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.


50th Annual Convention; Philadelphia, PA; 2024

Event Details

Previous Page


Paper Session #356
CE Offered: BACB
Data Analysis in Applied Behavior Analysis (ABA)
Sunday, May 26, 2024
6:00 PM–6:50 PM
Convention Center, 200 Level, 203 AB
Area: PCH
Instruction Level: Advanced
Chair: Carlos Rafael Fernandes Picanço (Universidade Federal de São Carlos)
CE Instructor: Jenny Foster, M.S.

Description, Prediction, and Control: Enhancing Scientific Understanding of Behavior by Integrating Novel Data and Sensor Technologies

Domain: Theory
JENNY FOSTER (The Center for Discovery), Johanna F Lantz (The Center for Discovery), Tania Villavicencio (The Center for Discovery), Conor Anderson (The Center for Discovery), Jennifer Ferina (Rensselaer Polytechnic Institute), Yash Kiarashi (Emory University), Hyeokhyen Kwon (Emory University), Ali Rad (Emory University)

Single case study design and the visual inspection of data have, for decades, served as pillars for the analysis of behavior and interpretation of applied research data. While these approaches allow for the precise understanding and control of variables affecting individual behavior, emerging technologies in the fields of sensor engineering and data science offer unique opportunities for behavior analysts to broaden our view. Sensing technology can automate data collection, and in some cases, give access to otherwise invisible data (e.g., heart rate, skin conductance). Machine learning algorithms and modeling can also shed light on the interaction between multiple variables and subsequent behavior, and even help predict the likelihood of future behavior. Successful utilization and integration of these technologies, however, requires careful collaboration between BCBAs and data scientists and biomedical engineers. The Center for Discovery is an educational and residential facility, serving individuals with complex autism and other developmental disabilities. Here we present several successful collaborative projects with researchers from Georgia Tech, Emory University, and Rensselaer Polytechnic Institute that have allowed us to capitalize on sensing technologies and machine learning to: enhance data collection and analysis, increase our understanding of setting events, and inform clinical judgement regarding treatment.

A Free Software and Open Database for Single-Model and Multimodel Matching-to-Sample
Domain: Theory
CARLOS RAFAEL FERNANDES PICANÇO (Universidade Federal de São Carlos), Fellipe Castro (Universidade Federal do Pará), João da Costa (Universidade Federal do Pará), Nicole Sauma Bentes Freitas (Universidade Federal do Pará), Olavo Galvao (Federal University of Para State), Elenice Seixas Hanna (Universidade de Brasilia), Deisy das Graças De Souza (Universidade Federal de São Carlos; Instituto Nacional de Ciência e Tecnologia sobre Comportamento Cognição e Ensino)
Abstract: Behavior analysis faces some challenges in conducting and replicating research, such as, at least in part, the lack of accessible and reliable software and materials. This paper proposes two technical solutions to address these challenges: a free software for matching-to-sample (MTS) presentation and a media database for an artificially created language with 256 cvcv pseudowords with human and robot recorded audio, a font file for text presentation, and high quality no-background abstract images. The free software code base has approximately 25 years of collaborative work and was adapted to present simultaneous and multi model MTS formats with auditory-to-visual, visual-to-visual, visual-to-auditory, auditory-to-speech, and visual-to-speech modalities. We present two use-cases for these tools: (1) to study generalized recombinative reading with simultaneous MTS and (2) equivalence class formation with multi-model MTS. These tools can be used to design, implement, and simulate behavioral experiments in both cases. We provide some examples of simulated data and results that illustrate the potential of these tools for basic behavioral research. These tools can facilitate the access and dissemination of basic behavioral research in the mentioned topics, and promote open science and collaboration, and enhance the validity, reliability, and applicability of behavioral knowledge.



Back to Top
Modifed by Eddie Soh