Further evaluation of functional analysis screening methods in early autism intervention.

A goal of some functional analysis (FA) variations is to reduce assessment time while still maintaining efficacy. This may be especially important when conducting FAs in early intervention programs, where time is a crucial commodity. To that end, we evaluated a model for using the results of the no-interaction condition as a screening for behavioral function and to guide selection of FA test conditions with 20 participants (22 assessments) aged 3 to 7 years old. We used the no-interaction condition to develop hypotheses for both automatic reinforcement and socially mediated reinforcement. The outcome of the no-interaction condition guided the selection of test conditions for the remainder of the FA. We also incorporated methods from prior FA studies (e.g., divided attention) to modify the test conditions. We obtained differentiated results in 91% of assessments, all within 70 min and, as such, extended evidence that an FA can be completed in little time without sacrificing efficacy.

Nathan H. Azrin: A Case Study in Research Translation in Behavior Science.

Nathan H. Azrin (1930-2013) contributed extensively to the fields of experimental and applied behavior analysis. His creative and prolific research programs covered a wide range of experimental and applied areas that resulted in 160 articles and several books published over a period of almost 6 decades. As a result, his career illustrates an unparalleled example of translational work in behavior analysis, which has had a major impact not only within our field, but across disciplines and outside academia. In the current article we present a summary of Azrin's wide ranging contributions in the areas of punishment, behavioral engineering, conditioned reinforcement and token economies, feeding disorders, toilet training, overcorrection, habit disorders, in-class behavior, job finding, marital therapy, and substance abuse. In addition, we use scientometric evidence to gain an insight on Azrin's general approach to treatment evaluation and programmatic research. The analysis of Azrin's approach to research, we believe, holds important lessons to behavior analysts today with an interest in the applied and translational sectors of our science. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s40614-020-00278-4.

Within-subject analysis of a prevention strategy for problem behavior.

Although decades of research on functional analysis methodology have identified common contingencies that maintain problem behavior and effective interventions, relatively little research has been conducted on strategies to prevent the initial development of problem behavior. We conducted a 2-part case study, the purposes of which were to illustrate the use of sensitivity tests as the bases for intervention (Study 1) and subsequently to assess the efficacy of a prevention strategy using a single-subject design (Study 2). Results showed that the sensitivity tests identified establishing operations that may set the occasion for the development of problem behavior and that interventions based on differential reinforcement prevented increases in the severity of problem behavior relative to untreated and control baselines. Benefits and limitations to this individualized approach to prevention are discussed.

Comparison of edible and leisure reinforcers.

Results of previous research have shown that individuals with intellectual disabilities usually prefer edible items over leisure items. Other research has shown that sensory (leisure) items facilitate response acquisition and maintenance better than edible items do for individuals with autism. The current studies examined preference and performance for edible and leisure reinforcers by children with and without autism. Results showed that edible items were more preferred (Study 1) and resulted in higher rates of responding under maintenance conditions (Study 3) in subjects both with and without autism. Edible and leisure items resulted in similar rates of response acquisition (Study 2) for both samples and for subjects who showed different patterns of preference in Study 1.

Metrics of Quality in Graduate Training.

This commentary, in response to Dixon et al. (Behavior Analysis in Practice, 8, 7-15, 2015), describes difficulties in defining metrics of quality in graduate training for different audiences (types of applicants). Outcome measures are preferred whenever possible, supplemented by subjective but frequently used opinion surveys.

Parameters of reinforcement and response-class hierarchies.

Shabani, Carr, and Petursdottir (2009) examined the effects of a response-response relation (effort) on the development of a response-class hierarchy using a laboratory model. Response-reinforcer relations may have similar influences. Using a similar translational approach, we examined the effects of reinforcer rate, quality, delay, and magnitude in a series of separate experiments conducted with 8 individuals with intellectual disabilities. Response-class hierarchies emerged along the dimension of rate for 3 of 6 subjects, quality for 5 of 5 subjects, delay for 2 of 8 subjects, and magnitude for 5 of 6 subjects.

Thirty years of research on the functional analysis of problem behavior.

Hanley, Iwata, and McCord (2003) reviewed studies published through 2000 on the functional analysis (FA) of problem behavior. We update that review for 2001 through 2012, including 158 more recent studies that reported data from 445 FAs. Combined with data obtained from Hanley et al., 435 FA studies, with line graphs for 981 FAs, have been published since 1961. We comment on recent trends in FA research and introduce the studies in the 2013 special issue of the Journal of Applied Behavior Analysis.

Effects of fixed versus random condition sequencing during multielement functional analyses.

It has been suggested that a fixed condition sequence might facilitate differential responding during multielement functional analyses (FAs) by capitalizing on or limiting sequence effects (Iwata, Pace, et al., 1994); however, the effects of condition sequence have not been examined empirically. We conducted fixed- and random-sequence FAs for 7 individuals with developmental disabilities to determine the relative effects that sequence may have on assessment outcomes. Experimental conditions during the fixed sequence were conducted in the following order: ignore, attention, play, and demand; condition order during the random sequence was determined randomly. Results showed that sequence had no influence on the FA outcomes for 3 subjects, whereas differential responding emerged either faster (1 subject) or only (3 subjects) under the fixed sequence for the remaining subjects. These results suggest that the fixed sequence, a simple modification, should be used when conducting multielement FAs to accommodate the influence of establishing operations across assessment conditions.

Test-specific control conditions for functional analyses.

Most functional analyses of problem behavior include a common condition (play or noncontingent reinforcement) as a control for both positive and negative reinforcement. However, test-specific conditions that control for each potential source of reinforcement may be beneficial occasionally. We compared responding during alone, ignore, play, and differential reinforcement of other behavior (DRO) control conditions for individuals whose problem behavior was maintained by positive or negative reinforcement. Results showed that all of the conditions were effective controls for problem behavior maintained by positive reinforcement; however, the DRO condition was consistently ineffective as a control for problem behavior maintained by negative reinforcement. Implications for the design of functional analyses and future research are discussed.

Functional analysis screening for problem behavior maintained by automatic reinforcement.

A common finding in previous research is that problem behavior maintained by automatic reinforcement continues to occur in the alone condition of a functional analysis (FA), whereas behavior maintained by social reinforcement typically is extinguished. Thus, the alone condition may represent an efficient screening procedure when maintenance by automatic reinforcement is suspected. We conducted a series of 5-min alone (or no-interaction) probes for 30 cases of problem behavior and compared initial predictions of maintenance or extinction to outcomes obtained in subsequent FAs. Results indicated that data from the screening procedure accurately predicted that problem behavior was maintained by automatic reinforcement in 21 of 22 cases and by social reinforcement in 7 of 8 cases. Thus, results of the screening accurately predicted the function of problem behavior (social vs. automatic reinforcement) in 28 of 30 cases.

Evaluation of the divided attention condition during functional analyses.

A common condition included in most functional analyses (FAs) is the attention condition, in which the therapist ignores the client by engaging in a solitary activity (antecedent event) but delivers attention to the client contingent on problem behavior (consequent event). The divided attention condition is similar, except that the antecedent event consists of the therapist conversing with an adult confederate. We compared the typical and divided attention conditions to determine whether behavior in general (Study 1) and problem behavior in particular (Study 2) were more sensitive to one of the test conditions. Results showed that the divided attention condition resulted in faster acquisition or more efficient FA results for 2 of 9 subjects, suggesting that the divided attention condition could be considered a preferred condition when resources are available.

Therapist and setting influences on functional analysis outcomes.

Functional analyses (FAs) of problem behavior typically are conducted in controlled settings to minimize potential sources of confounding. Several studies have reported that results of FAs conducted in controlled settings occasionally differ from those conducted under more naturalistic conditions, although little is known about factors that may contribute to the different outcomes. We examined correspondence between FAs conducted by staff in a clinic and those conducted either by caregivers as therapists or in the home setting. If results of the 2 analyses were dissimilar, we conducted further analyses to identify variables responsible for the different outcomes. Results showed that, in most cases, correspondence of function was observed across familiar and unfamiliar stimuli. Results are discussed in terms of implications for research and clinical practice for the evaluation of problem behavior.

Experimental analysis of precursors to severe problem behavior.

Some individuals engage in both mild and severe forms of problem behavior. Research has shown that when mild behaviors precede severe behaviors (i.e., the mild behaviors serve as precursors), they can (a) be maintained by the same source of reinforcement as severe behavior and (b) reduce rates of severe behavior observed during assessment. In Study 1, we developed an objective checklist to identify precursors via videotaped trials for 16 subjects who engaged in problem behavior and identified at least 1 precursor for every subject. In Study 2, we conducted separate functional analyses of precursor and severe problem behaviors for 8 subjects, and obtained correspondence between outcomes in 7 cases. In Study 3, we evaluated noncontingent reinforcement schedule thinning plus differential reinforcement of alternative behavior to reduce precursors, increase appropriate behavior, and maintain low rates of severe behavior during 3 treatment analyses for 2 subjects. Results showed that this treatment strategy was effective for behaviors maintained by positive and negative reinforcement.

Assessment and treatment of social avoidance.

Problem behavior maintained by social-negative reinforcement often is evoked by a specific type of social interaction--the presentation of task demands. This study involved assessment and treatment of a more general form of social avoidance in which the establishing operation (EO) for problem behavior consisted of social interaction per se. Four subjects exhibited high rates of problem behavior during the play or demand conditions of a functional analysis (FA). A subsequent FA in which problem behavior produced escape from social interaction confirmed social avoidance for all subjects. A series of interventions aimed at attenuating aversive characteristics of social interaction then was implemented with 3 of the subjects. These interventions included vicarious reinforcement, conditioning of social interaction as a reinforcer, stimulus fading, and differential reinforcement (DRA) plus extinction (EXT). DRA plus EXT was the only condition in which decreases in problem behavior and increases in social interactions were observed reliably.

Assessment and treatment of chronic hand mouthing.

Hand mouthing (HM) is a chronic problem in many individuals with intellectual disabilities. Although the prevalence of mouthing has been estimated, data on the frequency, severity, or functions of the behavior were not included. In Study 1, we examined the prevalence and risk of HM. Results obtained from interviews showed that the prevalence of HM in two institutional samples (N = 802) was 12.7%, whereas direct observation yielded a lower estimate of prevalence (8%). Moreover, a large proportion of observed HM (39.1%) was self-injurious in nature. In Study 2, we used modified functional analyses (FAs) to examine the HM of 64 individuals. Results indicated that maintenance by automatic reinforcement accounted for 98.4% of the cases (all but one case). In Study 3, we implemented a progressive series of interventions for HM exhibited by 14 individuals. The following interventions were implemented in sequential order: (a) noncontingent reinforcement (NCR, effective with 6 subjects), (b) either NCR plus differential reinforcement of alternative behavior (DRA) plus response blocking (effective with 5 subjects) or NCR plus response blocking only (effective with 2 subjects), and (c) NCR plus brief manual restraint (effective with 1 subject).

Does supplementary reinforcement of stereotypy facilitate extinction?

Results of several studies suggest that delivery of supplemental (social) reinforcement for stereotypy might facilitate its subsequent extinction. We examined this possibility with 9 subjects who engaged in stereotypy by including methodological refinements to ensure that (a) subjects' stereotypy was maintained in the absence of social consequences, (b) supplementary reinforcers were highly preferred and were shown to be reinforcers for some behavior, and (c) subjects were exposed to lengthy reinforcement and extinction conditions. In spite of these modifications, only 4 subjects' stereotypy increased when supplementary reinforcement was delivered contingent on stereotypy, and no subject's stereotypy decreased below initial baseline levels when social reinforcement was subsequently withheld. Decreases in stereotypy occurred with the implementation of noncontingent reinforcement. Thus, delivery of supplementary reinforcers either did not increase stereotypy or did not facilitate extinction of stereotypy maintained by automatic reinforcement. We discuss the practical and conceptual bases of these results with respect to our current understanding of function-based interventions.

Reliability and validity of the functional analysis screening tool.

The Functional Analysis Screening Tool (FAST) is a 16-item questionnaire about antecedent and consequent events that might be correlated with the occurrence of problem behavior. Items are organized into 4 functional categories based on contingencies that maintain problem behavior. We assessed interrater reliability of the FAST with 196 problem behaviors through independent administration to pairs of raters (Study 1). Mean item-by-item agreement between pairs of raters was 71.5%. Agreement for individual items ranged from 53.3% to 84.5%. Agreement on FAST outcomes, based on comparison of informants' highest totals, was 64.8%. We assessed the validity of the FAST by comparing its outcomes with results of 69 functional analyses (Study 2). The FAST score predicted the condition of the functional analysis in which the highest rate of problem behavior occurred in 44 cases (63.8%). Potential uses of the FAST in the context of a clinical interview, as well as limitations, are discussed.

Latency of response during the functional analysis of elopement.

Elopement is a dangerous behavior sometimes exhibited by individuals with intellectual disabilities. We conducted trial-based functional analyses in which latency was the index of elopement for 2 students. Two unobtrusive safety monitors were placed outside the building to eliminate potential confounding caused by having to retrieve a student. Results of both students' assessments indicated that elopement served multiple functions. Results are discussed in terms of the study's methodological and applied implications.

Effects of alternative responses on behavior exposed to noncontingent reinforcement.

Noncontingent reinforcement (NCR) may decrease the frequency of behavior by either inducing satiation or terminating the response-reinforcer contingency (extinction). Another possibility is that the target behavior is replaced by other behaviors maintained by preexisting contingencies. We conducted 2 experiments in which we allowed access to a target response and several alternatives. In Experiment 1, NCR, preceded by contingent reinforcement (CR) for the target, produced a reduction in the target and an increase in the alternatives in 2 subjects with intellectual disabilities. To separate the effects of NCR from the availability of alternative responses, we presented CR conditions to 4 subjects in Experiment 2 with and without the availability of alternatives. The availability of alternatives decreased the target in only 1 subject. Subsequent manipulations showed that reductions in the target were solely a function of NCR for the other 3 subjects. Thus, response competition may have marginal effects on response suppression during NCR.

Elimination of position-biased responding in individuals with autism and intellectual disabilities.

Five individuals with autism or other developmental disabilities participated in paired-stimulus preference assessments during repeated baseline probes. All subjects initially showed a pronounced bias by typically selecting the stimulus placed in either the left or right position. Biased responding for 3 subjects was eliminated when training trials were conducted in which a stimulus of known lesser quality was presented as one of the choices. Reinforcer-quality training was unsuccessful for 2 subjects, as was a condition in which reinforcer magnitude was modified to favor unbiased responding. These subjects' biased responding was eliminated only when a correction procedure (repetition of error trials) was implemented.