This multiple-study experiment evaluated the utility of assessing and treating severe self-injurious behavior SIB based on the outcomes of a functional analysis of precursor behavior. In Study 1, a precursor to SIB was identified using descriptive assessment and conditional probability analyses. In Study 2, a functional analysis of precursor behavior was conducted. Finally, study 3 evaluated the effects of a treatment in which precursor behavior produced the maintaining variable identified in the precursor functional analysis. Studies 1 and 3 were conducted in two settings in the participants natural environment, where data collection was ongoing throughout the course of the study. Results showed that it was possible to identify a precursor to infrequent but severe SIB, that a functional analysis of precursor behavior suggested a clear operant function, and that treatment based on the results of the precursor functional analysis reduced SIB in the natural environment.

We evaluated interrater agreement across multiple respondents on anecdotal assessments and compared cases in which agreement was obtained with outcomes of functional analyses. Experiment 1 evaluated agreement among multiple respondents on the function of problem behavior for 27 individuals across 42 target behaviors using the Motivation Assessment Scale (MAS) and the Questions about Behavioral Function (QABF). Results showed that at least 4 of 5 respondents agreed on the primary maintaining consequence for 52% (22 of 42) of target behaviors with the MAS and 57% (24 of 42) with the QABF. Experiment 2 examined correspondence between the anecdotal assessment results and functional analysis results for 7 individuals for whom at least 4 of 5 respondents showed agreement in Experiment 1. Correspondence with functional analysis results was observed in 6 of 7 cases with the QABF and in 4 of 7 cases with the MAS. Implications of these outcomes for the utility of anecdotal assessments are discussed.

Response variability is sensitive to antecedent and consequent manipulations. Researchers have investigated inducement, direct production through reinforcement, and stimulus control of response variability. Recently, researchers have shown that lag reinforcement schedules reliably increase variability but may also produce higher-order stereotypy. There has been limited investigation of appropriate variability levels and alternation between repetition and variation. In a three-part study, we evaluated levels of variability across a group of children, the effects of various procedures on producing response variability and novelty, and the use of schedule-correlated stimuli for producing rapid alternation between repetition and variation. In Study 1, there was a nearly bimodal distribution of children emitting either low or high variability. In Study 2, for most children, fixed lag 4 and variable lag 4 schedules produced the highest levels of variability and novelty. In Study 3, responding was brought under control of schedule-correlated stimuli, allowing for rapid alternation between repetition and variation.

Clinicians often conduct indirect assessments (IAs; e.g., Durand & Crimmins, 1988; Iwata, DeLeon, & Roscoe, 2013; Matson & Vollmer, 1995) such as questionnaires and interviews with caregivers to gain information about the variables influencing problem behavior. However, researchers have found poor reliability and validity of IAs with respect to determining functional variables. There are numerous variables that might influence the efficacy of IAs as an assessment tool, one of which is the skill set of the person completing the IA. For example, it may be possible to increase the validity and reliability of IAs by having individuals with certain skill sets such as a background in behavior analysis and FBA ("experts") complete them. Thus, the purpose of this study was to compare the reliability (i.e., agreement with respect to function and specific IA questions) and validity (i.e., agreement between the outcome of IAs and a functional analysis) of IAs completed by caregivers and "experts" for each of eight children who emitted problem behavior. We found that experts were more likely than caregivers to agree on IA outcomes with respect to (a) overall interrater agreement, (b) item-by-item agreement, and (c) the highest-rated function(s) of problem behavior. Experts were also more likely to correctly identify the function(s), based on comparisons of the results of the IAs and FAs. In addition, caregivers were more likely to (a) disagree on hypothesized functions and (b) identify multiple incorrect functions. The use of experts for completing IAs could have significant impact on their utility and provide a novel method for more rapidly completing the FBA process and developing a function-based treatment.