and self-report measures of staff burnout in developmental disabilities. Journal of Organizational Behavior Management, 14, 37-54.

Northup, J., Wacker, D., Sasso, G., Steege, M., Cigrand, K., Cook, J., & DeRaad, A. (1991). A brief functional analysis of aggressive and alternative behavior in an outclinic setting. Journal of Applied Behavior Analysis, 24, 509-522.

Powers, R. E. (2005). Primary care assessment and management of aggressive behavior toward others for the adult with Mental Retardation and Developmental Disabilities (MR/DD). Dementia Education Training Association, DDMED2. Retrieved from

Rooker, G. W., Jessel, J., Kurtz, P. F., & Hagopian, L. P. (2013). Functional communication training with and without alternative reinforcement and punishment: An analysis of 58 applications. Journal of Applied Behavior Analysis, 46, 708-722.

Ryan, J. B., Peterson, R., Tetreault, G., & Hagen, E. V. (2007). Reducing seclusion timeout and restraint procedures with at-risk youth. Journal of At-Risk Issues, 23(1), 7-12.

Scott, T. M., McIntyre, J., Liaupsin, C., Nelson, C. M., Conroy, M., & Payne, L. D. (2005). An examination of the relation between functional behavior assessment and selected intervention strategies with school-based teams. Journal of Positive Behavior Interventions, 7, 205-215.

Scott, T. M., Nelson, C. M., & Zabala, J. (2003). Functional behavior assessment training in public schools: Facilitating systemic change. Journal of Positive Behavior Interventions, 5(4), 216-224.

Smith, J. M. (2016). Strategies to position behavior analysis as the contemporary science of what works in behavior change. The Behavior Analyst, 39, 75-87. doi:10.1007/s40614-015-0044-3

Soenen, B., Goethals, I., Spriet, E., D'Oosterlinck, F., & Broekaert, E. (2009). Effects of the combination of life space crisis interventions and a level system at the therapeutic treatment center 'Heynsdaele': a special school and home for youth with behavioral and emotional problems. Therapeutic Communities, 30, 200-216.

Van Acker, R., Boreson, L., Gable, R. A., & Potterton, T. (2005). Are we on the right course? Lessons learned about current FBA/BIP practices in schools. Journal of Behavioral Education, 14, 35-56. doi:10.1007/s10864-005-0960-5

Williams, D. E. (2009). Restraint safety: An analysis of injuries related to restraint of people with intellectual disabilities. Journal of Applied Research in Intellectual Disabilities, 22(2), 135-139.

Caption: Figure 1. Percentage of intervals with challenging behavior.
Table 1
Cycle of Acting-Out Phases and Intervention Foci

Phase           Cycle of Acting Out

Calm            Behavior is cooperative
                and acceptable
Triggers        Behavior involves a series
                of unresolved problems
Agitation       Behavior is unfocused and
Acceleration    Behavior is staff-engaging,
                leading to further negative
Peak            Behavior is out of control
De-escalation   Behavior shows confusion
                and lack of focus
Recovery        Behavior shows eagerness
                for busy work and
                reluctance to interact

Phase           Cycle of Acting Out Foci

Calm            * Manage setting events (e.g., class
                structure, instructional methods)
                * Teach appropriate behaviors (e.g.,
                replacement behaviors, social
Triggers        * Prevention (e.g., minimizing
                triggers, precorrection)
Agitation       * Calming the student (e.g.,
                providing space, allowing access
                to calming activities, providing
Acceleration    * De-escalation (e.g., non-threatening
                interactions and limit
                setting, presenting consequences
                as choices)
Peak            * Maintain safety
De-escalation   * Prepare for reintegration (e.g.,
                isolate, engage in independent
                work, restore environment)
Recovery        * Reintegrate (e.g., reestablish
                rapport via debriefing, focus on

Table 2
Modified Definitions and Interventions

Phase           FBCI Definitions

Calm            Engages appropriately in
                the environment
Triggers        Environmental variables
                known to evoke challenging
                behavior(s) are present
Agitation       Exhibits behaviors
                indicating establishing
                operation(s) are in effect
Acceleration    Exhibits precursor
                behaviors, indicating
                dangerous behaviors are
Peak            Exhibits dangerous
Deescalation    Ceased exhibiting dangerous
                behaviors, but
                continues to exhibit
                precursor behaviors or
                other signs of distress (e.g.,
Recovery        Has not exhibited challenging
                behavior or any other
                signs of distress for at least
                30 s

Phase           FBCI Interventions

Calm            * Environmental enrichment
                * Create abolishing operations for
                challenging behaviors
                * Provide high-quality reinforcement
                for appropriate behaviors
Triggers        * Minimize triggers
                * Make triggers less aversive
                * Provide SD for appropriate/
                replacement behavior
Agitation       * Prompt the replacement behavior
                using least-to-most prompting
                * During schedule thinning: Indicate
                when reinforcement would be
Acceleration    * Prompt the replacement behavior
                using a controlling prompt
                * During schedule thinning: Indicate
                when reinforcement would be
                * Change-over delay
Peak            * Maintain safety using preplanned
                nonviolent physical intervention
                * Withhold or provide inferior
                quality reinforcement (e.g.,
                attention is provided due to
                physical intervention, but eye
                contact and vocal interaction are
Deescalation    * Shape appropriate behavior by
                praising appropriate responses
                (e.g., following directions)
                * Continue to withhold the identified
                reinforcer for the behavior
Recovery        * Recreate initial environment
                * Implement correction trial
COPYRIGHT 2019 West Virginia University Press, University of West Virginia
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Stevenson, Bradley S.; Wood, Charles L.; Iannello, Alana C.
Publication:Education & Treatment of Children
Article Type:Clinical report
Geographic Code:1USA
Date:Aug 1, 2019
Previous Article:Increasing Play Skills in Children with Autism Spectrum Disorder via Peer-Mediated Matrix Training.
Next Article:Assessment and Treatment of Aggression During Public Outings.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters