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.


48th Annual Convention; Boston, MA; 2022

Event Details

Previous Page


Symposium #144
CE Offered: BACB
Feasibility of Wearable Technology: First Steps Towards Automatic Measurement of Challenging Behavior
Saturday, May 28, 2022
5:00 PM–5:50 PM
Meeting Level 2; Room 251
Area: DDA/AUT; Domain: Applied Research
Chair: Mindy Christine Scheithauer (Emory University; Marcus Autism Center)
Discussant: Griffin Rooker (Mount St. Mary's University)
CE Instructor: Mindy Christine Scheithauer, Ph.D.
Abstract: Challenging behaviors, such as aggression, self-injury, and disruption, are generally measured using either direct observation or caregiver-report. Unfortunately, both of these measures have limitations and do not comprehensively measure the impact or severity of behavior. Wearable technology, including devices that include accelerometer, gyroscope, and magnetometer technology, have shown promise in their ability to automatically detect movement and behavior. The purpose of this symposium is to present research on the feasibility of using this technology to automatically detect challenging behavior. The first presentation (Neely) demonstrates the feasibility of this technology with typically developing adults and children, before generalizing findings to children with developmental disabilities during a functional analysis context. The second talk furthers evaluates accelerometer use during functional analyses with 33 children referred for very severe challenging behavior and evaluates the ability of caregivers to apply accelerometers in a home setting (N=10). Results are discussed in the context of importance of feasibility analyses and ways in which results set the stage for future large-scale studies that can evaluate the efficacy of automatically detecting challenging behavior using this technology.
Instruction Level: Intermediate
Keyword(s): Accelerometers, Challenging Behavior, Measurement, Problem Behavior
Target Audience: Attendees should have a basic understanding of the measurement of challenging behavior and limitations of current measurement strategies.
Learning Objectives: 1. Participants will be able to do describe limitations in current measurement systems for challenging behavior. 2. Participants will be able to explain the feasibility of using wearable technology with children with ASD. 3. Participants will be able to describe the benefit of using wearable technology to automatically detect challenging behavior in children with ASD.

Feasibility of Wearable Technology to Supplement Measurement During Functional Analysis

LESLIE NEELY (The University of Texas at San Antonio), Sakiko Oyama (University of Texas at San Antonio), Jordan Wimberley (Autism Treatment Center of San Antonio )

The intensity of a target behavior (severity/magnitude) may correspond to the social significance and priority of behavior for intervention. However, the extent to which researchers measure this dimension is unclear. Inertial measurement units (IMUs) are small portable motion capture systems that incorporate three types of sensors: accelerometer, gyroscope, and magnetometer. The data from the three sensors can be incorporated to calculate inclination angle of the sensor in space, and data from its components (gyroscope and accelerometers) provide data on angular (rotation) velocity and acceleration of the sensors about three axes. By fixing the units on the body segment, the sensor can measure human movement. This study investigates the feasibility of utilizing IMUs to quantify the occurrence and intensity of behavioral events by identifying peak acceleration and angular velocity of the segments. This study contains three experiments. Experiment one investigates the validity of the IMUs to capture common topographies of problem behavior (e.g., hitting, kicking, self-injury) with adults without disabilities. Experiment two investigates the validity of the IMUs to capture common topographies of problem behavior with children without disabilities. Experiment three investigates the validity of IMUs within the context of a functional analysis with two children diagnosed with a developmental disability.


Using Accelerometers With Children With Autism Spectrum Disorder and Challenging Behavior: A Feasibility Analysis

MINDY CHRISTINE SCHEITHAUER (Marcus Autism Center; Emory University), Shruthi Hiremath (Georgia Institute of Technology), Audrey Southerland (Georgia Institute of Technology), Agata Rozga (Georgia Institute of Technology), Thomas Ploetz (School of Interactive Computing, Georgia Institute of Technology), Chelsea Rock (Marcus Autism Center), Nathan Call (Marcus Autism Center; Emory University)

Behaviors such as aggression, self-injury, and disruption are common among individuals with autism spectrum disorder (ASD). Most research on these behaviors has relied on retrospective caregiver-report or direct observation, both of which have limitations. There is preliminary evidence that direct detection of these behaviors using accelerometers is a promising alternative, but additional research is needed to determine the feasibility during actual clinical assessment times and times when a therapist cannot be present for direct observation, as measurement during these times has the most applied significance. This study addressed these gaps by evaluating the feasibility of accelerometer use with children with ASD and severe aggression, self-injury, and disruptive behavior (N=33). We found that most passed habituation procedures (93.94% of participants) intended to promote tolerance with wearing accelerometers and continued to tolerate accelerometers during behavioral assessments (e.g., functional analyses). However, the necessary duration of habituation varied across individuals, an important consideration for planning future studies. Additionally, we identified that it is feasible for caregivers to apply sensors in the home-setting (N=10) but found variable fidelity in data-collection and wear-time duration. This study sets important groundwork for future large-scale studies to automatically detect aggression, self-injury, and disruption.




Back to Top
Modifed by Eddie Soh