The evidence
But does it really help students learn?
We teamed up with the departments of Mathematics and Computer Science at Dartmouth College to evaluate Phosphor's impact on student achievement. Our results indicate that Phosphor contributed to students' growth with an effect size of 0.71–1.30 standard deviations—sitting among the largest effect sizes for a study of this kind.
We published a peer-reviewed paper at the 7th International Workshop on Intelligent Textbooks in Seoul, South Korea in 2026. Full dosage of the Phosphor material is associated with an increase in exam performance of 8.0–14.7 percentage points. You can read the paper here.
Critically, we know that even the highest-quality textbooks are prone to the widespread problem of student reading noncompliance (Burchfield & Sappington, 2000): students rarely read the assigned text. That's why we designed Phosphor collaboratively with both students and faculty. In our study, reading compliance before Phosphor hovered between 10% and 15%. With Phosphor, over 90% of enrolled students engaged with the text—despite it being completely optional—with the median student completing 96% of course content.
Why does Phosphor work?
Recent research also shows that when students rely on ChatGPT, Claude, or other general-purpose AI tools, they retain even less than before (Bastani et al., 2025). We knew that in order to improve learning outcomes at scale, we needed to make a platform that students would rather use than talking to external AI. We had never heard of a platform students love—they rarely purchase print textbooks, detest scrolling through PDFs, and reject ages-old interactive platforms outright. So we tried something new.
We built Phosphor hand-in-hand with the students and faculty who depend on it. This way, we earn engagement rather than demanding it—and skyrocket learning outcomes along the way.
We've implemented the leading research from the learning sciences into every interaction students have with the text:
Constructed response over mere recognition
Phosphor emphasizes constructed response questions, in which students answer in their own words rather than picking from a list of options. Producing an answer from scratch—rather than recognizing a correct one—demands the effortful recall that builds genuine, durable understanding (Roediger & Karpicke, 2006).
Immediate formative feedback
Every response gets instant, personalized feedback while the concept is still fresh, so misconceptions are caught and corrected in the moment, not on a graded exam a week later (Hattie & Timperley, 2007).
Encouragement, not intimidation
Phosphor meets students where they are and keeps them going, treating a wrong answer as a normal step toward mastery rather than a verdict. Sustained, low-stakes engagement—not fear of failure—is what moves outcomes (Ryan & Deci, 2000).