Association for Behavior Analysis International

Differential reinforcement of compliance is a common treatment for children that present with escape-maintained problem behavior. This treatment involves providing positive and/or negative reinforcement following compliance with tasks. Traditionally, this arrangement involves brief, fixed-ratio schedules (i.e., small, distributed work requirements) to result in brief reinforcement periods. However, recent research has suggested that individuals may prefer to complete longer ratio schedules (i.e., larger, or accumulated work periods) to access longer reinforcement periods. However, for children with escape maintained problem behavior, accumulated work periods involve prolonged exposure to aversive events and delayed access to positive reinforcement, and thus may occasion increased problem behavior. In the current study, we exposed 3 children with escape-maintained aggression to distributed and accumulated work conditions to assess the efficacy of both arrangements. We then assessed these children's preferences for both arrangements using a concurrent-chains procedure. Across three participants, not only did accumulated work periods not occasion additional problem behavior, it was associated with decreased problem behavior relative to distributed work periods for 2 of 3 participants. Preferences for these three conditions were idiosyncratic across participants.

Functional analyses yield low and undifferentiated problem behavior rates if test conditions fail to include potentially idiosyncratic establishing operations to potentiate relevant reinforcement. Escape-maintained problem behavior during academic instruction may only occur during a functional analysis if sufficiently non-preferred tasks or teaching strategies are incorporated. In the current study, an initial functional analysis of aggression with a child with autism yielded low, undifferentiated rates across conditions. We modified our assessment based upon parental report to increase the likelihood of making errors and contacting an error-correction procedure. This modification led to the identification of aggression being maintained by negative reinforcement in the form of escape from demands where errors are likely to be made. This outcome led to the development of an effective function-based intervention involving differential reinforcement of alternative behavior.

In the current study, a functional analysis showed that problem behavior demonstrated by an 8 year-old male was maintained by adult attention in the form of social disapproval. Functional communication training (FCT) for a commonly used attention-type (i.e. praise), produced low rates of functional communication responses (FCRs) in the form of handing a specific card to a therapist. We conducted a Rapid Assessment of Attention-Types (RAAT) to identify attention-types that could be used to reinforce FCRs as functionally equivalent replacements for problem behavior. The RAAT identified two attention-types likely to produce higher rates of FCRs. We used a reversal design to evaluate the rates of FCRs when using praise to reinforce FCRs in comparison to RAAT-identified attention-types. Results suggested that the RAAT accurately predicted the attention-types that would reinforce FCRs. Additionally, we report outcomes of a function-based treatment for attention-maintained problem behavior that included the attention-types identified with the RAAT.

When problem behavior (PB) is maintained by multiple functional reinforcers, we often assume that different treatment components are needed. However, there are very few studies systematically testing this assumption. This study aimed to evaluate multiply-maintained PB (escape and automatic reinforcement) in a demand context following implementation of treatment targeting automatically-maintained PB with two participants. For participant one, self-injurious behavior (SIB) was multiply-maintained. Treatment included arm splints and differential reinforcement of other behavior (DRO) targeting SIB. For participant two, disruption was at least in part automatically-maintained while aggression was escape-maintained. Treatment included a DRO and punishment, targeting only disruption. Following demonstration of control in an austere environment, we returned to escape baseline. In baseline, the treatment for automatically-maintained problem behavior was in place, but we provided a break from demands contingent on SIB (participant 1) or aggression (participant 2). For both participants, low rates of PB occurred during escape baseline. Potential explanations include that a break was only reinforcing when combined with access to automatic reinforcement, that treatment increased the response effort of SIB (participant 1), or that treatment targeted PB lower in the response class for escape (participant 2). We discuss results in the context of future research and recommendations.

Problem behavior maintained by automatic reinforcement is difficult to treat because we are not able to manipulate the reinforcers maintaining the behavior. Piazza et al (1996) used a component analysis to address pica (eating cigarette butts) maintained by automatic reinforcement. In the current study a component analysis was conducted to identify the factors (sensation of heft and resulting noise) that reinforced destructive behavior (i.e., throwing items) maintained by automatic reinforcement as identified within a functional analysis. The participant, a 9 year old non-verbal boy diagnosed with autism and attention deficit disorder was referred to an intensive behavioral outpatient service to address destructive (throwing and kicking), aggressive, and self-injurious behavior. During the component analysis, the occurrence of throwing was compared across three conditions: (A) items with heft and noise, (B) items with heft but no noise, and (C) items with minimal heft and minimal noise. The results showed that throwing was maintained by heft and was not sensitive to noise. The treatment package provided access to heft using appropriate behaviors (e.g., throwing balls into a basket), a schedule to reduce down-time, and to teach appropriate play skills. Inappropriate throwing decreased to near zero levels as did kicking items.

Behavior analysts commonly employ functional analyses to determine the function of destructive behavior (Iwata, Dorsey, Slifer, Bauman, & Richman, 1982/1994). However, a traditional functional analysis is sometimes insufficient to clearly identify the function of destructive behavior. We evaluated the use of a transition assessment with two children for whom a traditional functional analysis produced inconclusive results. We first conducted an activity preference assessment to determine a hierarchy of preference for a variety of activities. We selected low, moderate, and high-preferred activities to use in the evaluation. We then systematically manipulated which of those activities were available in different contexts in which the client was required to transition to and from. Results indicated that both children engaged in destructive behavior when required to transition from a context with a high-preferred activity to a context with a low-preferred activity. We discuss the implications of the results and provide suggestions for future research.

Individuals with developmental disabilities may engage in severe problem behavior to escape social interaction. Patterns of responding within a standard functional analysis can serve as a source of evidence for social avoidance as reinforcement of problem behavior (Harper, Iwata, & Camp, 2013). In the current case study, an adolescent male admitted to an inpatient unit for assessment and treatment of severe problem behavior displayed frequent aggressive behaviors during the toy play and demand conditions of a functional analysis. Therefore, a social avoidance assessment was conducted to identify the antecedent condition that served as an establishing operation (EO) for problem behavior. High-quality social attention provided by a therapist while near the client was the condition with the shortest average latency to problem behavior. This condition was incorporated into an assessment to evaluate problem behavior related to escape from the proximity of individuals or their social attention. Results suggested the client engages in problem behavior to escape social attention, specifically when individuals are near him. Treatment targeted functional communication and proximity fading of low-preferred staff across different activities. Additional fading procedures were conducted in a demand context to decrease problem behavior maintained by escape from the presentation of demands by low-preferred staff.

Escape Extinction (EE) is a component in effective reinforcement-based treatments (Reed et al., 2004) such as Differential Reinforcement of Alternative Behavior for pediatric feeding disorders. Investigations of why extinction works have analyzed procedural form based on function of targeted behavior (Iwata, Dorsey, et al., 1994; Piazza, Fisher, et al., 2003). However, few studies evaluated the mechanisms underlying the effectiveness of extinction. The current study hypothesized that EE would result in exposure to foods and change preferences for the targeted foods; compliance would result in positive reinforcers. The hypothesis was tested using preference assessments (Pace et al., 1985) conducted before and during EE treatment. Participants are three individuals who engaged in inappropriate mealtime behaviors. Treatment results indicated that EE resulted in decreases in food refusal during treatment. Preference assessment results showed that the food targeted during the EE treatment changed from a nonpreferred food at the start of treatment to a preferred food during treatment, suggesting that positive reinforcement may have a role in maintaining food acceptance. These results support the hypothesis that positive reinforcement may play a role in the maintenance of food acceptance. Future research could evaluate the role of attention as a reinforcer for food acceptance.

Functional communication training (FCT) is an empirically-supported treatment for severe problem behavior in persons with intellectual disability. During FCT, a communicative response is differentially reinforced to provide an appropriate alternative to problem behavior. Selection of a communicative modality for FCT is often based on the individual's current skills and/or caregiver preference. However, research suggests that an individual's preference for communication modality can impact treatment outcome. In this study, a nonverbal 6-year-old male with a degenerative neurological disorder was trained to request a preferred item using two modalities: a picture card and an augmentative communication device (i.e., GoTalk). Proficiency was demonstrated with both modalities, and decreases in problem behavior were observed during FCT with both the picture card and the GoTalk. However, when both modalities were presented concurrently a clear preference for the GoTalk emerged. Additionally, when a blank distractor card was paired with each modality, accurate discrimination between the target and distractor card was observed with the preferred modality, but never with the non-preferred modality, even after multiple training exposures. Results suggest that including preferred communicative modalities into communication-based treatment packages may increase the speed of skill acquisition. Implications for assessment and treatment are discussed.

When presenting reinforcers to individuals with developmental disabilities, many researchers use food. One issue with using food as a reinforcer is that there may be other types of reinforcers which may be equally or more effective. Although preference assessment methods have been well-researched, one area that has not yet been resolved is concerned with whether food reinforcers are always more preferred than non-food reinforcers, when both are used in the same assessment. This study compared preference for food and non-food items with similar and dissimilar reinforcing values in people with developmental disabilities, in order to understand how food and non-food stimulus groups interact with reinforcing value. Participants were four individuals with developmental disabilities. The study first involved measuring the reinforcing value of each food and non-food item through the use of a reinforcer test. The food and non-food reinforcers identified were then paired based on their known reinforcing values, and a paired-stimulus preference assessment was conducted. The data from each participant was examined to determine how frequently each item was selected. A better understanding of how choice options from different reinforcer groups interact may provide insight on the best methods to select and present reinforcers to people with developmental disabilities.

Some individuals engage in aggression and self-injury to avoid challenging instructional situations (Horner et al., 1991). We used a multi-element design to evaluate the role that different forms of corrective feedback on frequency of problem behavior with four individuals admitted to an inpatient unit for the assessment and treatment of severe problem behavior. All participants engaged in problem behavior when presented with feedback during demand situations. During the control sessions, only praise was provided. In the test conditions, verbal (4 of 4 participants) or physical (3 of 4 participants) corrective feedback was provided, regardless of response accuracy. Contingent on problem behavior during all tests conditions, the participants were allowed to finish the task their way. This suggested that problem behavior was maintained by negative reinforcement in the form of escape from corrective feedback. Treatment for all four individuals consisted of positively-stated feedback with verbal and gestural error correction. With the treatment in place, overall rate of problem behavior decreased for all four participants as compared to baseline (see figure). These results suggest that positively-stated feedback with verbal and gestural error correction may be an effective treatment for individuals who exhibit problem behavior when corrected during demand contexts.

Spoon acceptance, defined as full lip closure over the spoon resulting in the bite being deposited into the mouth, is an important factor for children with feeding disorders. Although food can be deposited using alternative strategies, such as physical prompts, these procedures are unnatural and invasive, and may lead to increased meal durations and levels of refusal. The purpose of the current study was to increase spoon acceptance through the use of finger-to-spoon fading when escape extinction (EE) and differential reinforcement of alternative behavior (DRA) were ineffective. We used a reversal design with probe reversals to baseline following each fading step. The study included one 4-year-old male diagnosed with autism spectrum disorder and severe food selectivity who consumed nonpreferred foods with behavioral treatment, but required an invasive procedure (i.e. finger prompt and side deposit) in order to accept each bite because he would not accept any bites from a spoon. Prior to finger-to-spoon fading, spoon acceptance was 0% during baseline with EE and DRA, and increased to 100% following the intervention. These results indicate that using finger-to-spoon fading may be an effective intervention to increase spoon acceptance when other commonly used interventions, such as EE and DRA, are ineffective.

This study evaluated the use of a differential reinforcement of alternative behavior without extinction procedure to treat food refusal and oppositional behavior for an 8-year-old boy being treated through a combination of pharmacological and behavioral interventions. For this individual, we hypothesized escape from food and escape from instructions maintained refusal behavior. In addition, potential side effects due to psychotropic medications may have decreased appetite (Barkley, McMurray, Edelbrock, & Robbins, 1990). Specific reinforcement contingencies included negative reinforcement (i.e., removal of feeding context) and positive reinforcement (i.e., access to highly preferred activity). In baseline, the child consumed near zero levels of food. In treatment, completion of the response requirement resulted in access to a highly preferred activity (e.g., playground). Initial implementation of treatment procedures resulted in consistent consumption of the requisite response requirement (e.g., 10 g.). The child's food consumption was systematically increased to an age-appropriate serving size (e.g., 300 g.) using a changing criterion design. This study provides one example of the use of differential reinforcement of alternative behavior without extinction procedures to reduce food refusal behavior. These results offer an approach to treatment of food refusal, without the use of escape extinction.

The failure to maintain clinically significant reductions in problem behavior under prevailing treatment conditions is a recurring challenge for behavior analysts. Lerman et al. (1994) examined contributing factors for treatment relapse and suggested that degradation of treatment efficacy may be attributed to a change in function for problem behavior over time. To evaluate this hypothesis, functional analysis (FA) outcomes were examined for 61 cases readmitted to an inpatient unit for severe problem behavior. Participants were ages 2-24yrs; 68% were diagnosed with autism spectrum disorder. Time between admissions ranged from nine months to nine years. An exact match in function of problem behavior across admissions occurred in only 18% of cases. A partial function match (i.e., at least one correspondence and one noncorrespondence of function across admissions) occurred in 36% of cases, and a different function for problem behavior occurred in 46% of cases. Thus, 82% of cases had at least one new function for problem behavior identified at readmission. Finally, 75% of cases had either problem behavior maintained by automatic reinforcement, or undifferentiated FA outcomes. Results suggest that treatment relapse may occur with change in behavioral function, but may also be due to treatment challenges with specified behavioral functions.