AI2 Research Laboratory

To achieve our mission, the AI2 Research Laboratory builds an interdisciplinary and cross-institutional effort that unites experts in learning sciences, computer sciences, STEM educators and literacy researchers from multiple institutions. We pursue answers to two critical questions in education: (a) how can we personalize and advance learning experiences supported by emerging technologies such as AI and augmented reality? and (b) how can we design highly accessible learner experiences using learning technologies that deepen learner engagement? AI2 is led by Min Kyu Kim in partnership with colleagues across departments and colleges and his doctoral advisees.

Disciplinary Comprehension Lab

Researchers in the lab use quantitative, qualitative and computational approaches to investigate the cognitive processes that occur during reading comprehension and the types of activities and interventions that best support learning from text. The Disciplinary Comprehension Lab is led by Katie McCarthy, Ph.D.

Early Social Skills Curriculum Lab

The ESSC lab, led by Sarah Hansen, Ph.D., conducts research on early social communication skills for children with, and at risk for, developmental disabilities, specifically autism spectrum disorder. Our research focuses on teaching early pivotal skills like joint attention, play and communication. We also work to train natural change agents who may be parents, teachers, siblings and peers to implement interventions in everyday settings such as inclusive preschool classrooms and family homes.

Literacy Instruction Based on Evidence through Research for Adjudicated Teens to Excel (LIBERATE)

David Houchins, Ph.D. has received a four-year, $3.2 million grant from the National Center for Special Education Research to study a blended learning literacy program in juvenile justice schools.

Microcredentials for Integrating Computing Responsibly into Other Domains (MICRO)

Lauren Margulieux, Ph.D. and Brendan Calandra, Ph.D. received funding from the National Science Foundation called Micro-credentials for Integrating Computing Responsibly into Other Domains (MICRO). The team has developed and is testing an online professional development model for helping K-12 teachers in multiple subject areas to incorporate computing into their existing lessons.

Behavioral Early Education Scholars (BEES)

Chris Tullis, Ph.D., Sarah Hansen, Ph.D., and Claire Donehower, Ph.D. received support from the U.S. Department of Education to provide a specialized, fully-funded, master’s-level training experience in applied behavior analysis (ABA) and early childhood special education (ECSE) called Project Behavioral Early Education Scholars.

AI-Enhanced Scaffolding System (AISS)

Developed by Min Kyu Kim and his team, AISS is a web-based system that helps students write evidence-supported expository essays. AISS provides scaffolding (e.g., text completion examples as expert models) that supports students to improve information literacy, argumentation skills and writing skills in compliance with an academic writing style.

Interaction Geography Slicer (IGS)

Developed by Ben Shapiro, Ph.D. the IGS is an open-source, dynamic visualization tool that provides novel ways to visualize movement, conversation and audio/video data over space and time. This tool is designed to work with movement data collected either through location-based technologies or manually through Mondrian Transcription Software.

Learning How to Look & Listen

Ben Shapiro and colleagues developed this website that brings together resources from a conference supported by the Spencer Foundation at Arizona State University where an interdisciplinary group of older and younger scholars gathered to document and illustrate the basic patterns of visual and auditory attention that are employed by researchers who use video to study social interaction.

Mondrian Transcription

Developed by Ben Shapiro and colleagues, Mondrian Transcription Software is an open-source tool to transcribe fine-grained movement data from video by efficiently tracing the movement of people or things over floor plan representations to generate text files of positioning data — essentially, a transcript of movement. These files can be visualized in software such as the Interaction Geography Slicer. While there are increasingly many automated methods to interpret video or collect positioning data through location-based technologies, these methods continue to be imprecise and background the value of manual transcription practices that many qualitative researchers value. This page will get you started with how to use and do transcription with this tool. The more you use this tool, the easier and more fluid transcribing movement will feel.

Multi-Perspective Video Editor for Interaction Analysis (in progress)

Multiple-choice Online Cloze Comprehension Assessment (MOCCA)

Developed by Sarah Carlson, Ph.D. and colleagues, MOCCA is an assessment that identifies why students comprehend text poorly. MOCCA uses informative “distractors” to understand how students are approaching comprehending texts as they read. This information can help educators target instruction to best support students in improving their reading comprehension. Different versions of MOCCA are available for students in Grades 3-6 and in college. The Grade 3-6 MOCCA also has a computer-adaptive version that reduces testing time.

Mapping Self in Society (MaSelfS)

This website brings together free, open-source tools for gathering data on personal mobility, structuring these data to explore the “daily round” of personal experience at a community scale, and visualizing personal geography over time and thematic maps created with larger data sets in these communities. Sections of the website are organized as a sequence of five (5) activities that are summarized below. Concepts and tools introduced in each activity can be used separately or recombined to create novel forms of teaching or research. This work is a collaboration between researchers and teachers at Georgia State University and Vanderbilt University and is generously supported by the Spencer Foundation and the National Science Foundation.

Student Mental Model Analyzer for Research and Teaching Support (SMART)

Developed by Min Kyu Kim, SMART is a formative assessment and feedback system that analyzes students’ academic essays (e.g., summaries of a text) and provides feedback to help learners to build a solid knowledge structure of a complex problem situation or reading materials. SMART analyzes students’ mental models in three dimensions (surface, structure and semantic).