Research issue 06 · September 6, 2026

Weekly research brief · Dyslexia

What matters.
What to do with it.

A concise, evidence-ranked scan for educational therapists—without the research-paper fog.

7 minute read3 research resources3 practice prompts
This week’s signal

This week is about verification: separate evidence-backed instructional ingredients from appealing labels, check what practitioners can actually explain and apply, and judge growth against the learner's starting gap—not improvement alone.

What matters this week

Check the claim, test the knowledge, follow the gap

Vote on each one. Your feedback shapes the next scan.

01
Medium confidenceNew development · Narrative research synthesis · First published September 3, 2026

Supported ingredients matter more than a branded label—or an appealing add-on.

A new practitioner-focused synthesis reviews current dyslexia definitions and evidence for common remediation claims. It concludes that students with dyslexia need the same evidence-based structured literacy ingredients used broadly—explicit, systematic, sequential, diagnostic instruction—delivered at greater intensity when needed. It finds no established reading benefit for coloured overlays, dyslexic fonts, vision therapy, action videogames, or general cognitive training. Orton–Gillingham includes supported instructional features, but the multisensory element distinguishing it has not independently shown added benefit; a cited comparison was underpowered. This is a narrative synthesis, not a preregistered systematic review, and it generated no new participant data. It clarifies the 2025 International Dyslexia Association definition rather than announcing another 2026 definition. Authors reported no funding or conflicts.

Why it matters at edcahoots

Describe educational therapy by its observable instructional ingredients, intensity, progress checks, and transfer—not by a brand or an unsupported mechanism. Use the article as a source-backed family FAQ.

Read the primary source Austin et al. · Dyslexia
02
Early/Uncertain confidenceNew development · Exploratory cross-sectional study · Published September 4, 2026

Experience and confidence did not substitute for an objective knowledge check.

An exploratory study directly assessed 214 Brazilian public-school teachers and administrators serving children ages 4–10. Average accuracy was 67.1% on dyslexia items and 40.3% on phonological knowledge; 57.9% missed an item separating reading skill from intellectual ability. Additional formal training was associated with higher dyslexia knowledge (Cohen's d = 0.75), while years of service, teaching confidence, and—after school clustering was considered—experience with dyslexia were not significant predictors. The result supports checking knowledge rather than inferring it from tenure or confidence, but it does not show that a quiz or training improves student reading. One convenience sample from a single Brazilian municipality, Portuguese-language measures, uneven role groups, and no student outcomes limit transfer to EdCahoots. Publication charges were funded by CAPES; authors reported no research funding or conflicts.

Why it matters at edcahoots

Use a short, low-stakes staff calibration with applied phoneme, morpheme, orthography, and misconception scenarios. Let results guide coaching without treating the score as a performance verdict.

Read the primary source Pollo et al. · Reading and Writing
03
Medium confidenceNew development · Longitudinal measurement study · Published September 6, 2026

Growth can be real while an early skill gap remains stubbornly intact.

Researchers followed 196 Spanish-speaking kindergarteners at four Canary Islands schools across fall, winter, and spring, including 49 classified as at risk for reading difficulty. Five brief curriculum-based measures covering phonological awareness, alphabetic knowledge, print concepts, and vocabulary showed a stable underlying structure over time and detected growth on most indicators. Yet the at-risk group began lower and generally followed parallel growth trajectories under usual Tier 1 instruction, so initial gaps persisted rather than narrowing. The study supports repeated component-skill monitoring, not a specific intervention or dyslexia diagnosis. Convenience sampling, four schools, a Spanish transparent orthography, one measurement system, and no randomized targeted treatment constrain generalization to English-speaking Los Angeles clients. Accessible records did not surface funding or conflict details.

Why it matters at edcahoots

At each review, report both the learner's own growth and whether the gap to an appropriate benchmark is narrowing. A rising score alone may not justify maintaining the same plan.

Read the primary source Jiménez et al. · Reading and Writing

Worth watching

Promising, but not practice-changing yet.

A March 2026 systematic review and meta-analysis of 26 case-control studies found small average binocular and eye-movement differences in people with dyslexia. Those observational differences may help explain visual effort, but they do not establish a visual cause of dyslexia or show that vision therapy improves reading. Treat this as a mechanism to watch, not an intervention recommendation.

Questions for the team

Turn evidence into a useful conversation.

01

Could EdCahoots add a short evidence check before adopting a branded method, tool, or family-facing claim?

02

Which phoneme, morpheme, orthography, and dyslexia misconceptions should appear in a brief staff calibration this fall?

03

Do progress reviews distinguish absolute growth from whether a learner's relative skill gap is actually narrowing?

Educational therapy research, filtered for practice.

Research information only—not individualized medical, diagnostic, or legal advice. Please do not include student names in feedback.

Weekly dyslexia research briefSeptember 6, 2026

What matters this week

Check the claim, test the knowledge, follow the gap

Three new studies sharpen quality control: scrutinize unsupported dyslexia claims, calibrate staff knowledge objectively, and follow component-skill gaps across repeated progress checks.
1

Medium confidence · New development · Narrative research synthesis · First published September 3, 2026

Supported ingredients matter more than a branded label—or an appealing add-on.

Takeaway

A 2026 synthesis favors explicit, systematic literacy ingredients over branded methods, visual fixes, cognitive training, or unsupported multisensory claims.

Practice implication

Audit family-facing claims and describe observable instruction, intensity, progress, and transfer.
2

Early/Uncertain confidence · New development · Exploratory cross-sectional study · Published September 4, 2026

Experience and confidence did not substitute for an objective knowledge check.

Takeaway

Among 214 educators, formal training—not tenure or confidence—best tracked dyslexia knowledge; phonological accuracy averaged only 40.3%.

Practice implication

Use a low-stakes knowledge calibration to target coaching rather than assuming experience equals mastery.
3

Medium confidence · New development · Longitudinal measurement study · Published September 6, 2026

Growth can be real while an early skill gap remains stubbornly intact.

Takeaway

Repeated measures detected kindergarten growth, yet at-risk learners' lower starting position generally persisted across the year under usual instruction.

Practice implication

Track both absolute gains and whether the benchmark gap narrows before maintaining dosage.

Questions for the team

• Could EdCahoots add a short evidence check before adopting a branded method, tool, or family-facing claim?

• Which phoneme, morpheme, orthography, and dyslexia misconceptions should appear in a brief staff calibration this fall?

Primary sources

1. Austin et al. · Dyslexia: https://onlinelibrary.wiley.com/doi/10.1002/dys.70049

2. Pollo et al. · Reading and Writing: https://link.springer.com/article/10.1007/s11145-026-10899-x

3. Jiménez et al. · Reading and Writing: https://link.springer.com/article/10.1007/s11145-026-10890-6

Built by Brantley