Classroom Application of Functional Analysis
Bloom, S.E., Iwata, B.A., Fritz, J.N., Roscoe, E.M., & Carreau, A.B. (2011). Classroom application of a trial-based functional analysis. Journal of Applied Behavior Analysis, 44(1), 19-31.
Reviewed by: Laurie Brophy, LCSW, BCBA
Why this topic?
Researchers have shown that interventions for problematic behaviors are safest and most effective when selected based on an assessment of the function that the behaviors serve. Therefore, federal regulations now require schools to conduct functional behavior assessments for all students who need an individualized behavior support plan. However, the assessments often consist of indirect measures such as teacher ratings, which are easy to obtain but less useful than a direct observation method called functional analysis (FA) in which the evaluator systematically presents different experimental conditions and records data on the student’s behavior in each of these conditions. Thus, there is significant value in developing FA procedures that are practical to use in naturalistic settings, such as a student’s regular classroom.
What did the researchers do?
Research participants included 10 students (8 boys and 2 girls) who were 6-18 years old and were referred for assessment and treatment of problem behavior at one of two schools for students with developmental disabilities. Student diagnoses included autism (7), Down Syndrome (1), hearing impairment (1), and speech and language delay (1), and their level of intellectual disability was identified as mild (4), moderate (3), severe (2), and profound (1).
The researchers conducted FA using both “trial-based” and “standard” methodologies. The trial-based FAs took place in the student’s regular classrooms and were completed prior to the standard FAs, which were conducted in a designated session room. In both methodologies, experimenters recorded the presence or absence of a problem behavior, time from the start of a trial to the display of a behavior, and data on events that occurred immediately before or after the behavior (antecedents and consequences). In one condition, participants received attention when they displayed problem behavior; in another, participants were presented with demands and allowed to escape or avoid those demands if they displayed problem behavior. A tangible condition, in which participants gained access to a preferred item if they displayed problem behavior, was conducted with participants in which access to tangible was suspected to be a maintaining variable of problem behavior. An “ignore” condition, in which no consequences were given for problem behaviors, was also conducted unless the participant’s problem behavior included aggression.
Trial FA sessions spanned 4 -6 days with 8 to 16 trials conducted per day. Each session was 6 minutes long and was divided into three 2-minute segments: Segment 1 was a control condition in which reinforcers were freely available and problem behavior did not result in any planned consequence; Segment 2 was a test condition in which an establishing operation was generated and problem behavior produced a specific consequence; Segment 3 was a repeat of the control condition in segment 1. The occurrence of problem behavior in any segment resulted in a termination of that condition, with the exception of the “ignore” trials. Ignore trials also differed in that they were not alternated – they consisted of 3 consecutive 2-minute segments with the subject seated alone and no access to leisure/activity materials and no planned consequences for problem behavior. In all conditions, experimenters used materials chosen from participants’ classrooms.
During the standard FAs, the conditions were 15 minutes long and were presented in an alternating sequence. Materials were chosen based on an individual assessment of each participant’s preferences. The methods used were based on original studies by Iwata and, likewise, utilized a multi-element design (i.e., researchers quickly alternate two or more conditions enabling comparison of behavior across conditions). Prior to the FAs, paired stimulus preference assessments (a procedure in which pairs of items are presented systematically and data is collected on the participant’s selections in order to help determine potential reinforcers) were conducted.
What did the researchers find?
For 6 of the 10 participants, the trial-based FA results indicated the same maintaining function(s) as the standard FA procedures. Problem behaviors for these participants appeared to be maintained by escape from demands (2), access to tangible items (1), both escape from demands and access to tangible items (1), or by a combination of factors (2). For the other 4 participants, FA results did not correspond with standard FA results – that is, they only partially agreed or had no agreement. However, when the trial-based FA was modified and re-administered to two of these four participants, the results did correspond with the standard FA results.
What were the strengths and limitations of the study? What do the results mean?
A definite strength of the study is that it included a larger number of participants than most previous FA studies, thereby increasing confidence in the validity and generalizability of the results. Having participants with varying ages, diagnoses and severity is also beneficial for the study because it helps dispel the notion that FAs are primarily for children with autism or with severe to profound behavioral problems.
A limitation is that the trial-based FA methods in the classroom setting still required a significant amount of time and, all components considered, did not shorten the overall assessment. Also, the authors did not comment on the classroom size or identify other setting events or arrangements that may impact replication of a trial-based approach. An important area for further research would be to replicate the trial-based FA methodology in a general education classroom. It would also be of interest to test the utility of this approach with individuals who do not have an intellectual disability.
Although trial-based FA does not save time, it may make FA more practical or acceptable to parents and school personnel because it takes place in the classroom instead of requiring a separate space. This feature may decrease the possibility of missing important antecedent and consequence stimuli that the student encounters in the classroom. Despite the few limitations of the study, it is clear that the results do suggest that direct FA methodologies within the natural classroom environment are possible and can be a viable alternative to standard implementation.