The Delphi Dyslexia Definition 

Listening to the New Definition: Preventing Reading and Spelling Difficulties Before They Begin
 

The new Delphi Dyslexia Definition makes one thing clear: dyslexia is defined by struggles with reading and spelling, underpinned by weaknesses in phonemic awareness and phonological working memory. In other words, children aren’t diagnosed until they’re already struggling. But what if we could prevent that struggle from happening in the first place?
 

That’s exactly what our Phase 1 with Phonemies is designed to do.

Rather than diving straight into letters and sounds—as most synthetic phonics programmes now do—our 10-day Phase 1 plan ensures that all children develop the core listening and sound-processing skills they need before graphemes are introduced. This is especially important for children who are neurodivergent, non-speaking, or at risk of dyslexia.
 

Children learn to:

• Recognise and isolate individual speech sounds (phonemes),

• Hold sound sequences in working memory,

• Use visual phoneme anchors (Phonemies) to support auditory processing,

• Begin spelling and decoding using meaningful sound-to-letter connections.

By the end of the 10 days, children are already decoding and spelling with the letters s, a, t, p, i, n — but crucially, with the internal schema to understand what they’re doing, not just memorising or guessing.
 

Why This Matters
 

Most phonics programmes only teach around 100 grapheme–phoneme correspondences (GPCs) — as seen in the DfE-recommended scope and sequence — a fraction of what children need to read fluently in English. The Spelling Clouds show the whole code. For children to move into the self-teaching phase, where they begin to decode unfamiliar words independently, they must first understand how the code works. They can then use their existing knowledge the figure out new correspondences, and they learn more and more by exploring words - ideally through reading. 
 

But many don’t. They can’t hear the smallest sound units, segment, or blend them. Doing this with graphemes within an opaque orthography means it just gets harder and harder to understand. Children often try to memorise whole words as a result — or adults try to teach words that way. The difficulties are ignored, rather than addressed. They never get to the stage of reading and writing lots of words, or deep reading for pleasure.
 

An opaque orthography means that speech sounds can be represented by many different ‘sound pictures’ (graphemes). For example, the sound /s/ can be represented by at least 14 graphemes — see the Spelling Clouds. It also means that a single grapheme can represent multiple phonemes. Think of the letter a, which maps to at least nine different speech sounds.

In our Phase 1 with Phonemies plan, children see the speech sounds using Phonemies, without linking them to letters too early. The Innovate UK-funded tech introduced during the plan can then be used to make synthetic phonics learning easier — both at home and in school.

Even though 4 in 5 children pass the Year 1 Phonics Screening Check, 27% eleven-year-olds still leave primary school unable to read at the minimum expected level. That tells us passing the PSC doesn’t guarantee long-term reading success — especially if foundational skills were never secured. So much more is needed.
 

Phase 1 with Phonemies is designed to help Reception teachers identify which children are at risk, while offering an intervention even before phonics begins. It also empowers children to use the tech when words are difficult to map — in either direction. They can use it to check words they can’t (yet) decode, and those they cannot (yet) encode.
 

Activities are designed not only to develop phonemic awareness and phonological working memory, but to speed up the rate at which speech sounds, spelling, and meaning are ‘glued’ together and stored in the orthographic lexicon (the brain’s word bank). That way, the next time the child sees the word, they recognise it instantly — and can spell it without difficulty. They begin to know when words ‘look right’ — and when they don’t. We call this the Speech Sound Mapping Theory.