Recognizing the long-standing gap between research and classroom practice in special education, researchers at the UNC Frank Porter Graham Child Development Institute (FPG) have launched “Integrating AI and Evidence-Based Practices to Enhance Postsecondary Special Educators’ Preparation for Supporting Autistic Students.” This project draws on the strengths of artificial intelligence (AI) to make it easier for educators to identify evidence-based practices (EBPs) that are most effective for individual autistic students’ goals. By increasing access to information about EBPs, the project aims to prepare current and future teachers to design and implement a personalized system of instruction for autistic students. 

The project, which began on January 1, 2026, and is scheduled to conclude on December 31, 2029, is supported by an approximately $3.6 million funding award from the US Department of Education’s Office of Postsecondary Education. Ann Sam, PhD, and Samuel Odom, PhD, senior scientists at FPG, along with FPG Director Brian Boyd, PhD, serve as co-principal investigators. Key members of the team at UNC include FPG faculty fellows Kara Hume, PhD, an associate professor in the School of Education, and Jessica Steinbrenner, PhD, an associate professor in the Department of Health Sciences, who were former co-directors of the National Clearinghouse on Autism Evidence and Practice and will contribute extensive expertise in evidence-based practices for autistic students. Anita Crescenzi, PhD, an assistant professor in the School of Data Science and Society, will contribute specialized expertise in how users interact with information access systems, including generative AI.

To successfully carry out this large-scale project, the team is collaborating with nationally recognized leaders across research, higher education, and state systems. Key research partners include Kevin Chen-Chuan Chang, PhD, professor in computer science at the University of Illinois at Urbana-Champaign, and Jinjun Xiong, PhD, dean of the College of AI, Cyber and Computing at the University of Texas–San Antonio. Together, they bring expertise in artificial intelligence to support the systematic review, meta-analysis, and development of an IEP2EBP tool, a teacher-utilized program that will link the most effective EBPs to autistic students’ individual educational goals based on the student’s characteristics and learning needs. The team also includes Interim Associate Dean for Research and Professor at the University of Iowa Ariel Aloe, PhD, whose expertise in advanced meta-analytic methods strengthens the rigor of the project’s evidence synthesis. 

The new initiative builds directly on the FPG team’s prior national efforts that identified EBPs, including the original National Professional Development Center on Autism (initially funded in 2007), the National Clearinghouse on Autism Evidence and Practice, and the Autism Focused Intervention Resources and Modules (AFIRM), which emphasized translating and disseminating evidence-based practices to educators. AFIRM-AI is the working title of the new project, reflecting an expanded and more in-depth review process and the integration of new AI-driven features designed to advance and strengthen existing work.

The project will result in three primary products: a comprehensive review and meta-analysis of the autism intervention literature that will identify efficacious practices; the IEP2EBP tool; and a process for preparing current and future teachers to use the IEP2EBP system to provide an evidence-based system of instruction.

During the first two years of the project, team members will focus on completing an AI-assisted meta-analysis of autism intervention research and designing and evaluating the IEP2EBP tool. “To build effective educational programs for autistic students, we have to draw from both the history of effective practice and the most current science available,” said Odom.

The team’s previous reviews examined autism interventions published between 1990 and 2017. This project will update this previous work by including articles published from 2018 through 2027. The updated findings will directly inform the IEP2EBP tool, giving educators clearer and more precise guidance for matching evidence-based teaching strategies to individual student goals and needs, which will help make instruction more personalized and effective than ever before.

“By grounding guidance in strong research evidence, this AI-supported tool helps prepare educators to confidently select practices that align with a student’s individualized education program (IEP) goal,” said Sam. “It gives teachers a clear pathway from an IEP goal to a personalized instructional approach they can use effectively in real classrooms.”

“By grounding guidance in strong research evidence, this AI-supported tool helps prepare educators to confidently select practices that align with a student’s individualized education program (IEP) goal,” said Sam. “It gives teachers a clear pathway from an IEP goal to a personalized instructional approach they can use effectively in real classrooms.”

The development of the IEP2EBP program will employ a participatory design approach that actively involves educators throughout. Through surveys, focus groups, and usability testing, intended users will try out early versions of the tool and provide feedback during the iterative design process. The final IEP2EBP tool will allow educators to enter a student’s IEP goal along with key information about the student, such as strengths, needs, preferences, what has or has not worked well in the past, and available resources, to receive tailored guidance for selecting and using evidence-based practices and evaluating students’ goal attainment.

This developmental approach, informed by the fields of education science, improvement science, and human computer interaction, ensures the final product is practical, user-friendly, and responsive to the information needs of educators. Using similar methods, the team will also develop supporting resources for postsecondary faculty and local and state educational agency trainers to help current and future educators effectively use the IEP2EBP tool.

In the final year of the project, the team will conduct a comprehensive evaluation of the IEP2EBP tool, and the accompanying resources developed for postsecondary education faculty and local and state agency trainers. Local and state agency trainers will deliver the IEP2EBP program to current educators, while post-secondary teacher preparation programs will deliver the IEP2EBP program to future educators.

The evaluation will examine how the training is delivered and used, the usability of the tool, changes in teacher efficacy, and educators’ intentions to implement the practices in real-world settings. A national consortium of faculty from teacher preparation programs and leadership from state-level partners in California, Mississippi, and Missouri will collaborate in these implementation and evaluation activities.  

“We know that the demands placed on special education teachers have increased for many reasons, including teacher shortages,” says Boyd. “This project is an example of how AI and related technologies can be used to better prepare the teaching workforce, improve the efficiency and quality of teaching practices, and ultimately promote more positive outcomes for the autistic students they’re serving.”

“We know that the demands placed on special education teachers have increased for many reasons, including teacher shortages,” says Boyd. “This project is an example of how AI and related technologies can be used to better prepare the teaching workforce, improve the efficiency and quality of teaching practices, and ultimately promote more positive outcomes for the autistic students they’re serving.”

---

One hundred percent of this project, Integrating AI and Evidence-Based Practices to Enhance Postsecondary Special Educators' Preparation for Supporting Autistic Students, will be funded by federal funds. A funding amount of $3,669,980 was awarded to the project by the US Department of Education – Office of Postsecondary Education.