Functional Communication Training and Demand Fading Using Concurrent Schedules of Reinforcement
Demand fading, a schedule thinning procedure for escape-maintained behaviors, typically includes an escape extinction component. The purpose of this study was to examine the effectiveness of demand fading with alternative reinforcement utilizing concurrent reinforcement schedules without extinction. During demand fading, aggression and requests emitted prior to meeting the task completion criterion were reinforced with short, low-quality breaks, but requests emitted following the task completion criterion were reinforced with long, high-quality breaks. Results suggest that concurrent schedules of reinforcement may be an effective alternative to extinction as a component of demand fading.
KeywordsDemand fading Choice making Functional communication training
Functional communication training (FCT) involves teaching an individual to emit a socially appropriate functional communicative response (FCR) to access reinforcers that maintain problem behavior (Carr and Durand 1985). Although FCT has been demonstrated to be a highly effective treatment to reduce problem behaviors among individuals with intellectual disabilities (Tiger et al. 2008), some drawbacks of FCT have been identified. One drawback is that individuals may emit the FCR too often to be reinforced in the natural environment (Hagopian et al. 2011). For example, an individual may learn to mand for a break from work and that FCR will be strengthened through negative reinforcement, which prevents the individual from completing the undesired, but necessary, task. Hanley et al. (2001) found that post-training FCR rates often approximate pretreatment rates of the problem behavior. If the FCR is occurring too frequently to be practically reinforced in a natural setting, it is likely that the response will weaken or extinguish because the FCR no longer contacts a reinforcement schedule similar to that within FCT. Therefore, FCT alone may not promote generalization and maintenance of the FCR.
Schedule thinning following FCT involves systematically reducing the rate or density of the FCR reinforcement schedule until it meets a predetermined schedule of reinforcement (Hagopian et al. 2011). Hagopian and colleagues reviewed 76 studies on FCT published between 1985 and 2009 to find that only 29% described a schedule thinning phase. Four schedule thinning procedures identified included: (a) delay schedules; (b) demand fading; (c) multiple schedules; and (d) response restriction.
Of the studies reviewed, eight studies implemented demand fading for escape-maintained problem behavior, which involves increasing the number of demands that must be completed before the FCR will be honored. Demand fading was concluded to have the strongest support for problem behavior maintained by negative reinforcement (Hagopian et al. 2011). A common component across all eight studies was the implementation of escape extinction not only during the initial FCT, but also during demand fading. Escape extinction is the continued presentation of the demand after the display of the problem behavior; in other words, the problem behavior no longer results in escape from the demand.
Although escape extinction has been demonstrated to be effective, it is also associated with limitations, including the difficulty or inability to continue presentation of a task due to the individual’s strength, size, or nature of the problem behavior (Piazza et al. 1996), emotional responding, variability in the target behavior, and the possibility of extinction bursts (Lerman and Iwata 1995; Morgan and Lee 1996; Saini et al. 2016). Previous research indicates extinction bursts may occur in 11.7% (Saini et al. 2016) to 36% (Lerman and Iwata 1995) of all cases; however, considering all of the aforementioned limitations to extinction, it is worthwhile to explore possible effectiveness of alternative approaches that do not utilize extinction. A potential alternative to the use of escape extinction during FCT and demand fading is the implementation of concurrent schedules of reinforcement. This is applicable to FCT in that an individual selects between emitting the FCR, completing the task, or engaging in the problem behavior, each of which is associated with a concurrent schedule of reinforcement. Literature suggests that behavior selected in this concurrent schedule of reinforcement is influenced by variations in reinforcement type, quality, duration, immediacy, and magnitude, as well as response effort (Horner and Day 1991; Neef et al. 1992).
Athens and Vollmer (2010) evaluated the effects of manipulations of reinforcement for problem behavior and appropriate behavior within differential reinforcement of alternative behavior (DRA) without extinction. In the first three experiments, the duration, quality, and delay to reinforcement were individuality manipulated, across the three experiments, respectively. Although responding was sensitive to the manipulation of duration, quality, and delay, individually results were gradual and the specific manipulations in durations, quality, and delay that influenced differential responding were individualized. In Experiment 4, the effects of immediate, longer duration access to high-quality reinforcement for appropriate behavior and delayed, shorter duration access to low-quality reinforcement for problem behavior were evaluated. A larger and more consistent change in behavior occurred when various aspects of reinforcement were manipulated, relative to results found in the first three experiments. The results of this study indicate the effectiveness of variations in reinforcement to influence responding within DRA. Considering FCT is a form of DRA in which the alternative behavior is a communicative response specifically, it may be possible to implement FCT and demand fading without extinction if reinforcement is altered for each response (i.e., FCR, task completion, and problem behavior) across one or more of the aforementioned dimensions.
Concurrent schedules of reinforcement may not only manipulate dimensions of reinforcement, but also types of reinforcement. For example, Hoch et al. (2002) evaluated the effects of concurrent schedules of reinforcement on negatively reinforced problem behavior and task extinction by manipulating access to type of reinforcement for challenging behavior and task completion. When challenging behavior and task completion both contacted negative reinforcement in the form of a break from tasks, challenging behavior remained high and task completion remained low. However, when challenging behavior contacted negative reinforcement, but task completion contacted both negative and positive reinforcement in the form of a break from tasks with access to preferred activities, challenging behavior was reduced to near zero levels and task completion increased. Schieltz et al. (2017) implemented a similar procedure with participants whose challenging behavior was maintained by negative reinforcement. During treatment, challenging behavior continued to contact negative reinforcement, but task completion contacted both negative and positive reinforcement in the form of a break from the task and access to a leisure activity. These studies indicate that concurrent schedules of reinforcement containing positive reinforcement may be successful in reducing negatively reinforced challenging behavior.
Peck Peterson et al. (2005) demonstrated the utility of a choice-making component within FCT. Participants’ problem behavior was maintained by negative reinforcement. FCT effectively decreased problem behavior, but participants continuously manded for breaks. A choice-making contingency was added; participants could mand for a break and receive a low-quality break in which they received a 15-s break from work, but were not delivered toys or attention. If the participants manded for work and then completed the task, the participants were allowed to take a 2-min break in which toys and adult attention were continuously available. When higher quality breaks were contingent upon choosing and completing work, task engagement increased without an increase in problem behavior. However, problem behavior was placed on extinction during FCT with the choice-making component.
The current study replicates and expands on this research in three ways. First, Peck Peterson et al. (2005) implemented extinction for problem behavior; the current study eliminated extinction altogether. Second, this study expanded upon the potential merit of procedures outlined by Peck Peterson and colleagues by systematically altering concurrent reinforcement contingencies within a demand fading sequence. Finally, a discriminative stimulus was added to signal the FCR reinforcement contingencies in place. The purpose of this study was to evaluate the effectiveness of demand fading procedures using concurrent schedules of reinforcement in lieu of extinction.
Participant and Setting
The participant, Noah, was a 7-year-old male diagnosed with attention deficit hyperactivity disorder, oppositional defiant disorder, and disruptive mood dysregulated disorder. He was homeschooled by his mother who reported he was on grade-level (i.e., 1st grade) for most academic subjects, but displayed a marked level of problem behavior during academic work. Noah had no speech or language impairments; he was capable of communicating clearly in full sentences. He was referred to a university-based applied behavior analysis clinic for treatment of aggressive behavior.
An informal phone interview with Noah’s family was conducted to gather more information about the topography and potential function of Noah’s problem behavior. Noah had a history of aggression toward others. Noah’s parents removed him from public school because they were concerned teachers would be unable to manage his behavior in a typical classroom setting. Noah’s parents reported he was likely to engage in aggression when asked to perform a nonpreferred task. Noah’s mother reported specific concerns with handwriting tasks. Although Noah had no known fine motor limitations and was capable of writing all upper and lower case letters of the alphabet in a legible fashion, he displayed frequent and severe aggression during any handwriting task. In collaboration with Noah’s parents, handwriting, specifically writing his name, was targeted for intervention.
Sessions were conducted at a university-affiliated clinic in a room approximately 3 m × 3 m. The room contained a child-sized table, chairs, and research-specific materials.
Response Definitions, Measurement, and Interobserver Agreement
Target responses were aggression, task completion, and the FCR. Aggression included hitting, kicking, pencil stabbing, biting, and eye gouging. Due to the severity of aggression, several attempts of aggression were blocked or dodged to protect the experimenter. Blocking or dodging was implemented if the experimenter believed she was at risk for injury that would result in tissue damage; therefore, stabbing, biting, and eye gouging attempts were blocked or dodged. Attempted stabbing was defined as Noah moving quickly toward an experimenter while holding a pencil in a fist with the sharpened end of the pencil facing the experimenter. Attempted biting was defined as Noah leaning his head within 0.3 m from the experimenter with an open mouth. Attempted eye gouging was defined as Noah’s hand, with fingers extended, being within 0.3 m of the experimenter’s face. Blocking was conducted by the experimenter placing her hands in front of her body to prevent Noah from making contact. Dodging involved moving quickly to the side of Noah’s attempted aggression (e.g., stepping quickly aside when he came at her with the sharpened end of a pencil). If necessary, a second experimenter would steady Noah’s hands and remove the pencil to ensure the safety of the experimenter. Attempted and completed aggressions were included in the total percentage of aggression measured using a 10-s partial-interval recording system.
The FCR was the independent vocal emission “I want a break” or any other functionally equivalent two-word or longer phrase (e.g., “Break please.”). The FCR was considered prompted if the break request was made after the experimenter prompted the request by instructing Noah, “Say ‘I want a break’.” FCRs were measured with a 10-s partial-interval recording system. Task completion was defined as the number of upper or lower case letters written legibly on lined paper, measured using a frequency count.
A second observer independently collected data on 83% of sessions. Interobserver agreement for aggression and communication was calculated by dividing the number of intervals in which observers agreed by the total number of intervals and converting to a percentage. Mean percentage agreement for aggression and communication was 99.1% (range 87–100%) and 99.6% (range 93–100%), respectively. Interobserver agreement for task completion was calculated by dividing the smaller number of observed occurrences by the larger number of observed occurrences and converting to a percentage. Mean percentage agreement for task completion was 99.7% (range 93–100%).
Experimental Sequence and Design
A function analysis was first conducted to confirm the variables maintaining aggression. The functional analysis was analyzed using a pairwise design (Iwata et al. 1994b). Next, the effects of FCT were evaluated using an ABABC design with the C phase representing demand fading (Kennedy 2005).
Due to parent interview results indicating a likely escape function, a functional analysis was conducted to confirm the hypothesis that aggression was maintained by negative reinforcement. Functional analysis sessions were 5 min in duration. Escape and control conditions were conducted similarly to those described by Iwata et al. (1982/1994a). During escape conditions, Noah was asked to write his name. Contingent upon aggression, the task was removed and the experimenter turned away for 15 s. During control sessions, no demands were placed and Noah was provided with his highest preferred item, an iPad, and provided experimenter attention approximately every 10 s and aggression was ignored.
Baseline sessions were 5 min in duration. During baseline, the experimenter placed lined paper and a pencil in front of Noah and instructed him to write his name. Praise was provided contingent upon each correct letter. If Noah engaged in aggression, the experimenter removed the paper and pencil and turned away from Noah for 30 s. All nontargeted problem behavior (e.g., passive noncompliance) and the FCR were ignored.
Functional Communication Training
FCT sessions were 5 min in duration. During FCT, the experimenter taught Noah to utilize the FCR to solicit a break from work tasks. No prior training sessions were conducted to teach Noah to use the FCR. Prior to beginning each session, the experimenter briefly explained, “When the green card is out, if you ask for a break you can have a long break with your iPad.” The experimenter began the session by placing the paper and pencil in front of Noah and instructing him to write his name. Immediately after placing the demand, the experimenter also placed a 10-cm by 15-cm laminated green card on the table and prompted Noah to request a break (i.e., “Say, ‘I want a break.”). If Noah did not request a break within 5 s, the prompt was repeated; however, this was never implemented because Noah always responded to the initial prompt. Prompts were faded using a 2-s progressive time delay. After Noah emitted three consecutive correct FCRs, the experimenter increased the delay between the demand and the prompt by 2 s. The increase in prompt delay continued across sessions and conditions rather than beginning at a 0-s delay for the start of a new session. Noah began independent communication after the delay was increased to 4 s; therefore, the prompt was never delayed more than 4 s.
The green card served as a discriminative stimulus for the availability of a high-quality break; however, the green card was available at all times during FCT. Contingent upon prompted and independent break requests, the experimenter provided a high-quality break, by removing work materials, offering an iPad, and turning away from Noah for 30 s. The increase in duration of the break and the inclusion of the iPad was provided during the high-quality break in order to improve the features of the high-quality break across two dimensions: duration and quality. Moreover, this potentially improved discrimination between break choices. Contingent upon aggression, the experimenter provided a low-quality break, by removing work materials and turning away from Noah, without offering an iPad, for 10 s. Contingent upon the completion of his name, Noah was praised and directed to write his name again. If Noah was passively noncompliant (i.e., did not complete the task, but did not display aggression), the experimenter re-prompted him to write his name every 10 s.
Terminal Schedule Probe
Before implementing demand fading, Noah’s tolerance for delayed reinforcement was assessed. The terminal schedule probe session was 5 min in duration. The experimenter began the session by placing the paper and pencil in front of Noah, telling him, “Write your name four times and then I will place the green card on the table; when the green card is on the table and you ask for a break, you will get a long break with the iPad.” His first and last names consisted of 10 letters; therefore, the experimenter essentially asked Noah to write 40 letters. Contingent upon the completion of the task, the experimenter placed the green card on the table. When the green card was present, independent break requests resulted in a high-quality break; however, when the green card was not present, independent break requests resulted in a low-quality break. Contingent upon aggression at any time, with or without the green card present, a low-quality break was delivered.
Demand fading sessions were 5 min in duration. During demand fading, the experimenter systematically increased the number of letters Noah was required to complete before presenting the discriminative stimulus for the high-quality break. Contingent upon two consecutive sessions with no aggression, the number of letters required for the presentation of the green card was doubled. However, in two phases (the second 1 letter condition and the third 2 letter condition), an additional session was conducted by error for a total of three sessions per condition. Similarly, after the first 2 letter condition, the number of letters required for the presentation of the green card was erroneously increased to 5 letters, as opposed to 4 letters. This systematic increase in letters Noah was required to complete before the presentation of the discriminative stimulus for the availability of a high-quality break continued until Noah reached 16 letters; at this time, rather than doubling to 32 letters, we simply asked him to write his full name 3 times (i.e., 30 letters). Demand fading from this point on consisted of increasing the number of times he wrote his full name by one (i.e., 4 times, 5 times). If Noah displayed aggression during any session, demand fading was decreased to the previous contingency.
During demand fading, the experimenter began the session by placing the paper and pencil in front of Noah, telling him, “Write [‘the first N letters of your name’ or ‘your name N times’] and then I will place the green card on the table; when the green card is on the table and you ask for a break, you will get a long break with the iPad. You can ask for a break before the green card is out, but it will be a short break with no iPad.” Noah was able to request a break at any time during the session. Contingent upon completion of the specified contingency, the experimenter placed the green card on the table. When the green card was present, break requests resulted in a high-quality break; when the green card was not present, break requests resulted in a low-quality break. If Noah did not ask for a break immediately after the green card was presented, the experimenter continued to present instructions to write until he requested a break. Contingent upon aggression at any time, with or without the green card present, a low-quality break was delivered.
During the terminal schedule probe, aggression increased to a mean of 93% of intervals, no FCR was emitted, and no letters were completed. During demand fading, with the exception of an escalation during the 6th and 7th sessions, aggression stayed at zero levels. The FCRs gradually decreased from a mean of 29% of intervals during FCT, to a mean of 18% of intervals (range 13–20%) during the initial demand fading contingency (i.e., write 1 letter) to a mean of 7% of intervals during the final contingency (i.e., write 50 letters). Task completion rose from 0 letters completed during the initial FCT sessions to a mean of 100 letters per session during the final contingency (i.e., write 50 letters). Demand fading reached the final contingency within 24 sessions, which was approximately 2 h of intervention.
These results are consistent with previous findings exposing the limitations of FCT for negatively reinforced problem behavior, which include emissions of FCR at such high levels they result in continuous access to breaks and minimal task completion during FCT (e.g., Peck Peterson et al. 2005). During FCT, Noah’s problem behavior quickly decreased, but the FCR was emitted so often that he never completed a single task because he accessed nearly continuous breaks. In other words, although aggression was successfully treated, task completion was unaffected.
Demand fading has been successful at addressing these limitations associated with FCT, but previous studies have utilized escape extinction within the demand fading procedure. With its aforementioned limitations, escape extinction may not be a viable option when problem behavior is dangerous and/or makes escape extinction difficult to implement with fidelity. This was the case with Noah as his aggression was severe enough to cause injury to the experimenter. It would have been unreasonable to expect an adult to continue presentation of the writing task if Noah were attempting to stab or eye gouge, as the adult would need to protect him or herself, which would naturally result in an escape from the writing task.
This study expands upon the research conducted by Peck Peterson et al. (2005) demonstrating the advantages of concurrent schedules of reinforcement within FCT. Peck Peterson and colleagues demonstrated that the manipulation of reinforcement dimensions across mands for work and break increased task completion; however, problem behavior was placed on extinction. In this study, dimensions of reinforcement were altered across three responses: work, mands for a break, and problem behavior. In fact, Noah contacted reinforcement for aggression in the 6th and 7th sessions of demand fading when problem behavior reemerged. This reemergence was quickly followed by a decrease in problem behavior, demonstrating that Noah sampled the reinforcement contingency for problem behavior, but subsequently chose to display task compliance and FCR to access the associated reinforcement contingencies in future sessions.
Not only was escape extinction eliminated in this study, but demand fading was also implemented without manipulating the ability to emit the FCR or placing appropriate requests on extinction. Demand fading typically involves increasing the number of demands that must be completed before (a) the ability to mand is presented or (b) the mand is reinforced. Perry and Fisher (2001) implemented the first approach by delivering the picture card needed to request a break only after the participant completed the specified number of tasks. This procedure is only applicable when an external device (e.g., picture card, microswitch) is necessary to emit the FCR. Fisher et al. (1993) implemented the second approach with a participant who was taught to sign for a break. Within the demand fading procedure, the sign was only reinforced after the specified number of tasks had been completed. In the current study, Noah communicated vocally; therefore, the ability to emit the FCR could not ethically be manipulated. Moreover, all FCRs were reinforced, but the reinforcement quality was altered in that high-quality breaks were delivered contingent upon an FCR emitted after the specified number of tasks were completed, but low-quality breaks were delivered contingent upon an FCR being emitted before the specified number of tasks were completed. This study expands upon the current literature by demonstrating demand fading can be conducted without restricting the ability to emit the FCR or placing it on extinction until the task criterion has been met.
Although the demand fading procedure was successful without the use of response restriction or extinction, it is unclear the potential influence of the vocally delivered rule describing contingencies to this success. Noah had no known receptive communication delays; therefore, based on his age and development, the vocally delivered rule may have served as a response prompt. Leon et al. (2010) utilized rule statements to facilitate discriminated responding within FCT. Requests for attention were placed on extinction when the experimenter was busy but reinforced when the experimenter was engaged in nonbusy activities. Requests for attention persisted when the experimenter was busy, so a rule statement was added at the beginning of each busy period, which resulted in discriminated responding. It is possible the rule statement contributed to the success of discriminated responding in the current study as well.
The inclusion of demand fading was critical to address the fact within the FCT intervention, Noah never completed any tasks. While clearly FCT was effective at reducing challenging behavior, it lacks social validity if task completion remained unaffected. For the most part, during demand fading Noah’s task completion systematically increased across increased response contingencies; that is he generally completed more letters with higher response contingencies (e.g., 30 letters) relative to lower response contingencies (e.g., 2 letters). However, in some sessions, Noah completed more letters relative to sessions associated with higher response contingencies. For example, during the first sessions with a 30-letter contingency he completed 90 letters, but he completed only 80 letters during both sessions with a 40-letter contingency. The current data do not allow for evaluation of the causes of this phenomenon. Anecdotally, this variation appeared to be related to two factors: (a) task completion efficiency and (b) task completion above the minimum contingency to access a high-quality break. In other words, during some sessions, Noah initiated letter writing after the explanation of the contingency faster than others, but this short delay to letter writing initiation was not associated with challenging behavior. Similarly, in some sessions, Noah completed letters more quickly than other sessions. Moreover, in some sessions, Noah opted to complete more letters than necessary to access the high-quality break. For example, he may have completed 35 letters prior to requesting a break within a 30-letter contingency.
The results of this study are also consistent with previous research that confirms the ability to bias responding to task completion for children whose problem behavior is escape-maintained (e.g., DeLeon et al. 2001; Kodak et al. 2007). Gardner et al. (2009) successfully biased responding toward academic tasks for two participants with escape-maintained problem behavior by providing high-quality attention for task completion. When high-quality attention was provided contingent upon task completion and escape was continuously available, participants selected to complete work and receive attention rather than escape and not receive attention. The results of this study are consistent with this literature which suggests negatively reinforced problem behavior can be addressed without escape extinction if the quality of reinforcement is altered.
These results suggest that concurrent schedules of reinforcement may ameliorate limitations associated with FCT and offer an alternative to extinction during demand fading. However, some limitations should be considered. First, this study altered two dimensions of reinforcement: duration and quality. This was done to ensure contingencies were discriminable; however, the individual effects of each dimension are unknown. Second, although systematic, the procedure to increase task demands within the demand fading condition (i.e., doubling the number of letters after two consecutive sessions with no aggression) was somewhat arbitrary. Relatively little research has examined schedule thinning procedures (Hagopian et al. 2004), and future research should continue to explore the most efficacious schedule thinning approaches. Third, the functional analysis only evaluated the possibility of negative reinforcement maintaining aggression, yet the control condition contained establishing operations associated with contingencies other than negative reinforcement (e.g., noncontingent access to attention was provided in the control condition, but would serve as an establishing operation for attention-maintained behavior). It is possible that aggression was maintained by additional contingencies, such as access to attention or tangibles. The lack of functional analysis regarding additional maintaining contingencies limits the ability to hypothesize why the inclusion of the iPad within the high-quality break biased responding toward task completion. Specifically, it is unclear if the iPad was a functional reinforcer or simply a preferred stimulus. Hanley et al. (1997) compared effectiveness of two forms of noncontingent stimulus access on challenging behavior: (a) noncontingent presentation of a stimulus maintaining challenging behavior and (b) noncontingent presentation of a preferred stimulus. Noncontingent access to both stimuli reduced challenging behavior. In this study, it is unclear if the inclusion of the iPad within the high-quality break was successful due to it being a preferred stimulus or a due to it being another stimulus maintaining challenging behavior. As a result, the lack of functional analysis data to determine if iPad access maintained challenging behavior limits the replicability of this procedure. However, the initial interviews with Noah’s mother suggested problem behavior was maintained by negative reinforcement; therefore, in an attempt to expedite assessment, only the test condition which was supported by the initial hypotheses (i.e., escape) was included in the functional analysis.
Another limitation of this study was that data were collected only on independent FCR; therefore, we were unable to report data on prompted FCR. However, it can be noted from anecdotal observation that Noah required only a few prompts to emit the FCR across both FCT and demand fading sessions across the entire study. Similarly, we did not differentiate FCRs for low- versus high-quality breaks in the graph; however, it was anecdotally noted that he emitted the FCR for a low-quality break no more than a few times across the duration of the study. Fifth, few sessions were conducted during the initial baseline and FCT conditions, which limits experimental control. Finally, only one child participated in this study, which may decrease the external validity of the obtained results. Therefore, it is important these results are replicated with additional participants.
Future research should continue to explore the efficacy of concurrent schedules of reinforcement as an approach to demand fading. An evaluation of the effects of altering a single dimension in isolation is warranted. In the current study, two dimensions were altered within the high- and low-quality breaks, duration, and quality. Future research should examine if altering only duration or only quality would produce similar effects. Moreover, the effects of other types of reinforcers, (e.g., attention) should also be explored. Finally, future research should explore the impact demand fading with concurrent schedules or reinforcement has on long-term maintenance of treatment effects. Wacker et al. (2011) evaluated the persistence of treatment effects of FCT among participants whose destructive behavior was also maintained by negative reinforcement. After participants were exposed to prolonged FCT (5–15 months), destructive behavior no longer recurred during treatment challenges (i.e., extended extinction, introduction of novel tasks, removal of FCR modality, and mixed schedules of reinforcement). Future research should evaluate the persistence of treatment effects of FCT with subsequent demand fading implemented with choice making to determine if similar prolonged treatment is necessary for long-term maintenance of treatment effects.
In summary, FCT has many advantages and some disadvantages. Although demand fading has been implemented across many studies to address these aforementioned disadvantages, such procedures have consistently incorporated an extinction component. This study contributes to the literature by eliminating extinction within demand fading. Results indicate that concurrent schedules of reinforcement are a viable treatment component to demand fading.
Compliance with Ethical Standards
Conflict of interest
The authors declare that they have no conflicts of interest.
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
Informed consent was obtained from all individual participants included in the study.
- DeLeon, I. G., Neidert, P. L., Anders, B. M., & Rodriguez-Catter, V. (2001). Choices between positive and negative reinforcement during treatment for escape-maintained behavior. Journal of Applied Behavior Analysis, 34, 521–525. https://doi.org/10.1901/jaba.2001.34-521.CrossRefPubMedPubMedCentralGoogle Scholar
- Gardner, A. W., Wacker, D. P., & Boelter, E. W. (2009). An evaluation of the interaction between quality of attention and negative reinforcement with children who display negative reinforcement with children who display escape-maintained problem behavior. Journal of Applied Behavior Analysis, 42, 343–348.CrossRefPubMedPubMedCentralGoogle Scholar
- Hagopian, L. P., Toole, L. M., Long, E. S., Bowman, L. G., & Lieving, G. A. (2004). A comparison of dense-to-lean and fixed lean schedules of alternative reinforcement and extinction. Journal of Applied Behavior Analysis, 37, 323–337. https://doi.org/10.1901/jaba.2004.37-323.CrossRefPubMedPubMedCentralGoogle Scholar
- Kennedy, C. H. (2005). Single-case designs for educational research. Boston, MA: Allyn and Bacon.Google Scholar
- Peck Peterson, S. M., Caniglia, C., Royster, A. J., Macfarlane, E., Plowman, K., Baird, S. J., et al. (2005). Blending functional communication training and choice making to improve task engagement and decrease problem behavior. Educational Psychology: An International Journal of Experimental Educational Psychology, 25, 257–274. https://doi.org/10.1080/0144341042000301193.CrossRefGoogle Scholar
- Piazza, C. C., Moes, D. R., & Fisher, W. W. (1996). Differential reinforcement of alternative behavior and demand fading in the treatment of escape maintained destructive behavior. Journal of Applied Behavior Analysis, 29, 569–572. https://doi.org/10.1901/jaba.1996.29-569.CrossRefPubMedPubMedCentralGoogle Scholar
- Wacker, D. P., Harding, J. W., Berg, W. K., Lee, J. F., Schieltz, K. M., Padilla, Y. C., et al. (2011). An evaluation of persistence of treatment effects during long-term treatment of destructive behavior. Journal of the Experimental Analysis of Behavior, 96, 261–282.CrossRefPubMedPubMedCentralGoogle Scholar