Is the multisensory component of reading instruction necessary?

[The following post was originally shared in 2023, in the private Facebook group Effective Education Practices. This is the first time it has been published on a public forum. It contains small edits and one new reference, but is substantially the same as the 2023 version.]

[TL;DR summary: There is insufficient evidence to justify the use of time and resources on inserting a multisensory component into reading instruction.]

In this post, I’m going to look at where the idea that reading instruction had to be multisensory came from, and whether it is supported by evidence or just theory. I will also address why people are confused about its necessity, and what might be happening when people incorporate multisensory elements into reading instruction.

Effective reading instruction for beginning readers, requires the same approach that is supported by wider research into effective education practices: the approach should be direct and explicit, cumulative and sequential, systematic, structured, include timely feedback…

But the list for reading instruction frequently adds on the word “multisensory” as essential to the overall approach as well. In fact, the multisensory component is considered so essential by some that they use it as short-hand for the entire package – it has become widespread to simply use the term “multisensory” to capture all of the above components. This broad use of the term is not just seen in training and websites and services, it’s in the research too; many researchers have used “multisensory” to describe the entire “evidence-based” approach to reading instruction. This has caused significant confusion when people are looking for evidential support of just the multisensory component, as I’ll outline in more detail below.

So where did this all start? The idea that a multisensory component for reading instruction was essential, is commonly traced back to Orton of the Orton-Gillingham approach. In brief, the Orton-Gillingham approach uses all the above listed components to deliver phonics-based reading instruction. Orton had a theory about the dyslexic brain and why it reversed letters, to do with the interaction between the brain’s hemispheres. The requirement to use a multisensory component rested on that theory. His theory was wrong (see Snow & Bowen, 2017, p. 243, and Elliott & Gringorenko, 2014, p. xvi).

The rationale for continuing to use the multisensory component, started to be linked to the theory of “learning styles” – that different children learned best through different sensory modalities so we need to alter our teaching to take this into account by including as many senses as possible. Learning styles theory is also wrong (see my previous post on this topic for a detailed discussion and references). Though the learning styles theory in particular continues for many as the rationale for using a multisensory component, there is a stronger case put forward now that actually we need a multisensory component because of how the brain processes memory.

The memory theory is certainly stronger than the other rationales. It has some good brain-based research to support it (though there is still a lot of controversy over the use of brain scans to justify changes in teaching approaches, e.g. see Shanahan, 2024). There are some sound theories and related evidence in other fields, to support the idea that using a multisensory approach to reading could improve outcomes due to memory.

However, the move from the theory to finding support for it in evidence, hasn’t been so smooth. In fact, the lack of evidence for the multisensory component at all has become one of the most persistent complaints against modern, popular “evidence-based” approaches to teaching reading:

“Despite the enthusiasm for multisensory approaches held by many specialist dyslexia teachers… the theoretical grounds and scientific rationale for their use are questionable” (Elliot & Grigorenko, 2014).

“[R]esearch has offered no compelling evidence to suggest that a multisensory approach to instruction is advantageous for students with learning or reading disabilities” (Vaughn & Linan-Thompson, 2003).

“The appeal of multisensory instruction endures even though it has been poorly defined and is not well validated in existing intervention studies” (Farrell & Sherman, in Birsch, 2011).

“[A]lthough the modality-matched approach (e.g., programs for “visual learners”) and the multisensory approach (e.g., using visual, auditory, and/or kinesthetic cues in instruction) are popular with many teachers, they have not been supported empirically” (Truscott et al, 2004).

“The one component of OG [Orton-Gillingham] approaches for which there is less research evidence is the kinesthetic/tactile instructional component, often called the “multisensory” component (Al Otaiba et al., 2018). There is little research to suggest this component adds value to explicit and systematic phonics-based instruction” (Solari et al, 2021).

“[T]here is not sufficient evidence for the benefit of reading instructional programs that describe themselves as multisensory to require that school districts use them in place of other evidence-based instructional approaches (i.e., other explicit, systematic approaches to foundational skills instruction) that do not describe themselves as multisensory)” (Hall et al., 2022).

Those are only a selection of the many similar statements I came across in the research; you get the idea. (The reason they often refer to dyslexic students above, is because the strongest advocacy for a multisensory approach is specifically for the high-risk readers, with many claiming it is advantageous for any learning reader but crucial for dyslexic students. The research isn’t showing clear support yet for a multisensory component for either group of learners.)

Before we go any further, we need to add some clarity to what is meant by the multisensory component in reading instruction. It’s most consistently used to refer to the simultaneous use of two or more sensory modalities, and sometimes explicitly used to refer to the simultaneous use of three modalities, specifically visual, audial, and kinaesthetic. It requires particular effort to go into making sure a sensory modality is being used. Some organisations offer training for teachers specifically targeted to the integration of the sensory modalities.

The multisensory requirement goes beyond making sure the teaching modality matches the actual task. If you read my piece on learning styles, you’ll remember that research supports using a sensory approach that is directly relevant and linked to the content of what is being taught (e.g., using a visual and kinaesthetic approach to map-making, and audio and visual for music etc.). That is to say, if the task genuinely and clearly requires a particular sensory approach, you should absolutely use that approach – there is nothing controversial there at all. That’s just sensible (and somewhat obvious, I hope).

Taking an extra step of intentionally finding a way to insert another simultaneous sensory modality that is not otherwise required by the task, is what’s at issue. (Some of the most notorious examples are shaving foam letter activities, creating words out of play-dough, and rainbow letters for spelling, but they are the tip of the iceberg.) This extra step can be time-consuming, not just for the teacher doing the lesson-plan, but for the student receiving the lesson. So, the use of that extra time and resources (especially for children with special needs who have fallen behind), must be clearly supported by evidence. In the case of the multisensory component for reading, it isn’t.

But what about all those studies that found benefits from using a multisensory approach? There are heaps of them, and they are widely used in support of including a multisensory component. But have a look at those studies. What you find, almost universally, is they actually used an evidence-based approach to teaching reading (structured, cumulative, explicit, included phonics etc.) that happened to include a multisensory component, and compared it to an approach to teaching reading that is a Whole Language or Balanced Literacy approach. To clarify, they took an approach that includes very well-supported elements for any reading approach (including phonics) and compared them to approaches that don’t include those elements. You cannot conclude from those studies that the multisensory component was what made the difference, especially when we have decades of evidence already establishing the effectiveness of the other components that were being used.

If you want to reach the conclusion that the multisensory aspect is crucial, you need to take two approaches that include all the evidence-based components (phonics, structured etc.) with only one difference between them: the multisensory component. While the other elements have solid decades of research from multiple fields supporting them, the multisensory component does not, so you need to isolate that component if you want to draw a conclusion about its independent necessity.

Why haven’t people done this type of study? They have! And so far, it hasn’t found that the multisensory component makes a reliable or significant difference, either for typically-developing readers or for dyslexics. There aren’t nearly enough of these studies to draw a firm conclusion either way yet, but until there is it’s hard to justify extra time and money being spent on incorporating something without clear evidence behind it.

Studies have also tested the idea that it isn’t the use of multiple sensory modalities that is the key to improved memory and learning, but the use of multiple exposures and cues – regardless of the sensory modality (e.g., it could be a repeat of the same modality) – that might lead to improved outcomes. The idea of multiple exposure to something supporting memory is well-established, in theory and in evidence. It looks like the number of cues and exposures could be key rather than sensory modalities (though there may be some differences in different ages of children in regards to sensitivity to sensory modalities – something that again requires more research).

Another important factor for learning is attention: using multiple sensory modalities could theoretically capture the attention of children better. This is another claim made in support of using a multisensory approach. In that case though, the factor that should be focused on is what genuinely increases children’s attention to the task. It may be that the classroom environment, the noise and light levels and distractions in the classroom, the physical interactions by the teacher and other students, are all affecting attention, or perhaps the child requires medication to improve their attention to learning tasks – those are not addressed by adding sensory components to the content being taught. Furthermore, using multiple simultaneous sensory modalities could overwhelm a child, especially a child with special needs, making it harder for them to focus on the content itself and creating distractions both within the classroom and within the lesson. So, we need to be careful that if we’re adding sensory modalities because we think it will improve attention, that we are actually achieving that outcome and not working against it.

It's possible that a multisensory approach improves memory and improves attention and therefore improves outcomes when used for reading instruction. It’s possible. It’s backed by various theories and some expert opinions. But, at this stage, the studies trying to find the actual measurable improvement caused directly by the multisensory component, are struggling. There is certainly insufficient evidence at this time to require teachers to train in the multisensory component, to require them to implement the multisensory component in their teaching for reading, and to pour extra money and time into it for that purpose. In turn, there is insufficient evidence to reject approaches that do not explicitly use a multisensory component.

In summary, be very wary of claims that studies back using the multisensory component for teaching reading – studies back the use of the other evidence-based and well-established components, not the multisensory aspect. Not yet, anyway. It’s theoretically useful to explicitly use simultaneous multisensory components, and studies in other areas and fields certainly support exploring it, but there are reasons to doubt its efficacy too and to be careful about its implementation.

Research absolutely supports using appropriate sensory modalities as dictated by the task, and in the task of reading those are most clearly visual and audial approaches. There is nothing wrong with using the appropriate sensory modalities. If there are other tasks involved in literacy tuition which absolutely require, for example, a kinaesthetic approach, such as for handwriting and forming the mouth correctly to create certain sounds, then again that is good and fine; it’s the forcing in of extra sensory modalities unrequired by the actual task, which should make you pause and dig deeper before you spend any more of your, and your student’s or child’s, precious and limited time on them.

***

Special thanks to Cathryn Bjarnesen for helping me locate some of the readings I’d misplaced, and for suggesting further readings on this topic.

Helpful resources to explore the issues discussed above:

• “Multisensory Teaching Basic Language Skills” (Birsch, 2011)

• “Making Sense of Interventions for Children with Developmental Disorders (Bowen & Snow, 2017)

• “The Dyslexia Debate” (Elliott & Grigorenko, 2014)

• “Are two cues always better than one? The role of multiple intra-sensory cues compared to multi-cross-sensory cues in children's incidental category learning” (Broadbent et al, 2020)

• “Forty Years of Reading Intervention Research for Elementary Students with or at Risk for Dyslexia: A Systematic Review and Meta-Analysis” (Hall et al., 2022)

• “The impact of multisensory instruction on learning letter names and sounds, word reading, and spelling” (Schlesinger & Gray, 2017)

• “What Does Brain Science Have To Say About Teaching Reading? Does It Matter?” (Shanahan, 2024).

• “What Does Science Say About Orton-Gillingham Interventions? An Explanation and Commentary on the Stevens et al. (2021) Meta-Analysis” (Solari et al, 2021)

• “Special Education” (Truscott et al., 2004)

• “What Is Special About Special Education for Students with Learning Disabilities?” (Vaughn & Linan-Thompson, 2003)