Association for Behavior Analysis International

The Association for Behavior Analysis International® (ABAI) is a nonprofit membership organization with the mission to contribute to the well-being of society by developing, enhancing, and supporting the growth and vitality of the science of behavior analysis through research, education, and practice.


41st Annual Convention; San Antonio, TX; 2015

Event Details

Previous Page


Symposium #23
CE Offered: BACB
Staff Training for Human Service Settings: Analyses and Evaluations
Saturday, May 23, 2015
1:00 PM–2:50 PM
Grand Ballroom C2 (CC)
Area: AUT/OBM; Domain: Applied Research
Chair: Ellie Kazemi (California State University, Northridge)
Discussant: Stephanie M. Peterson (Western Michigan University)
CE Instructor: Ellie Kazemi, Ph.D.
Abstract: Behavior analysts who develop behavioral programming are typically faced with the challenge of effectively training others to implement the programming. Such training involves a number of considerations, including how to structure the training, how to ensure adequate practice of the skill without endangering clients, how to ensure that training will maintain, and ensuring the skills will generalize to novel situations. This symposium will include discussions on each of these areas with talks on a parametric analysis of rehearsal on functional analysis implementation fidelity, an evaluation of the utility of a humanoid robot as a simulated client, and, a pyramidal approach to teach staff effective interaction techniques for older adults with neurocognitive disorder, and using video modeling to teach staff a prompt fading procedure.
Keyword(s): Rehearsal, Simulation, Staff Training
A Parametric Analysis of Rehearsal and Feedback Opportunities during Training of Functional Analysis Conditions
SARAH R. JENKINS (The University of Kansas), Florence D. DiGennaro Reed (University of Kansas)
Abstract: The number of rehearsal with feedback opportunities varies widely in staff training research, ranging from one to 10 in distributed and massed role-plays. Ward-Horner and Sturmey (2012) suggest that rehearsal may be unnecessary to train functional analysis methodology; however, the researchers omitted feedback, which allowed participants to practice errors and does not reflect use of rehearsal in applied settings. Thus, the purpose of this investigation was to conduct a parametric analysis of rehearsal with feedback. We evaluated the effects of massed (one, three, or 10) and distributed rehearsals with feedback within a behavioral skills training package on undergraduate participants’ fidelity of three functional analysis conditions. In general, fidelity was low following instruction and increased with the video model, but participants did not demonstrate mastery in either condition. Participant fidelity increased following massed rehearsals with feedback for a majority of participants; however, most required additional rehearsals to meet criterion. On average, participants required 4.6 rehearsals to meet criterion. Several participants demonstrated high fidelity in the attention condition across nearly all phases of the study suggesting it may be a relatively easier condition to implement accurately. In sum, these findings demonstrate that rehearsal with feedback is beneficial to promote high-fidelity performance.
Pyramidal Training For Supervisors and Caregivers of Aging Adults
ROCKY HAYNES (Southern Illinois University - Carbondale), Jonathan C. Baker (Southern Illinois University), Hannah Ritchie (Southern Illinois University - Carbondale)
Abstract: Alzheimer’s Disease (AD) is more prevalent than any other disease under the umbrella of Neurocognitive Disorder (Alzheimer’s Association, 2013). Certified Nursing Assistants (CNAs) are the typical front-line care staff who care for individuals in aging care (Sengupta, Harris-Kojetin, & Ejaz, 2010). The present study investigated the use of a pyramidal training model to teach aging facility staff to be able to conduct trainings and to teach direct care staff antecedent strategies shown to be effective when communicating with individuals with AD. Pyramidal training resulted in two tiers of staff successfully implementing training for subsequent tiers of staff and subsequent staff demonstrated mastery of the trained material. However, during maintenance observations, some decreases were observed both with regard to training integrity as well as implementation of the trained material.
Using Video Modeling with Voiceover Instruction to Train Staff to Implement a Most-to-Least Prompt Fading Procedure
ANTONIA GIANNAKAKOS (Caldwell College), Jason C. Vladescu (Caldwell College), Sharon A. Reeve (Caldwell College), April N. Kisamore (Caldwell College)
Abstract: Most-to-least (MTL) prompting procedures are often an important component of teaching methods based on the principles of applied behavior analysis. However, only two studies have trained staff to implement MTL (Lerman, Tetreault, Hovanetz, Strobel, & Garro, 2008; Lerman, Vorndran, Addison, & Kuhn, 2004) and both incorporated training procedures that requiered the presence of a staff trainer. In the present study a training modality that did not require the presence of a staff trainer, video modeling with voiceover instruction (VM), was evaluated in training staff trainees at a clinic for individuals with autism to implement a MTL prompt prompt/prompt-fading procedure. Generalization to novel prompt fading procedure was also assessed. Results indicated that VM was effective in teaching participants to use MTL and their skills generalized to a untrained prompt delay procedure.
Further Evidence That a Robot Can Simulate a Client in Staff Training Research
LISA STEDMAN-FALLS (California State University, Northridge), Ellie Kazemi (California State University, Northridge)
Abstract: Researchers can face methodological challenges when they evaluate staff training interventions, as trainee performance often depends on client responses. If clients respond differently within an experiment, trainees may have unequal opportunities to perform target skills. One possible solution may be to use a humanoid robot as a simulated client in training research. In Experiment 1, we used multiple baseline across participants designs to train 5 undergraduate students to implement a paired-stimulus preference (PS) assessment with either a robot (3 participants) or a human simulated client (2 participants). Following training, all participants met our mastery criteria and skills generalized across simulated clients. We conducted Experiment 2 using a multi-element design and 10 participants implemented both the PS and the multiple-stimulus without replacement preference assessments. After participants reached our mastery criteria conducting both assessments (one with the robot and the other with the human simulated client) we probed for generalization across simulated clients and across children. We found that participants performed similarly regardless of the assessment type or the simulated client with whom they worked and that skills generalized to working with children. We conclude that a robot may be an effective simulated client in training research.



Back to Top
Modifed by Eddie Soh