Original title: Dose, Not Access: Digital Reading Tutor Use and Early Reading Growth
Nicholas A. Gage · September 2026
Learning to read is essential, and young kids learn best with direct instruction in phonics, which connects letters to sounds. Human tutoring helps, but it takes lots of time and resources. Researchers tested an artificial-intelligence tool called the Project Read Tutor across four schools. Kids practiced phonics directly on the digital tutor, but how much they actually practiced made all the difference. Most children used it only a little bit. However, students who practiced fifteen minutes or more each week scored much higher on end-of-year reading tests than those who barely used it. Adding just ten minutes of practice each week led to clear gains. The practice helped struggling readers and students with disabilities the most. It boosted scores strongly on the specific letter-sound skills taught by the tutor, and it also improved reading fluency. Merely having the software was not enough; the actual practice time drove growth.
Imagine you have a set of flashcards for learning new words. If you flip through them for just a minute each week, you barely remember any. But if you practice for 15 minutes every week, you start to recognize the words quickly. Now suppose the flashcards are an AI reading tutor, and the practice time is how many minutes a student uses it each week. In this paper, the researchers found that the amount of time spent using the tutor, not just having it available, is what actually improves reading skills.
For schools with limited budgets, this means the key isn't just buying a reading app—it's making sure students actually use it enough. Struggling readers and students with disabilities benefit most, suggesting targeted, high-dosage use could narrow achievement gaps without expensive one-on-one tutoring.
The study is observational, not a randomized experiment, so unmeasured factors (like teacher quality or home support) could influence results. It was conducted in one Indiana district with one specific tutor, so findings may not generalize to other programs or regions. The abstract does not specify how usage was encouraged or whether the tutor was used during class time or at home.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
Dose, Not Access: Digital Reading Tutor Use and Early Reading Growth — Nicholas A. Gage — EdArXiv — doi:10.35542/osf.io/4a6sk_v1
Abstract licence: CC BY
Explanation generated by Synthesis from the abstract — it is not the original paper. This is a preprint, so it hasn't necessarily been peer reviewed.