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.


46th Annual Convention; Washington DC; 2020

Event Details

Previous Page


Symposium #441
CE Offered: BACB
To Vary or not to Vary: Advances in Behavioral Variability Research
Monday, May 25, 2020
8:00 AM–9:50 AM
Walter E. Washington Convention Center, Level 1, Salon G
Area: DDA/EAB; Domain: Translational
Chair: Clodagh Mary Murray (Emirates College for Advanced Education)
Discussant: Allen Neuringer (Reed College)
CE Instructor: Clodagh Mary Murray, Ph.D.

This symposium presents some of the latest developments in behavioral variability research. The opening paper outlines a comparison of video-modelling (VM) and VM plus lag schedules to increase variability of intraverbal responses to social questions among adults with autism spectrum disorder (ASD). Issues with generalization and maintenance of intervention outcomes and how to measure same will be discussed. The next study used lag schedules to increase variability in toy selections of children with ASD. There was an associated increase in appropriate play and decrease in stereotypy, while multiple generalization probes revealed generalization across people and settings but not stimuli. Next is a long-awaited investigation into the level of variability in activities of neurotypical preschoolers that details a novel measurement system. This study provides important information for those working on increasing variability in play and other activities in children with developmental disabilities; these are increasingly frequent topics of applied variability research. The final paper describes a basic study investigating generalization of reinforced variability in rats. In the applied field, researchers are striving to understand generalization effects of variation across multiple repertoires; this study adds to our understanding of these processes. This set of papers, together with our Discussant, the leading expert in this field, aims to address some of the many interesting questions that have been raised as more and more researchers identify behavioral variability as an important field of study.

Instruction Level: Intermediate
Keyword(s): Generalized Variability, Lag Schedules, Measuring Variability, Reinforced Variability
Target Audience:

BCBAs, BCBA-Ds, anyone interested in variability in basic or applied fields.

Learning Objectives: At the conclusion of the symposium, attendees will be able to: 1. Explain the importance of variability as a dimension of behavior, particularly when working with people with developmental disabilities 2. Describe the factors associated with generalization of reinforced variability 3. Describe two different ways to measure behavioral variability

Increasing Variability of Intraverbal Responses to Social Questions in Adults With Autism Spectrum Disorder

(Applied Research)
Aibhin O'Neill (National University of Ireland Galway), CLODAGH MARY MURRAY (Emirates College for Advanced Education)

This study investigated the effects of video modelling alone (VM) and video modelling plus a lag 2 schedule of reinforcement (VM + lag 2) on the variability of intraverbal responses to social questions for two adults with autism. It used an alternating treatments design with embedded multiple baseline. The questions ‘How are you?’, ‘What do you like to do?’, and ‘Can you tell me something about you?’ were targeted in this study, with different questions being randomly allocated to either VM, VM + lag or best treatment conditions. Variability in responding was measured by calculating, for each session, the mean number of responses from which each one differed. For both participants, variability was higher in the VM + lag 2 condition, hence, this was used in the best treatment condition. Novel responses emerged in both treatment conditions so, while variability did not increase in the VM alone condition, participants did learn new responses from the video models. A one-month maintenance check revealed that responses to questions targeted with VM+ lag 2 were more variable than those targeted by VM alone. During generalization sessions in novel settings, variability was concordant with that observed in the training location for all questions.

Using Lag Schedules to Increase Variability in Toy Selections of Children with Autism Spectrum Disorder
(Applied Research)
Megan Davis (National University of Ireland Galway), KATHERINE MARISSA CLARKE (National University of Ireland Galway), Clodagh Mary Murray (Emirates College for Advanced Education)
Abstract: Children with Autism Spectrum Disorder (ASD) have been shown to have lower variability in their play-related behaviors than their neurotypical peers and one way this is evident is in the limited number of toys they engage with. A multiple-baseline design across participants was used to examine the impact of lag 2 and lag 3 schedules on toy-selection variability and on stereotypical behaviors during play. Results showed that all three participants displayed an increase in toy-selection variability in addition to a decrease in stereotypical behaviors and an increase in appropriate play behaviors when interacting with the toys. Maintenance checks after 1-month showed that the increase in variability was maintained by all three participants while levels of stereotypical behaviors remained low. A series of generalization probes revealed that the skills generalized across settings and staff members and, to a far lesser extent, to novel toys, though this varied by session, according to the type of toys used. The impact of the type of toys introduced during generalization probes (preferred vs non-preferred) will be discussed along with the implications of this work for researchers and clinicians.
Examining Variability of Item Interaction and Activity Selection in Preschool Classrooms
(Applied Research)
JARED T ARMSHAW (University of North Texas), Joseph D. Dracobly (University of North Texas), Gabriela Arias (University of North Texas)
Abstract: Creativity is a core skill that schools target beginning in early pre-school settings. To have different patterns of responding that represent different ideas, interests, fantasies, etc., one must first have a variety of ideas, interests, fantasies, etc. There is a growing body of research on methods to produce response variability. Despite this promising research, it is not clear, however, how much alternation is socially appropriate. Therefore, the purpose of this study is to develop, across a larger number of children, an understanding of the different ways children interact with activities commonly found in pre-school environments. We measured children’s repeat item interactions, novel item interactions, and allocation of time across five concurrently available activity centers. Both within and across children, there was diversity in the number of items with which children interacted and some interaction between levels of repeat item interaction and levels of novel item interaction. Although the relations were predominantly correlated with centers, there were some differences across children within some activities. We will discuss the implications for our understanding of variability and creativity within school environments, including how everyday arrangements of preschool environments may support or hinder variable responding.
Generalization of Variability Training Across Responses in Rats
(Basic Research)
KAILEY MORRISSEY (Utah State University), Annie Galizio (Utah State University), Jeremy Haynes (Utah State University), Diana Michelle Perez (Utah State University), Caroline Towse (Utah State University), Amy Odum (Utah State University)
Abstract: Research has shown that variability is an operant. If so, reinforced variability should translate across contexts. This study was designed to see if variability training would generalize across response topographies. Phase 1 of this experiment included rats producing four-response patterns across two nosepokes (e.g.,LRLR, where L and R indicate left and right responses). Food was delivered probabilistically during this phase, so variability was not required, and low levels of variability were observed. In Phase 2, one group of rats earned food by producing varied response sequences. The control group was yoked, meaning these rats earned food at the same rate as the experimental rats but did not have to vary. This phase showed high levels of variability for Vary rats, and low levels for Yoke rats. Phase 3 included all rats pressing levers to earn probabilistic rewards, resulting in low levels of variability. In Phase 4, all rats earned food for variable lever pressing. If Vary rats acquire variable lever pressing more quickly than Yoke rats, it is possible that variability is generalized across topographies. If so, the hypothesis that behavioral variability is an operant is supported. The results suggest limited evidence of generalization of variability across responses in rats.



Back to Top
Modifed by Eddie Soh