What Does Brain Science Have To Say About Teaching Reading? Does It Matter?

  • 13 January, 2024

Teacher question:

I am the principal of a small primary grade school (350 students). I want to hire a consultant/ professional development specialist who could school my faculty in brain science so they will be able to teach reading more effectively. We all earned our credentials in colleges of education so none of us knows these new brain-based methods of teaching reading. Could you please provide some guidance?

RELATED: Does the science of reading include middle school?

Shanahan responds:

It may be hard to believe given news media reports and the numerous books that now purport to translate neuroscience into pedagogy, but there are not any new and effective instructional methods, approaches, techniques, or materials that have been developed based on “brain science.”

Save your money.

Invest in something more certain to help your school – like buying lottery tickets.

When people are talking about “teaching the brain to read”, they are typically touting phonics instruction. You know, phonics, an instructional method developed in the early 19th century. Not exactly the spawn of modern neuroscience.

Teaching phonics is teaching the brain.

But then so is teaching word memorization.

I don’t mean to be cavalier here – I do understand that neurologists have identified some provocative distinctions between decoding and word memorization (we’ll get to that) – but let’s be honest: all cognitive learning is housed in the brain.

Much is made in those books and articles about how phonics is the right approach because it alters the brain. That latter claim is true as far as I can tell (I’m not a neuroscientist so reading such research gives me the heebee jeebees). However, it is not just phonics that changes the brain. The same can be said about any kind of learning, education, physical exercise, meditation and so on. They all alter the brain in terms of the circuits that are formed and the brain’s physical properties (such as thickening the hippocampus).

So far, no instructional method has resulted from the study of the brain.

Probably the best treatment of the neurological study of the reading brain aimed at a general audience is the now somewhat dated book (first published in 2009), Reading in the Brain: The New Science of How We Read by Stanislas Dehaene.

That book has a bit of a split personality – it starts out writing checks that it can’t cash and ends up getting real. On page 2, Dehaene claims: “The insight into how literacy changes the brain is profoundly transforming our vision of education and learning disabilities. New remediation programs are being conceived that should, in time, cope with the debilitating incapacity to decipher words known as dyslexia.”

Sounds great! That’s the kind of assertion that leads to letters like yours.

If neuroscience is leading to new ways of teaching, then teachers want to get their hands on those innovations.

But if you were tantalized by that page 2 promise, you’re going to be disappointed by the practical directions that neuroscience proposes. Dehaene argues for instruction in phonemic awareness (PA) and concedes that PA is not a prerequisite to reading (kids are likely developing PA and decoding simultaneously). I agree with all of that, but none of those pedagogical conclusions come from brain science – Dehaene usually cites psychological studies to support that type of claim.

Other insights that he shares are that kids learn complex rules or patterns later than simple ones, and that repetition matters when it comes to learning.


Dehaene’s own characterization of these pedagogical claims: “A great many teachers will consider my recommendations redundant and obvious—but it does no harm to specify them” (Dehaene, 2009, p. 229). 

Let’s get real.

Neuroscientific research can do one of two things when it comes to the teaching of reading.

One possible outcome is that it will identify a structural difference (say, between the brains of normal readers and those with dyslexia) or some puzzling neurological process – such as a circuit implicating an unexpected region of the brain. These kinds of findings could, theoretically, lead to the development new assessments for the early identification of reading problems or suggestions for new and different teaching methods.

Neurological science has not yet led to such practical innovations. They might someday – that research should continue to be funded – but at this stage it hasn’t happened.

A second possible contribution that brain study can make is that it confirms what we already know. This kind of confirmational study is more about understanding the brain than how to teach reading. Such research offers possible explanations for why things work the way they do.

These studies have revealed that when we read words, we activate visual-phonological circuits in the brain. Such observations have led neuroscientists to conclude that phonics would possibly be more effective and/or more efficient than the teaching of whole words.

More recent studies (studies that were not yet available to Dehaene) go even further.

For example, in one fascinating investigation, subjects were either led to memorize whole words (with a made-up set of orthographic symbols) or to decode those symbols. The decoding instruction led to neural processing like what is observed in the brains of proficient readers (Yoncheva, et al., 2015). Word memorization led to processing more like what we do with pictures than with language.

The conclusion from such studies has been that it makes sense to teach phonics.

I certainly agree with that conclusion, but not because those studies are definitive. My assent comes from the fact that those conclusions are consistent with what psychological and pedagogical studies have repeatedly demonstrated for more than 60 years.

My reasoning isn’t:

“Oh wow, the brain coordinates both visual and phonological information when we read words. Man, I think we should try to teach kids to do that.”

It is more:

“That’s cool. These images of the brain show that kids coordinate visual and phonological information when they read words. I wonder if that is why reading instruction works better when phonics is included?”

I advocate phonics because so many studies show that kids do better in learning to read when that is part of their instruction. I do appreciate that these neurological findings appear to be consistent with those studies of teaching. This concurrence may give me greater confidence, but it would not make any difference in my practice. (Of course, it should be noted that the instructional studies can do more than just suggest possible benefits or efficiencies that could result from phonics – unlike the brain studies. No, instructional studies will also provide me with guidance as to what the content of those lessons should be, the types of examples and explanations I should provide, the actions the students should be engaged in, their duration, and other practical specifics that are pedagogically essential if I am to teach something, but that are unheard of in brain studies.

Think about it. What if we had no instructional evidence that phonics improved reading achievement, but neuroscientists had scads of photographs showing that we connect visual and phonological information when we read words?

If that were the case, I would not be advocating the teaching of phonics.

Instead, I’d be calling for further research to evaluate this fascinating hypothesis in classrooms. The same way such information is handled by the medical community.

Neuroscientists identify unusual accumulations of plaque in the brains of Alzheimer patients. Based on that information, physicians don’t immediately start prescribing anti-plaque medications. They wait until there are medical studies showing that reducing plaque works. Despite the obvious conclusion from brain images that plaque causes this disease, further study was required and that showed that plaque removal (or plaque removal alone) is neither a cure nor a palliative. 

Neuroscience is largely a correlational enterprise. Scientists analyze brain images and look for patterns and consistencies. That information is then translated into hypotheses and possible explanations for how those patterns connect to external behaviors and conditions.

In reading, most neural studies have explored how children read, not how they learn to read. Longitudinal studies, for instance, have been unusual (Wang, et al., 2023). Until recently, fMRIs could be used only with the reading of single words. Because those studies couldn’t look at connected text, they were unable to consider the impact of semantic context (Junker, et al., 2023; Terporten, et al., 2023), how ambiguous words are processed (Mizrachi, et al., 2023), the role of morphemes (Marks, Eggleston, & Kovelman, 2024), font differences (Wu, et al., 2023), or anything else about how we process written language. The newer studies, as they have looked at phenomena more like real connected reading, have not contradicted the explanations formulated from the images of single word reading, but time will tell.

Back in the 1960s and 1970s, there were studies that compared children who received little or no phonics with those who received a heavy dose of it. Most kids in both groups learned to read (albeit with less failure, greater average achievement, and better spelling ability in the phonics groups). But what about those kids who learned to read successfully without phonics? How do brains take such different learning paths to get to the same neural processing outcome?

I don’t know the answers to those kinds of questions, but I do know that the explanations that have been provided so far tend to neglect variations in learning and processing (Debska, et al., 2023; Wat, et al., 2023).

My advice?

I wouldn’t look for a consultant who knows the neuroscience, but for one who has a deep understanding and appreciation of the findings of instructional study. Your teachers don’t need to know how the brain processes single words, but what content if taught and what instructional methods if used are likely to be most successful in raising students’ reading achievement. Except in the most general terms (e.g., teach phonics, encourage kids to read a lot), neuroscience has few practical suggestions that do any more than confirm what you and your teachers already probably know.


Debska, A. M., Wang, J., Dziegiel-Fivet, G. K., Chyl, K. M., Wojcik, M. P., Jednorog, K. M., Booth, J. R. (2023). The development of orthography and phonology coupling in the ventral occipito-temporal cortex and its relation to reading. Journal of Experimental Psychology: General. https://dx.doi.org/10.1037/xge0001495

Dehaene, S. (2010). Reading in the brain: The new science of how we read. New York: Penguin Books.

Junker, F. B., Schlaffke, L., Lange, J., Schmidt-Wilcke, T. (2023). The angular gyrus serves as an interface between the non-lexical reading network and the semantic system: evidence from dynamic causal modeling. Brain Structure & Function. https://dx.doi.org/10.1007/s00429-023-02624-z

Marks, R. A., Eggleston, R., Kovelman, I. (2024). Brain bases of morphological awareness and longitudinal word reading outcomes. Journal of Experimental Child Psychology. https://dx.doi.org/10.1016/j.jecp.2023.105802

Mizrachi, N., Eviatar, Z., Peleg, O., Bitan, T. (2023). Inter- and intra- hemispheric interactions in reading ambiguous words. Cortex, 171, 257-271. https://dx.doi.org/10.1016/j.cortex.2023.09.022

Terporten, R., Huizeling, E., Heidlmayr, K., Hagoort, P., Kosem, A. (2023). The interaction of context constraints and predictive validity during sentence reading. Journal of Cognitive Neuroscience, 1-13. https://dx.doi.org/10.1162/jocn_a_02082

Wang, F., Kaneshiro, B., Toomarian, E. Y., Gosavi, R. S., Hasak, L. R., Moron, S., Nguyen, Q. T. H., Norcia, A. M., McCandliss, B. D. (2023). Progress in elementary school reading linked to growth of cortical responses to familiar letter combinations within visual word forms. Developmental Science. https://dx.doi.org/10.1111/desc.13435

Wat, E. K., Jangraw, D. C., Finn, E. S., Bandettini, P. A., Preston, J. L., Landi, N., Hoeft, F., Frost, S. J., Lau, A., Chen, G., Pugh, K. R., Molfese, P. J. (2023). Will you read how I will read? naturalistic fmri predictors of emergent reading. Neuropsychologia. https://dx.doi.org/10.1016/j.neuropsychologia.2023.108763

Wu, Y., Luo, C., Wang, Z., Xie, H., Huang, Y., Su, Y. (2023). A further specification of the effects of font emphasis on reading comprehension: evidence from event-related potentials and neural oscillations. Memory & Cognition. https://dx.doi.org/10.3758/s13421-023-01457-9

Yoncheva, Y. N., Wise, J., & McCandliss, B. (2015). Hemispheric specialization for visual words is shaped by attention to sublexical units during initial learning. Brain and Language, 145-146, 23-33. doi: 10.1016/j.bandl.2015.04.001. 


LISTEN TO MORE: Shanahan On Literacy Podcast


See what others have to say about this topic.

Gaynor Jan 13, 2024 04:24 AM

I read Timothy's comments and other contributors on this blog and thought about why phonic instruction at levels above K -1 are of apparently little help.

My main experience of this dilemma is based on what I experienced with my mother's 1000 of remedial students in a high income area last century. The first problem in teaching older students is maybe a developmental one . If a student has not absorbed the phonemes and phonic spelling patterns by age seven then possibly it is much more difficult for them to do so at a later.age .They have also acquired wrong strategies in approaching an unfamiliar word and this has to be countered . This is very difficult . But I think the worst problem is children who become resistant to learning because although they had enthusiasm when they were younger they have with time given up because they have tried hard in the past but still continued to fail.

These resistant children develop all sorts of psychological problems which Kerry Hempenstall of RMTI in Melbourne Australia has written extensively about on line. I have seen for myself in my mother's school room the disturbing behaviours these children display.

This makes remedial reading for older children really difficult. They require an extraordinary amount of patience and expertise by the tutor. I remember the day my own dyslexic and dysgraphic child , in frustration threw his books , into the bathing pool and stormed off down the street declaring he was leaving home. Establishing legible handwriting was very difficult for him when .It was something the rest of his class seemed to do effortlessly. Many hundreds of hours of individual tuition were required to overcome his dyslexia which finally is now no longer . evident. Tantrums, deviant behaviours including defiance and crying can be standard for these failing children. Remedial reading instruction should be seen as a bit of a nightmare and everything possible done earlier to prevent its necessity .

For me , an advocate of Universal Literacy the answer, is to ensure every child achieves in reading in the first two years of school. The various stages like Knowing all the consonants and short vowels should be achieved thoroughly by everyone individually in the class before they move to the next stage . Those who are slower to learn should be given extra help until they have achieved in even small stages. This help could be given by family or other or even older responsible students in the school ,The advanced children could possibly jump up a year.

When I was at school in mid -last century everyone in my class of 50 students , could read the same book at the correct level for their age , although some read a bit slower. There was extension work for the more able. We all worked through the SRA comprehension cards , daily which were graded in difficulty. Phonics instruction was in spelling which was both phonic based and whole word . Everyone in the class received daily he same ten spelling words which were for homework and all tested at the end of the week. These words promoted building up vocabulary .as well. In English morphology and grammar were covered as well as a page of written work every day.

Melanie Jan 13, 2024 03:12 PM

So, can I conclude that it would be more impactful if teachers are trained in Orton Gillingham (the how) than investing two years in LTRS training (the why)?

Mark Pennington Jan 13, 2024 03:24 PM

Two responses: 1. Your writing is more humorous post Mt. Everest... any guru meetings we don't know about? 2. If your article has removed even a single dissected reading brain graphic from any PD, you've earned our heart-felt thanks.

Ruth Jan 13, 2024 04:14 PM

I enjoyed reading your thoughts. I have just finished reading a book about cognitive load theory and how we can take their research and apply it to the classroom. I found some of the ideas different to my usual way of teaching (eg not reading to students whilst they follow the text). Do you have any thoughts on cognitive load theory and teaching practice? It does seem in this case that neuroscience has given us a different perspective on how we pay attention and this has led to suggested instructional changes . Possibly we already knew these ways would work better but it seems our teaching practice still hasn’t followed that route.

Dr. Bill Conrad Jan 13, 2024 04:43 PM

As usual, your clarity and evidence-based discussion dispels the huge fog around the relationship between neural science and instructional practices that are essential to help children learn to read! A true fog buster are you!

Read The Fog of Education!

Timothy Shanahan Jan 13, 2024 04:45 PM

Cognitive load theory came out of cognitive psychology, not neuroscience. They weren't looking at pictures of the brain, they were examining how people process information. In terms of whether kids should follow along while you read or just listen, that depends not on load but on what you are trying to teach. If you are teaching a foreign language or are trying to improve kids' oral reading fluency, having them follow along is quite sensible. (If you are trying to teach reading, then reading to the kids would be an occasional activity. If you are trying to teach listening or to convey certain information to the students, then reading to them makes sense).


Timothy Shanahan Jan 13, 2024 04:48 PM

Just looking at the research on reading instruction, neither of those would be my choice -- though both has been made to work, at least in some instances. I don't recommend programs in this space, but I do encourage professionals to turn to the research when making such decisions -- does this improve reading achievement?


Sherry Bandy Jan 13, 2024 04:56 PM

I love your common sense approach to teaching and learning. Thank you.

Miriam Giskin Jan 13, 2024 05:08 PM

I am just starting to learn about Structured Word Inquiry but feel that it is comprehensive in a way other methods are not because it instructs the reality of how the system works without relying on mnemonics or creating irregularities or exceptions that must then be dealt with. Would be interested in what others make of this work. Thanks.


Shellie Jan 13, 2024 06:29 PM

LETRS is the WHY, but it also includes the HOW. Quite frankly, I believe knowing WHY we do what we do is vital in making decisions along the way, regarding the specific needs of students and evaluating materials.

Timothy Shanahan Jan 13, 2024 07:53 PM

That certainly sounds right and I'll never come out against teachers not knowing stuff, but results are pretty mixed on that very expensive program, most studies of phonics instruction that have been successful haven't required such extensive professional development, and reading is more than phonics (how many years of training in language and language instruction have you enrolled in?).


Barbara Schuh Jan 14, 2024 12:24 PM

I agree with Shellie about learning the why and how that comes with LETRS. It is also a programs that talks about more than phonics instruction. Thinking we can do excellent phonics and solve all our reading problems is going to be short sighted, I fear.
That is why that program is strong.

I do appreciate that you are trying to avoid the problems of research to practice. But I think learning about learning as an educator keeps people from teaching programs and encourages teaching people.
Also, while those of us who have been in education a long time are aware of phonics instruction, we have a generation of teachers who were not given phonics instruction as students in grade school and don't know how to do it as educators. They will need professional development as we continue to dig our way out of the literacy crisis.

Judy Jan 14, 2024 03:30 PM

Is there a neuroscientist who has taught children to read outside of clinical or experimental constructs? Knowing the why is important for educators, but we need more than one why. Receiving training and support in one method or program will not empower us to teach ALL our students. We teach children, not programs and theories. Teachers need a repertoire of the how. In my opinion, in my state, we have gone so overboard with phonics and the "science of reading" in K-6 that we are marginalizing students who struggle. If a child doesn't learn to read with explicit, systematic phonics instruction, they are tracked to struggle. It's frustrating, and the older the child, the more frustrating for everyone involved. I wish we could grasp the notion that children will learn when we teach children.

Melanie McNulty Jan 14, 2024 04:32 PM

Such good discussion. So, what types of pd are effective? k-5.

Timothy Shanahan Jan 14, 2024 04:35 PM

One thing that I want to make clear is that for the most part neuroscientists have usually not weighed in on instruction beyond general statements that their findings seem to favor decoding over memorization -- nothing unreasonable there. It is usually educators who make big claims based on brain science (overstating and failing to use the hedges and qualifications that the scientists use). It is science as marketing device rather than science.


Timothy Shanahan Jan 14, 2024 04:37 PM

Presentations on practical matters (How to do particular things), with opportunity for teachers to go back the classroom and then to revisit the topic.


Peter Bowers Jan 14, 2024 06:52 PM

Tim, I very much enjoyed this post of yours. I have long been very dubious about educational neuroscience in general. If our question is about whether instruction A or instruction B is more effective, I could not imagine how any images of where blood oxygen flows in the brain could tell us anything more than what we could find through studying the behaviour that results from instruction A or B. As you suggest, it makes sense to the that we could learn about how the brain works by measuring it during reading and reading instruction. But I cannot imagine getting information from neuroscience that would convince me to do something in contradiction to findings from behavioural outcomes.

For those interested, my brother wrote an article on this titled "The Practical and Principled Problems with Educational Neuroscience" in Psychological Review that you can access here:


I also appreciated your response to a question about Cognitive Load Theory (CLT) showing that this is not neuroscience. I have long been fascinated with CLT since I first studied the review by (Shnotz & Kurshner, 2007) when my supervisor John Kirby suggested I look at that article. As I read it, it seemed to me that every teacher knows about the ideas they unpack, but that we didn't have "names" for these key ideas. While I don't always succeed, I can say that every time I teach or present since, it is influenced by my learning from CLT. I was developing my work with Structured Word Inquiry (Bowers & Kirby, 2010) at the time, and I was struck by how many of the practices I saw emphasized based on CLT fit with what I was doing with SWI. Must be something about 3-letter acronyms! Anyone interested in the connections I see between this instructional practice and the theory of learning from CLT is welcome to read the term paper (never tried to publish) outlining the theory of CLT and how I see it fitting with SWI. Link to that paper is here: https://files.realspellers.org/PetesFolder/Articles/Best_CLT_Vocap_paper_for_Lesly_dec_14.pdf


Harriett Janetos Jan 14, 2024 09:18 PM

From Pete Bowers:

"But I cannot imagine getting information from neuroscience that would convince me to do something in contradiction to findings from behavioural outcomes."

But can you imagine feeling more confident about doing something based on the findings from behavioral outcomes when information from neuroscience supplies the "why" behind these findings? That's what happened to me when I found myself teaching kindergarten for the first time eight years ago due to unforseen circumstances with staffing. Isabel Beck's Making Sense of Phonics guided my phonemic awareness/phonics instruction (based on the findings from research), and Stanislas Dehaene's Reading in the Brain explained why it was working. This was the first experience I had with actually teaching children how to read, and I found this dynamic duo extremely helpful in guiding my guidance of students.

Timothy Shanahan Jan 15, 2024 07:50 PM


I would suggest that you search this website for such information.


Timothy Shanahan Jan 16, 2024 04:30 PM


I couldn't agree with you more. Unfortunately, that isn't what a lot of schools are being sold.


Mary Baker-Hendy Jan 15, 2024 05:18 PM

Dr. Shanahan,
While I agree that a course on the brain science of learning to read is not a necessary step for faculty professional development, I was disappointed to read your dismissal of the importance of phonics. Your summation of the research on how we learn to recognize words as "provocative distinctions between decoding and word memorization” does not tell the story. In the world of teacher education that fact remains that teachers have not been taught that unlike spoken language, word recognition is not a natural process. The introduction of symbols does require instruction that links the visual and auditory networks of the brain and is best taught explicitly and cumulatively to assure that all children learn to recognize words.

PS There is no reason why vocabulary, morphology, sentence structure and content cannot be integrated in phonics instruction. Those element taught together produces young children that can read and write and are ready to learn.

Timothy Shanahan Jan 15, 2024 07:01 PM


I think you might want to read this again. I spoke up for phonics because studies show that teaching it provides learning advantages to children as shown in a scad of pedagogical and psychological studies. What I spoke up against were claims that we need to teach phonics because phonics teaches the brain or because studies show that readers when reading single words connect auditory and visual processing parts of the brain.


Kionna Squires Jan 15, 2024 07:29 PM

I agree with you 1000%! It’s not about what or who you teach, it’s about HOW you teach. Where can we find more research on instructional studies? Research on things like the gradual release model and explicit teaching vs inquiry based teaching? Any tips in the right direction would be much appreciated.

Kionna Squires (DaLiteracyLady) Jan 15, 2024 07:48 PM

And are teachers in a time where they need to push back against these pacing guides that ask them to teach 1 grade level ELA standard a week? Are teachers asked to move on to a new standard too quickly? I ask that because it seems like the best instructional way to teach reading is to teach it explicitly but in my experience teaching knowledge from the text AND how to perform the standard explicitly takes more than 5 days. It’s like teachers have to choose between ignoring doing what’s right by children or speaking up and getting fired. So just looking for a bit of insight from someone outside of the school.

Sebastian Wren Jan 16, 2024 03:38 PM

I feel like neuroscience research has certainly reinforced a few points that we know, but that rarely find their way into classroom practice. First, explicitly teaching the code is better than letting kids figure it out on their own (or, more likely, letting their parents teach them). Second, starting early is critically important -- kids should be reading confidently and fluently by 2nd grade. And third, focusing on automatic processing at every stage of development is especially beneficial.

Enrique Andres Puig Jan 16, 2024 04:18 PM

Spot on Tim! Thank you for the straightforward response. Lately, I think of the late Erma Bombeck's "the grass is always greener over the septic tank."

Vicki Collet Feb 23, 2024 05:16 PM

Thanks for this important, rationale caution.

What Are your thoughts?

Leave me a comment and I would like to have a discussion with you!

Comment *

What Does Brain Science Have To Say About Teaching Reading? Does It Matter?


One of the world’s premier literacy educators.

He studies reading and writing across all ages and abilities. Feel free to contact him.