Identifying Precursors to Severe Problem Behavior

The primary objective of this research is to evaluate the escalating sequence of behavior that culminates in severe problem behavior (SPB, such as aggression, self-injury, and disruption). To do so, we will investigate the precursors of SPB, which are defined as behaviors that are not as severe or dangerous as SPB, but that reliably precede the occurrence of SPB. Identifying and knowing about these precursors has the potential to complement existing behavioral treatments, making treatments safer (because SPB may occur to a lesser extent or not at all). However, existing research on precursors has been opportunistic and lacks systematic study. This research will examine methods to identify and better understand precursors.

The primary objective of this research is to evaluate the escalating sequence of behavior that culminates in severe problem behavior (SPB, such as aggression, self-injury, and disruption). To do so, the investigators will investigate the precursors of SPB, which are defined as behaviors that are not as severe or dangerous as SPB, but that reliably precede the occurrence of SPB. Identifying and knowing about these precursors has the potential to complement existing behavioral treatments, making treatments safer (because SPB may occur to a lesser extent or not at all). However, existing research on precursors has been opportunistic and lacks systematic study. This research will examine methods to identify and better understand precursors. In Aim 1, researchers will conduct second-by-second review of videos when participants engage in SPB and identify behaviors that reliably precede SPB. The research team will then conduct quantitative analysis of these behaviors that precede SPB to identify likely precursors. In doing so, researchers will also identify the environmental factors that give rise to and maintain SPB (the reinforcer). In Aim 2, the investigators will experimentally confirm that the likely precursors identified in Aim 1 are precursors to SPB by examining how both behaviors change across two different conditions that researchers arrange for the participants to experience. First, investigators will examine these behaviors in a contingent reinforcement condition, in which the investigators provide reinforcer for each instance of the precursor. Second, researchers will examine these behavior, in a noncontingent reinforcement condition, in which the researchers provide the reinforcer on a time based schedule. A behavior will be identified as a precursor if it occurs only in the same context as SPB (i.e., the contingent reinforcement condition) and not in other conditions (i.e., the noncontingent reinforcement condition). AIM 3: The research team will identify whether the precursor is a member of a response class (two behaviors that have same goal) or whether the precursor is part of a behavioral chain (a sequence of behaviors where each is required before the next can occur) culminating in SPB by examining how SPB changes when investigators block the occurrence of the precursor (i.e., physically prevent the precursor from occurring) but provide the reinforce for SPB. If SPB occurs when the precursor is blocked, researchers conclude the precursor is part of a response class. If the SPB does not occur when the precursor is blocked, the investigators will conclude the precursor is part of a response chain. Each recruited individual will participate in Aim 1. Based on our pilot data and the published literature, the researchers hypothesize that approximately 80% of individuals will have a behavior that occurs frequently before SPB, relative to how often this behavior occurs in general. Based on pilot data and published literature, the researchers hypothesize these likely precursors will be verified to be a precursor to SPB 80% of the time in Aim 2. For all individuals where a precursor verified in Aim 2, investigators will block the occurrence of this behavior (physically prevent this behavior from occurring) and assess the effects on SPB.

A consecutive controlled case series design will be used across the three proposed studies. In this design, all individuals who meet the inclusion criteria will experience the procedures described in that study. Across studies, within-subject experimental designs with repeated measures will be used to examine the effects of the independent variables (environmental conditions) on the dependent variables (responding). Within-subject experimental designs are well suited for this type of research because the individual's SPB is used as his or her own control. Well-validated structural criteria will be employed (Hagopian et al., 1997; Roane et al., 2013) in these studies to determine when a series of conditions has a sufficient sample of behavior. The Study Team members have published numerous treatment evaluations employing this methodology (see Hagopian et al., 2011; Rooker et al., 2013).

The precursor behavior occurs as a part of a sequence, or for the same maintaining variable, leading up to the severe problem behavior.

Data on topographies of SPB (e.g., aggression, disruptions) will be collected in real time using direct observation of the behavior. Data will be collected on the frequency of problem behavior which will be transformed into responses per minute. Data will be collected during baseline (no blocking) and treatment phase (blocking). The outcome measure for SPB will be the percentage change during treatment relative to baseline.

Data on topographies of precursors (behavior defined in Studies 1 & 2) will be collected in real time using direct observation of the behavior. Data will be collected on the frequency of precursors which will be transformed into responses per minute. Data will be collected during baseline (no blocking) and treatment phase (blocking). The outcome measure for precursors will be proportional change in attempts to engage in precursor behavior.

Inclusion Criteria: Children aged 6 -17 Individuals with intellectual and developmental disabilities Individuals who engage in severe problem behavior (e.g., aggression or self injury) Exclusion Criteria: No exclusion criteria

Hagopian LP, Rooker GW, Rolider NU. Identifying empirically supported treatments for pica in individuals with intellectual disabilities. Res Dev Disabil. 2011 Nov-Dec;32(6):2114-20. doi: 10.1016/j.ridd.2011.07.042. Epub 2011 Sep 8.

Hagopian LP, Fisher WW, Thompson RH, Owen-DeSchryver J, Iwata BA, Wacker DP. Toward the development of structured criteria for interpretation of functional analysis data. J Appl Behav Anal. 1997 Summer;30(2):313-25; quiz 326. doi: 10.1901/jaba.1997.30-313.

Rooker GW, Jessel J, Kurtz PF, Hagopian LP. Functional communication training with and without alternative reinforcement and punishment: an analysis of 58 applications. J Appl Behav Anal. 2013 Dec;46(4):708-22. doi: 10.1002/jaba.76. Epub 2013 Aug 22.

Roane HS, Fisher WW, Kelley ME, Mevers JL, Bouxsein KJ. Using modified visual-inspection criteria to interpret functional analysis outcomes. J Appl Behav Anal. 2013 Spring;46(1):130-46. doi: 10.1002/jaba.13.