How can we bring the reading sciences closer to the classroom?

“Although reading science is highly relevant to learning in the classroom setting, it does not yet speak to what to teach, when, how, and for whom at a level that is useful for teachers.” (Seidenberg et al., 2020)

This observation captures one of the central challenges facing literacy education today. Teachers are increasingly encouraged to draw on evidence from the reading sciences and cognitive psychology, yet translating research findings into practical classroom decisions is often far from straightforward.

In a recent discussion paper, we consider this challenge in the area of phonics instruction. Phonics involves teaching children how letter(s) map onto sounds: for example, “ea” is pronounced as a long /i/ sound in words like tea or bean. This enables children to decode unfamiliar words independently, creating opportunities for wider reading practice and the development of fluency and comprehension. Substantial research has shown that systematic phonics instruction provides the most effective route into reading in alphabetic writing systems.

However, to teach phonics optimally, teachers must have answers to several practical questions: Which letter-sound correspondences should be taught? How should they be illustrated by words that exemplify the correspondences? And when should irregular “tricky words” be introduced?

The Phonics Wild West

Our paper shows that existing phonics programmes provide very different answers to these questions. Widely used programmes such as Jolly Phonics, Letters and Sounds, and Read Write Inc. differ considerably in the number of letter-sound rules they teach, the sequence in which they introduce them, the example words they use, and the exception words they prioritise. The rationale behind these design decisions is often unclear and not explicitly linked to existing research.

Differences in the teaching content of three major phonics programmes: Jolly Phonics (JP), Letters and Sounds (LS), and Read Write Inc. (RWI). Source: https://doi.org/10.17605/OSF.IO/6WV9C

What the Reading Sciences Can Tell Us

How can reading research help schools make informed decisions about what phonics knowledge to teach, as well as inform improvements to phonics programs themselves?

One major issue concerns the selection of letter-sound rules. English contains hundreds of possible regularities, but classroom time limits how many can be taught explicitly. The challenge is therefore not simply to teach more phonics, but to identify the most useful content. Highly frequent and reliable correspondences such as “ee”, which are almost always pronounced the same (e.g., see, need, indeed), may offer greater instructional value than rare or inconsistent ones such as the different pronunciations of “ou” in words like out, soup, could, and mould. Research suggests that effective instruction is a balancing act, wherein letter-sound correspondences that are highly frequent and consistent are taught in the classroom and others that are less frequent and/or inconsistent are acquired through reading practice. 

Using Examples and Teaching Exceptions

In our paper, we also address a surprisingly underexplored aspect of phonics instruction: the example words used to illustrate spelling-sound correspondences. 

Research from language acquisition and category learning suggests that diversity in examples is crucial. Learners acquire patterns more effectively when they encounter them across a wide range of contexts. For phonics, this means that the pronunciation of a letter such as “a” should be practised using varied examples (e.g., pat, fan and cap) rather than highly similar words (e.g., bat, pat and hat). Diverse examples help children identify the underlying pattern and apply it more successfully to new words.

Another question concerns the teaching of exception words—sometimes called “tricky words”—whose pronunciations do not follow the regular patterns children have learned. For example, be does not match the usual pronunciation of “e” (as in bed), and said deviates from the typical pronunciation of “ai” (as in rain).  Studies from cognitive psychology indicate that learners typically benefit when exceptions are introduced after a strong understanding of the regular patterns has been established. At the same time, many common English words are irregular, making it difficult to postpone them for too long. Phonics instruction must strike an appropriate balance between strengthening word decoding skills and preparing children to read authentic texts.

A New Roadmap for Educationally-Relevant Research

The broader vision of our paper lies in showcasing the potential of research on reading and learning to inform day-to-day decisions in educational contexts. For this purpose, we outline a road map for how to strengthen the ties between current and future research and the concerns of various educational stakeholders, including teachers, program publishers, education boards and educational technology (EdTech) companies. In particular, we highlight the merits of combining multiple research methods, including classroom interventions, laboratory experiments and computational modelling, to create a reliable evidence base that can guide educational practice. Other important elements of the road map are an emphasis on open science, a mutual feedback cycle between researchers and practitioners, and greater attention to individual differences among learners.

Ultimately, our paper is not simply about phonics. Instead, it provides a model for how reading science—and cognitive psychology more broadly—can support evidence-informed teaching.

Ungerer, T., Rastle, K., & Armstrong, B. C. (2026). Bringing the Reading Sciences into the Classroom: Insights for Phonics Instruction. Perspectives on Psychological Science, https://doi.org/10.1177/17456916261438004