Carolyn (Carol) Forsyth is a research scientist in foundational research in the ETS Research Institute. She earned a Ph.D. in cognitive psychology with a cognitive science graduate certification from the University of Memphis in 2014. Her research at ETS focuses on creating and evaluating applications of AI for education, including for learning and assessment. She created a methodology for making inferences from log files to update such systems known as theoretically grounded educational data. She applies this methodology to investigate and improve assessment and learning processes and outcomes in interactive simulations and tasks by employing machine-learning algorithms, advanced statistics, and theory in an iterative process. Furthermore, she has applied her expertise to various contexts and constructs, including empathy, collaborative problem-solving, conversation-based assessment, intelligent tutoring systems, discourse processes, and serious games.
Carol intertwines her training as a cognitive scientist with a range of methodologies that span AI, computational linguistics, computer science, statistics, educational data mining (EDM), and cognitive psychology to make inferences about cognition based on actions performed by learners in simulated environments.
Carol has achieved international recognition in the AI and EDM field, with active professional participation in the top conferences in this field: the International Conference on Educational Data Mining and The International Conference on Artificial Intelligence in Education. She has served these communities in various capacities over the past 15 years, including as a presenter, program committee member, and track chair. She has been invited to 15 talks and panelist activities both nationally and internationally on AI and data mining techniques, and has provided 80 peer-reviewed publications and over 100 valued contributions (e.g., invited talks/lectures) to the field of AI and data mining.
Date updated: 1/27/2026