Diane Litman
University of Pittsburgh, PA·Computer Science
How can computers that understand language help students learn better, without adding work for teachers? Students in large lecture courses and in writing classes often struggle to reflect on what confuses them or to act on feedback they receive.
- Fields
Computer Science
Education Research
- Topics
natural language processing
artificial intelligence in education
educational technology
- Grants
- 2 known grants, through Aug 2028
What they work on
- Design and refine adaptive prompts, motivational nudges and automated reflection summaries in CourseMIRROR, a mobile reflection app, through iterative usability and feasibility testing.
- Test in field experiments across multiple introductory STEM courses and institutions how CourseMIRROR's features affect student motivation, emotion, self-regulated learning and course performance.
- Determine whether gains in self-regulated learning explain achievement gains, using questionnaire scores, app logs, interviews and grades analyzed with multilevel and structural-equation mediation models.
- Develop reliable, valid measures of how well students revise the use of evidence in argument essays after receiving feedback, and automate their scoring with natural language processing models.
- Provide students with automated feedback on their essay revisions and evaluate in classrooms whether it improves revision and writing, comparing NLP models for reliability, transparency and fairness.
More about the research
Both projects build and test tools that use natural language processing (NLP), the branch of artificial intelligence that lets computers analyze human text, to give students timely, personalized support. One project targets reflection in big STEM lectures. The other targets how students revise argument essays.
Why it matters
These tools could help students in crowded introductory STEM courses and in K–12 or college writing classes get personalized feedback without extra grading for instructors. Better reflection and revision skills may improve course performance and retention in science and engineering. They may also prepare students for the writing their future schooling and jobs demand. Instructors would gain class-wide insight into what confuses students.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Collaborative Research: Harnessing Student Reflections with AI-Driven Personalization in Large STEM Lectures
NSF · Sep 2025 – Aug 2028
Collaborative Research: Development of Natural Language Processing Techniques to Improve Students' Revision of Evidence Use in Argument Writing
NSF · Jun 2022 – May 2027
Closest labs
Sidney D'Mello
University of Colorado Boulder, CO·Education Research
Co-design and iteratively refine automated feedback interfaces for English and language arts teachers with a diverse teacher panel, in partnership with TeachFX, a provider of technology-enabled instructional feedback.
Glen Miller
Texas A&M University-College Station, TX·Education Research
Catalog current and planned AI ethics education interventions across diverse U.S. colleges, including community colleges and minority-serving institutions, through a national survey.
Elizabeth Dvorsky
University of Puerto Rico-Rio Piedras, PR·Education Research
Design a blueprint, including curriculum structure and system architecture, for a gamified virtual experience that teaches scientific research skills to Spanish-speaking STEM undergraduates.
The school
University of Pittsburgh: 83 labs on PI Pages, $1.5B a year on research, 403 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.