The scientific analysis of behavior seeks to demonstrate a functional relationship between an independent variable (IV, i.e., the intervention) and the behavior or dependent variable (DV; Baer, Wolf, & Risely, 1968; Skinner, 1953). Experimentation is designed to demonstrate how the IV impacts or changes the DV; in other words, how does an intervention change behavior. In order to accomplish this goal of determining the relationship between the IV and the DV, single-case research designs (SCD) are most often utilized. One basic, but powerful SCD is the withdrawal or ABAB design. Show Withdrawal designs, also known as ABAB designs, rely on the comparisons between conditions when an intervention is in place and conditions when that intervention is not being implemented. This comparison demonstrates the impact of the IV on the DV (Ledford & Gast, 2018 p. 217). Within these designs, typically an initial baseline condition (A1) is first implemented in which the IV is not in place. Usually baseline recording continues until a stable trend is established. Then the IV or intervention (B1) is put into place, with data again being recorded until a stable trend is displayed. The next condition withdraws the IV and returns to the baseline condition (A2), which if successful in establishing a functional relationship, should show a data pattern similar to the first baseline (A1). The last condition is reinstating the IV (B2), which should be similar to the data in (B1) if a functional relationship is to be demonstrated. A traditional withdrawal design allows for comparisons over time between two baseline and two intervention conditions, though variations with additional A or B conditions can be utilized (e.g., ABABAB). The power of these comparisons is that if responding (i.e., the dependent variable) shows consistent levels during both baseline conditions that are different than consistent responding in the two intervention conditions, then a strong case for the IV being the change of behavior can be made. For example, if the behavior of calling out in class follows the pattern in figure 1 in which baseline responding in A1 and A2 is significantly higher than responding during intervention phases B1 and B2, then we have strong evidence that a functional relationship exists. When the intervention is first put into place (B1) it shows a change in the frequency of the behavior. However, there is the possibility that other changes in the environment occurring at the same time are what caused the decrease in frequency. The withdrawal of the IV (A2) and resulting increase in responding makes the case that the other variables which may or may not still be in place become less likely to be a cause of the change in rate. The second implementation of the IV (B2) and resulting decrease of responding provides further evidence that changes in the DV are primarily due to the implementation of the IV. Due to the consistency of these changes over time, the research design presents a strong case for ruling out the impact of other variables besides the IV. Thus, the withdrawal design can be a highly effective tool to demonstrate that the IV is the variable that is causing the change in frequency and provides evidence that the intervention is effective. Figure 1. Graphical representation of fictitious data to demonstrate the effects of the IV within a withdrawal (ABAB) research design. In applied practice, withdrawal designs can have significant value in verifying the effectiveness of the intervention, so it can be continued to be utilized with confidence. There are, however, practical and ethical concerns which must be considered. The first practical concern is that withdrawal designs are not effective in demonstrating a functional relation if the DV or behavior will continue once learned regardless of the intervention. For example, if the behavior being taught is reading, then once the behavior is learned in condition B1, the natural automatic contingencies of reinforcement (i.e., the enjoyment of reading) are likely to keep the rate of reading behavior high even if the intervention is withdrawn. Thus, when examining behaviors that will receive natural reinforcement outside of the intervention, a withdrawal design is likely not the best option. In addition, the withdrawal of treatment that has begun to demonstrate success can raise both practical and ethical concerns. It may be challenging to convince caregivers and team members of the value in withdrawing the treatment. One way to overcome this challenge is to withdraw for a short duration, though this may impact your ability to demonstrate a functional relation. For a discussion on how to advocate for the use of experimentation in treatment see Caldwell (2019). An even bigger concern is the use of a withdrawal design when the goal of the intervention is to decrease an unsafe behavior (e.g., physical aggression or self-injurious behavior). Withdrawing the treatment even for a brief period of time presents the opportunity for an increase in the unsafe behavior. Withdrawal of the interventions in this situation should be limited to cases in which determining that the IV is the cause of behavior is necessary in order for treatment to be successful. In many cases, if the behavior goes down after the IV is implemented, then that is likely enough evidence to continue the intervention. The use of the withdrawal design can be beneficial in practical decision making. Kleinman and Saigh (2011) utilized a withdrawal design to show the impact of the Good Behavior Game on disruptive behavior among regular high-school students. The data demonstrated a functional relationship with the IV displaying a decrease in disruptive behavior, but with both baseline and withdrawal phases displaying increased levels of disruptive behavior. A withdrawal design was also used to study the rate of correct responding and disruptive behavior among a classroom of students with emotional and behavioral disorders (EBD) when increased opportunities to respond in class were presented (Sutherland, Alder, & Gunter, 2003). The baseline and withdrawal phases showed lower correct responding and higher rates of disruptive behavior than in both intervention phases. A third example can be seen in examining the use of video modeling to teach behaviors related to independent transitions between settings in four elementary age students diagnosed with Autism Spectrum Disorder (Cihak, Fahrenkrog, Ayres, & Smith, 2010). In this study, the withdrawal design demonstrated that when the intervention was withdrawn, independent transitions decreased; however, they increased again when the video-modeling was reintroduced. Thus, the withdrawal design not only verified that the IV was the cause of change, but also that the behavior was not yet sufficiently conditioned to remove the intervention. All three studies show how the withdrawal design can be used to verify the effects of the IV and also how decisions about continuing the intervention can be made. In conclusion, the use of single-case designs to verify treatment effects and to guide decision making in treatment can have significant impact on behavioral programming (Ledford, Barton, Severini, & Zimmerman, 2019; Ledford & Gast, 2018). The withdrawal or ABA design is a powerful tool in confirming the relationship between an IV and DV, as well as determining whether the intervention needs to be or should be continued. Practical application of withdrawal designs should be considered when feasible and ethical to inform ongoing implementation within client treatment. For additional information and description on the use of SCD’s, including withdrawal designs, see (Alnahdi, 2015; Krasny-Pacini & Evans, 2018; Ledford et al., 2019; Ledford & Gast, 2018). References Alnahdi, G. H. (2015). Single-subject designs in special education: Advantages and limitations. Journal of Research in Special Educational Needs, 15, 257-265. doi:10.1111/1471-3802.12039 Baer, D. M., Wolf, M. M., & Risley, T. R. (1968). Some current dimensions of applied behavior analysis. Journal of Applied Behavior Analysis, 1, 91-97. Caldwell, T. D. (2019). Intervention assessment. Retrieved from https://drive.google.com/file/d/1pKN1nw8Q2C_WzVbxS8BTFjpyo1W_ghj4/view Cihak, D., Fahrenkrog, C., Ayres, K. M., & Smith, C. (2010). The use of video modeling via a video iPod and a system of least prompts to improve transitional behaviors for students with autism spectrum disorders in the general education classroom. Journal of Positive Behavior Interventions, 12, 103-115. doi:10.1177/1098300709332346 Kleinman, K. E. & Saigh, P. A. (2011). The effects of the good behavior game on the conduct of regular education New York city high school students. Behavior Modification, 35(1), 95-105. doi:10.1177/0145445510392213 Krasny-Pacini, A. & Evans, J. (2018). Single-case experimental designs to assess intervention effectiveness in rehabilitation: A practical guide. Annals of Physical and Rehabilitative Medicine, 61, 164-179. doi:10.1016/j-rehab.2017.12.002 Ledford, J. R., Barton, E. E., Severini, K. E., & Zimmerman, K. N. (2019). A primer on single-case research designs: Contemporary use and analysis. American Journal of Intellectual and Developmental Disabilities, 124, 35-56. doi:10.1352/1944-7558.124.1.35 Ledford, J. R. & Gast, D. L. (2018). Single case research methodology (3rd ed.). New York, NY: Routledge. Skinner, B. F. (1953). Science and human behavior. New York, NY: The Free Press. Sutherland, K. S., Adler, N., & Gunter, P. L. (2003). The effect of varying rates of opportunities to respond to academic requests on the classroom behavior of students with EBD. Journal of Emotional and Behavioral Disorders, 11, 239-248. Which design is better ABA or ABAB?Some researchers believe ABAB is a stronger design since it has multiple reversals.
What are the advantages of an ABAB design?The advantages of reversal or ABAB design in psychology include seeing whether an intervention actually has an effect, and offering evidence of how people respond without the intervention.
What is the difference between ABA and ABAB?Simply, the ABAB design helps to solve problems of behavior by adding a benefit and then taking it away to return to the baseline to see what changes. The ABA design provides stronger evidence of how a used treatment has an effect.
When would you use an ABAB design?An ABAB research design, also called a withdrawal or reversal design, is used to determine if an intervention is effective in changing the behavior of a participant.
|
