What about tracing and other multi-sensory teaching approaches?

  • 22 June, 2024

Blast from the Past: This blog first appeared on May 16, 2020, and an updated version was released on June 22, 2024. The research references have been updated. If you would like to read the 26 comments that the original release attracted click here:


Teacher question: 

I have read the work of researchers like Louisa Moats, Stanislas Dehaene, and Linnea Ehri and understand of how reading works in the brain. I understand the critical role of connecting graphemes to phonemes. My question is what is the true role of the kinesthetic activities promoted in many intervention programs? In a webinar that I

watched the speaker mentioned several times how critical it we to have students trace

the words because this created neural pathways. What does the research say about this?

Shanahan response:

The idea of tracing words to improve literacy has been around for a century. You’d think by now, we’d have a clear idea on whether tracing (and all the other haptic and kinesthetic training procedures) help and, if so, how and why.

But you’d be wrong.

This method was first described by Grace Fernald and Hellen Keller in 1921. Fernald, a clinical psychologist, with a practice focused on reading improvement, applied the method with severely disabled readers. By all accounts, she was a remarkable teacher and her article described what she did and how well it worked (the kids she worked with learned to read). She didn’t devote much space to explaining why tracing was such a boon.

Her idea caught on and ended up in several remedial reading programs, most notably in the one created by Gillingham & Stillman (now referred to as the Orton-Gillingham or O-G method). And, via that route, there are now several commercial instructional programs aimed at dyslexia that include tracing and air writing and that sort of thing.

Over time, these V-A-K-T (visual, auditory, kinesthetic, tactile) practices accumulated several explanations of their effectiveness (Shams, & Seitz, 2008). Many of these focus on memory, claiming that tracing builds neural pathways or that it reinforces visual-auditory pathways in the brain through physical movement and touch. There are also attentional and perceptual explanations and there have been many a rationale based on whatever the current thinking on brain architecture and neural processing may be at the time. Some of these explanations have fallen by the wayside as it has become apparent that they were out of sync with the how the brain works. Many of these are still unresolved.

Personally, I’m more in the attention camp (D’Mello & Gabrieli, 2018), I’m not convinced that these practices create special neural routes or facilitate the paring of alternative paths that is typical of any kind of learning.

My own guess – and this is no more than that and mine is not necessarily better than yours – is that the various kinesthetic schemes simply increase the amount of time readers spend looking at the letters and words. Better attention often translates to more learning.  

Attention is not trivial, so if tracing gets kids to look better or longer, it could be mathemagenic (an action that give rise to learning). Think of many study skills; they often work because they get students to spend more time thinking about the ideas in text by highlighting or note taking (Rothkopf, 1970).

When youngsters simultaneously look at a word, say its name, and trace its letters, it is possible that they are improving word memory for some subtle neurological reason, but it seems more likely that they are just spending more time looking at the words and this may encourage or facilitate phonological stretching (drawing out the pronunciation of a word to highlight the phonological parts and make them more phonetically accessible).

Of course, providing a rationale for why tracing works, assumes that it does, which raises a big, “Not so fast.”

Unfortunately, though educators have tilled these fields for a hundred years, it is unclear whether it works. I often hear from parents and educators who tell me that O-G is the “gold standard” of reading interventions, however the research studies do not provide as glowing an endorsement. Kids don’t seem to read better if they are taught with O-G, tracing or not (Stevens, et al., 2021).

Part of the problem is that these instructional procedures have never gained much sustained attention from the scientific community, so deep understanding is not likely. Often when there is a lot of study of an issue, we start to figure them out a lot better. That has not been the case here.

Making it even more difficult to sort out is the fact that many of the studies have been small (often with only 2-3 children), quite diverse in outcomes and types of students served.

Certainly, some studies support the idea of teaching reading (or aspects of reading such as letter recognition or blending) with multisensory approaches (Campbell, Helf, Cooke, 2008; Connor, 1994; Gentaz, Cole, & Bara, 2003; Ho, Lam, & Au, 2001; Itaguchi, Yamada, & Fukuzawa, 2015; Itaguchi, Yamada, Yoshihara, & Fukuzawa, 2017; Nash, Thorpe, & Lamp, 1980; Thomas, 2015; Xu, Liu, & Joshi, 2019). Many of the studies were conducted with non-alphabetic languages like Japanese or Chinese. None of the studies done in Western languages controlled for the time differences in how long the students were looking at the words. That is not supportive of my supposition about attention, but it does not refute it.

There are some studies that support tracing, but many more and better done studies have reported no clear or consistent benefits (Hulme, 1981; Lee, 2016; Myers, 1978; Schlesinger & Gray, 2017; Wilson, Harris, & Harris, 1976). There are still other studies showing that tracing can be distracting or irrelevant, leading to lower performance when compared to more traditional visual-auditory approaches to decoding (Berninger, Lester, Sohlberg, & Mateer, 1991; Rau, Zheng, & Wei, 2020); Vandever, & Nevelle, D. 1972).

After 100 years, I still can’t tell you if tracing improves learning when it comes to reading.

Of course, several instructional programs incorporate tracing, and some of those programs are effective. Unfortunately, studies of them can’t reveal the impact of tracing because those programs include much more than that. Maybe the tracing is an effective ingredient, or maybe it is inert (just a wasting a bit of time). It is even possible that it is disruptive, but if so, it is not so damaging as to outweigh the program benefits.

Perhaps if tracing supports more thorough and careful looking and listening, it is beneficial. When it doesn’t, it may have no impact whatsoever. And, when learners get all wrapped up in rubbing the letters or dipping their fingers in goop, it could be a distraction that reduces learning.

As a teacher I would not seek out multisensory programs, though I wouldn’t necessarily avoid them either.

If I were using such a program, I’d do what I could to ensure that the tracing wasn’t distracting the students from matching sounds and spellings by ear and eye. I prefer having students seeing the words and hearing the sounds while tracing; I’m not a big fan of “air tracing” despite its effectiveness in Japanese character memorization.

Tracing, if it is to be used at all, should slow students down, focusing their attention on the letters and helping them to think about the letters and sounds more thoroughly and carefully. The teacher who uses this method must be vigilant to make sure that it delivers.


Berninger, V., Lester, K., Sohlberg, M. M., & Mateer, C. (1991). Interventions based on the multiple connections model of reading for developmental dyslexia and acquired deep dyslexia. Archives of Clinical Neuropsychology, 6(4), 375-391.

Campbell, M. L., Helf, S., Cooke, N. L. (2008). Effects of adding multisensory components to a supplemental reading program on the decoding skills of treatment resisters. Education & Treatment of Children, 31(3), 267-295. 

Connor, M. (1994). Specific learning difficulty (dyslexia) and interventions. Support for Learning, 9(4), 114-119.

Fernald, G. M., & Keller, H. (1921). The effect of kinaesthetic factors in the development of word recognition in the case of non-readers. Journal of Educational Research, 4, 355-379.

Gentaz, E., Cole, P., & Bara, F. (2003). Évaluation d'entraînements multisensoriels de préparation à la lecture pour les enfants en grande section de maternelle: Une étude sur la contribution du système haptique manuel. L’Annee Psychologique, 103(4), 561-584.

Ho, C. S., Lam, E. Y., & Au, A. (2001). The effectiveness of multisensory training in improving reading and writing skills of Chinese dyslexic children. Psychologia: An International Journal of Psychology in the Orient, 44(4), 269-280.

Hulme, C. (1981). The effects of manual tracing on memory in normal and retarded readers: Some implications for multi-sensory teaching. Psychological Research, 43(2), 179-191.

Itaguchi, Y., Yamada, C., & Fukuzawa, K. (2015). Writing in the air: Contributions of finger movement to cognitive processing. PLoS One, 19(6).

Itaguchi, Y., Yamada, C., Yoshihara, M., & Fukuzawa, K. (2017). Writing in the air: A visualization tool for written languages. PLoS ONE, 12(6).

Lee, L. W., (2016). Multisensory modalities for blending and segmenting among early readers. Computer Assisted Language Learning, 29(5), 1017-1032.

Myers, C. A. (1978). Reviewing the literature on Fernald’s technique of remedial reading. Reading Teacher, 31(6), 614-619.

Nash, R. T., Thorpe, H. W., & Lamp, S. (1980). A study of the effectiveness of the kinesthetic-tactile component in multisensory instruction. Corrective & Social Psychiatry & Journal of Behavior Technology Methods & Therapy, 26(2).

Rau, P.P., Zheng, J., & Wei, Y. (2020). Distractive effect of multimodal information in multisensory learning. Computers & Education, 144. https://doi.org/10.1016/j.compedu.219.103699

Schlesigner, N. W., & Gray, S. (2017). The impact of multisensory instruction on learning letter names and sounds, word reading, and spelling. Annals of Dyslexia, 67(3), 219-258.

Shams, L., & Seitz, A. R. (2008). Benefits of multisensory learning. Trends in Cognitive Science. https://www.cell.com/trends/cognitive-sciences/fulltext/S1364-6613(08)00218-0?_returnURL= 

Stevens, E. A., Austin, C., Moore, C., Scammacca, N., Boucher, A. N., & Vaughn, S. (2021). Current state of the evidence: Examining the effects of Orton-Gillingham reading interventions for students with or at risk for word-level reading disabilities. Exceptional Children, 87(4), 397-417. DOI: 10.1177/0014402921993406

Thomas, M. (2015). Air writing as a technique for the acquisition of sino-Japanese characters by second language learners. Language Learning, 65(3), 631-659.

Vandever, T. R., & Nevelle, D. D. (1972). The effectiveness of tracing for good and poor decoders. Journal of Reading Behavior, 5(2), 119-125.

Wilson, S. P., Harris, C. W., & Harris, M. L. (1976). Effects of an auditory perceptual remediation program on reading performance. Journal of Learning Disabilities, 9(10), 671-678.

Xu, Z., Liu, D., & Joshi, R. M. (2019). The influence of sensory-motor components of handwriting on Chinese character learning in second- and fourth-grade Chinese children. Journal of Educational Psychology. https://doi.org/10.1037/edu0000443


 Shanahan On Literacy Podcast 


See what others have to say about this topic.

Mary Ann Carter Jun 22, 2024 12:41 PM

Thank you, Dr. Shanahan, for your thoughts on tracing and on using O.G. for intervention. I have been using the O.G. Method through the IMSE Intervention Program for 6 years. I’ve been in teaching for 36 and in Reading Intervention for 12. I have found that O.G. for students with Dyslexia can be lacking in some areas and have needed to supplement it with other strategies. Teaching “red words” hasn’t worked well with this method. I wonder what your thoughts are in teaching Sight Words to students with Dyslexia and other students who struggle with learning to reading.

Timothy Shanahan Jun 22, 2024 01:04 PM

Mary Ann--
I know some "experts" opposed the teaching of sight vocabulary or much sight vocabulary (limiting this to only words that are exceptional) to struggling readers, but it seems to me the safest bet is to build a collection of sight words in students' memory while simultaneously familiarizing them with the most common/useful spelling patterns. The combination -- to my way of thinking -- is more likely to end in success.


Yaneth Jun 22, 2024 01:20 PM

Can you give some examples of “traditional visual-auditory approaches to decoding” that are indeed effective and should be incorporated into a teacher’s intervention strategies toolbox?

Heidi Jun 22, 2024 01:32 PM

This was clear as mud….
So tracing may work as a strategy or it may not.
I am much more competent after reading this post.

Rick Jun 22, 2024 02:45 PM

As always, I come away from a Shanahan post thinking more about what I think I know about reading. :) After reading this review, I have 3 quick thoughts...
1). As a Wilson trained teacher (similar to OG), I also use multi-sensory methods and have often wondered about the theory behind the practice. I agree - "attention" is a key variable in this discussion.
2). Relatedly, I would also say multi-sensory techniques like air-tracing "might" reinforce neural processing because many kids (but not all!) find them fun. Let's not forget that motivation is also an important variable in learning to read (and one many educators forget about as the Science of Reading becomes the new phrase du jour). Duke and Cartwright's emphasis on self-regulation (Active View of Reading, 2021) is a valuable extension to Scarborough's Rope.
3). Finally, Thank You, Mr. Falker is a great book, but it does reinforce "beliefs" that merit further investigation. :)

Mari Buys Jun 23, 2024 05:12 PM

Unfortunately I cannot reference this, but my understanding of the MOTOR part of multisensory input is actually SPEECH itself. By looking in a mirror when articulating a sound and being aware of what your articulators are DOING, the motor system is activating the phonological system. Any ideas on this?

Anonymous Jun 22, 2024 03:19 PM

I’m a parent of a child with auditory processing disorder. Learning to read in a classroom has a lot of obstacles if you’re auditory processing operates differently. With that in mind, I do think seeing, hearing, feeling the movement of letter formations can help a lot of kids who have a variety of sensory and processing differences grab on to what’s being taught. On the surface it may seem like attention issues, but underneath that it could be processing variation from person to person. It may be easy to assume that multi sensory approaches are for kids with diagnosed disabilities… but that stuff is on a continuum. We’re all neurodiverse. We all have variation. Kids with disabilities have just crossed the line that was drawn in the sand… but there’s another kid who’s just on the other side of that line. And many others who have not been tested. Multi sensory is following a universal design model.

Suzie Jun 22, 2024 03:43 PM

Tracing vs mapping should be considered. Students should sound out the words and form the letters as they write NOT TRACE. Mapping sounds also allows the teacher to observe the students sound-letter correspondence. The act of writing letters (not tracing) as they say the sounds (not letter names) reinforces the sound letter correspondence. This is further applied in larger words where students say the chunks and then spell chunk by chunk. This reinforces the reading strategy of reading by chunks. Then when writing or copying connected text students say phrases/groups of words as they write. This reinforces phrasing while reading.

Gale Jun 22, 2024 03:56 PM

The advice to teachers should be use pencil and paper and skip the sand and shaving cream. Golf pencils for K-1 are great. Practice letter formation a lot.

Multi sensory approaches skirt too close to learning styles which we know definitely are pseudoscience.

Harriett Janetos Jun 22, 2024 05:58 PM

David Share (the self-teaching hypothesis) speculates that the reason why spelling instruction improves reading even better than decoding instruction is because students are uniting seeing the letters with writing the letters, simultaneously activating the visual with the kinesthetic.

Spelling as a self-teaching mechanism in orthographic learning, Daphna Shahar-Yames and David L. Share, Journal of Research in Reading, 2008

"Even when spelling highly familiar words, the writer is obliged to retrieve the elements of the visual form of the word whereas reading only requires recognition (Perfetti, 1997). When spelling, furthermore, the writer must process each and every letter. In reading, on the other hand, the orthographic representation may be less than fully specified yet sufficient for word identification, particularly when encountered in meaningful context (Holmes & Carruthers, 1998)."

Having students say the phoneme while writing the grapheme further reinforces the sound-spelling connection necessary for orthographic mapping.

Dana Jun 22, 2024 08:48 PM

I have seen students who view tracing like making an art project, with no connection to the real purpose of the activity. They can complete pages of traced letters, and have no more letter knowledge at the end. For some of my students, sand trays, textured letter cards, and shaving cream (I hate the stuff!), do help kids to learn letters. Whether it is because of the novelty and paying closer attention or the act/motion of writing the letters, I don't know. Perhaps these techniques also slow down the process, again teaching kids to look at the sub-letter level. I have also had great success with students writing letters in the air, especially my MLs, who may struggle with directionality.

Gaynor Jun 23, 2024 12:30 AM

After the experience of having my own child with quite severe dysgraphia, which severely interfered with his story writing and spelling, I now put more emphasis on letter formation at an early age including tracing. I have seen in articles strong recommendations in the value of handwriting and its likely benefits for hyperactivity as well.

I am keen on proper pencil grip since I have observed older students who held a pencil with an iron grip as if they were stirring porridge and consequently needed to stop writing and shake their hand because they had cramp.

My conclusion is that letter formation should be an exercise combining the importance linking sounds and graphemes and fine motor control since both, for small children can take some time to develop. I am also a bit of a stickler for explicit instruction in sensibly formed letters .which includes chanting up down and around etc. I have experimented with children doing repeated lazy eights and the clockwise and anti clockwise classification of letters in desperation to try and improve the ghastly illegible handwriting of dysgraphics.

Beth Hankoff Jun 23, 2024 02:04 AM

I’m a handwriting specialist with Handwriting Without Tears training. Some have mentioned handwriting instruction with tracing. If you are working one-on-one you may be able to monitor the child and make sure they are tracing the letters properly. I would NOT recommend tracing in the classroom for learning letter formation. Kids will trace the letters any which way and they will not learn proper formation. Even with numbers and arrows on the letters, they ignore them. Much better to have them practice making each stroke properly and checking that they’re doing it correctly, perhaps in small groups.

Cecelia Bockenstedt Jun 23, 2024 05:14 AM

Again, as a former classroom teacher, I find myself bristling with frustration as I read this.

As a recovering person with German ethnicity, I want to put in all caps: Well, get on with it then, researchers!

Release thousands of teachers from the burden of doing something that is not necessary!

I've personally let go of doing air-writing (I'm a Barton tutor), and instead do a sound-association thing where I tell the student that we're going to work on 3 words that are spelled in a way that you wouldn't expect. I say the first word, and ask how many sounds they hear in it. (So they need to have digraph sounds, letter sounds, and segmenting down pretty well.)

Then we draw horizontal lines for the sounds. Together, we fill in the lines with what makes the most sense. Sometimes, it's an opportunity to mention that English words don't end with a v, or something equivalent.

The student highlights the "tricky spelling, and then erases the word, sound by sound. (dry-erase board) Finally, they write it themselves, saying each sound as they write the letters.

Based loosely on Reading Simplified's lessons on orthographically irregular words.

It seems to be working just as well as the other procedure I'd been taught to do, but I'd sure love to see some research, especially with retention of these words by kids with dyslexia.

That is all.

Thank you for broaching the topic.

Timothy Shanahan Jun 23, 2024 02:25 PM

There is some research showing that when it comes to teaching handwriting, having students trace letters does make a positive contribution to their development.


Timothy Shanahan Jun 23, 2024 02:33 PM

Writing does everything that you describe -- tracing does none of those.


Timothy Shanahan Jun 23, 2024 02:35 PM


You can believe all of that, but there is no research evidence supporting it, neurological theories have moved on from that, and there is some empirical evidence on tracing suggesting that it is more likely to confuse children than to help them learn.


Timothy Shanahan Jun 23, 2024 05:26 PM


There is no question that seeing a word and saying a word simultaneously. I suspect the verbal/linguistic part of that is what is important rather than the motor portion of the activity. This could easily be studied by having some kids think a word when they saw it or imagine making that sound. You could get a small fall off from some subjects who didn't think the word/sound (you couldn't tell for sure), but otherwise I'd bet they would do equally well.


Timothy Shanahan Jun 23, 2024 05:31 PM


Basically, teaching decoding minus tactile-kinesthetic exercise, getting kids to look at words and say them, having kids trying to write words from memory and then checking their spellings against the standard, encouraging invented spelling.


Lise L'Heureux Jun 24, 2024 07:01 PM

Bonjour Tim !

I've been teaching dyslexic kids for 20 years. I did use tracing following OG training but found that it made no difference, and took up precious time that could be used for more useful work. What does seem to work much better is saying, listening and writing the words (not tracing letters).

The tracing of letters, not knowing if it really works or not, reminds me of one of my high school students who's teacher spent hours if not entire weekends correcting her students spelling tests with a sophisticated colour coding. When I ask M-A what "classe de mots" (not sure if you call these "word classes" in English : noun, verb, article, adjective ...), he answered without skipping a beat : C'est le orange, Mme Lise. I still laugh about it 10 years later. He was much more into colours than "classes de mots" lol

Always interesting reading you!


TS: "After 100 years, I still can’t tell you if tracing improves learning when it comes to reading."
LLH : I've met you. You are definitely not that old ! ;-)

Karen Sonday Jun 25, 2024 05:18 PM

We use tracing as a strategy to decode a word. When the student encounters a word they have the skills to decode but they aren't able to read it just by looking at it, we prompt them to trace it on the tabletop as they say the sounds. It works like magic. They've written the grapheme countless times so perhaps that movement triggers the sound for them (especially helpful with vowels), or perhaps it just slows them down enough to get the sounds in the right order (bulb instead of blub), or perhaps it helps them focus, but regardless, they're far more accurate (and confident that they're right) than when they utilize their stand-by of guessing and hoping.

That does not seem to be the process that's being discussed here. Are you referring to looking at words and copying them in an effort to build automaticity or accuracy?

Timothy Shanahan Jun 26, 2024 12:51 PM

Yes, trying to "copy" words but without the words being visually available. Look at the word, remove the word, and then try to write the word. In other words, what you are doing is trying to fix a detailed representation of the word in memory (detailed in the sense that it is based on knowing the sequence of letters and not shape or some distinctive feature).


Michele Minor-Corriveau Jun 25, 2024 08:00 PM

Muscle memory is built through repetition. In sport, the movements requiring fine tuning are often accompanied by verbal instruction that we repeat in our heads while memorizing these movements. This is true when learning a new choreography. It was true when we learned to associate the letters of a keyboard to the finger required to be most efficient. James & Engelhardt (2012) did in fact measure 5-year old children's abilities to associate symbols to sounds. "Preliterate, five-year old children printed, typed, or traced letters and shapes, then were shown images of these stimuli while undergoing functional MRI scanning. A previously documented “reading circuit” was recruited during letter perception only after handwriting—not after typing or tracing experience. These findings demonstrate that handwriting is important for the early recruitment in letter processing of brain regions known to underlie successful reading. Handwriting therefore may facilitate reading acquisition in young children."

Palmis, Velay, Fabiani, Nazarian, Ancton, Habib, Kandel & Longcamp (2019) also measured the contribution of handwriting to orthographic process. "These results constitute direct evidence that orthographic and motor processes occur in a continuous and interactive fashion during writing."

In a typically developing brain and body, automated motor movements require less and less time in order to be executed accurately. In time, cognitive load is reduced and access occurs without much effort. If one can argue that seeing the letters more frequently can help the learner retrieve the pertinent information more quickly, then one can hypothesize that this also applies to frequent motor practice involving tracing letters while learning about the letter sounds and names (teaching both in unison yields more benefits then teaching one or the other). There are many studies that have studied the process of the graphomotor movement required to become a fluent writer, and they do include building up to automatized letter writing on command through activities like tracing or using dotted lines at first - like anything else, a template always helps set the stage for precise movement : the key is knowing when to remove the prompt and foster independence (Goigoux, et al. 2016). With proper instruction with respect to directionality, writing fluency will lend itself well enough to cursive movements, for the most part.

Goigoux et al. (2016). Lire et écrire au CP. Institut français de l'éducation. https://ife.ens-lyon.fr/recherche/thematiques-de-recherche/projets/la-recherche-lire-et-ecrirecp

James, K. H. & Engelhardt, L. (2012). The effects of handwriting experience on functional brain development in pre-literate children, Trends in Neuroscience and Education, 1(1), pp. 32-42, ISSN 2211-9493, https://doi.org/10.1016/j.tine.2012.08.001

Palmis, S., Velay, J.-L., Fabiani, E., Nazarian, B., Ancton, J-L., Habib, M., Kandel, S., & Longcamp, M. (2019). The impact of spelling regularity on handwriting production: A coupled fMRI and kinematics study. Cortex, 1113, pp. 111-127.

Timothy Shanahan Jun 26, 2024 12:39 PM

There is no reason to believe that the effects of writing and spelling on reading has anything to do with graphomotor movement. Cognitively it is markedly different than tracing.


What Are your thoughts?

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

Comment *

What about tracing and other multi-sensory teaching approaches?


One of the world’s premier literacy educators.

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