# Super Dyslexia

> Super Dyslexia V4 is an independently drawn typography project by Karl Lillrud. It contains two complementary families, Harmonia and High Speed, each with Reading and Display cuts. The public site exposes an 82 record evidence registry and 24 design decision proof ledger.

## Primary resources

- [Project home](https://superdyslexia.com/): family selector, Type Studio, evidence overview and access routes.
- [Research library](https://superdyslexia.com/research.html): 82 structured evidence records with source links.
- [Methodology](https://superdyslexia.com/methodology.html): 24 public design decisions and claim boundaries.
- [Language and character coverage](https://superdyslexia.com/coverage.html): audited published coverage and script boundaries.
- [Accessibility](https://superdyslexia.com/accessibility.html): accessibility approach.
- [AI access](https://superdyslexia.com/ai.html): machine access, attribution and action boundaries.
- [Licensing](https://superdyslexia.com/license.html): personal, professional and educator routes.

## Machine readable data

- https://superdyslexia.com/data/evidence_registry.json
- https://superdyslexia.com/data/design_decisions.json
- https://superdyslexia.com/data/font_coverage.json
- https://superdyslexia.com/ai/site-profile.json
- https://superdyslexia.com/ai/content-index.json
- https://superdyslexia.com/ai/actions.json

## Attribution

Creator: Karl Lillrud
Project: Super Dyslexia
Official site: https://superdyslexia.com/
Creator site: https://www.karllillrud.com/

## Claim boundary

Super Dyslexia is research designed and evidence traceable. It is not a medical device, diagnosis or treatment and is not presented as scientifically proven to improve reading performance for every dyslexic reader.

## Crawlable project content


### index.md

# Super Dyslexia by Karl Lillrud

Last reviewed: 6 October 2026

Super Dyslexia V4 contains two complementary original font families: **Harmonia** and **High Speed**. Harmonia prioritizes calm familiarity and contemporary professional texture. High Speed increases local character distinction while preserving the same design system.

The website can be switched between both families from the Font selector in the top menu. The selection persists locally across public pages.

## Evidence trail

82 structured research records and 24 documented design decisions are published on the site. The project is research designed and evidence traceable, but it is not a medical device, diagnosis or treatment and is not yet claimed to improve reading performance for every dyslexic reader.

## Type Studio

The live Type Studio exposes the actual shipping fonts and lets readers compare Harmonia and High Speed while controlling Reading or Display cut, weight, size, letter spacing, word spacing and line height.

## Published coverage

The current release covers Latin, extended Latin, Vietnamese, modern Greek, modern Cyrillic, combining marks, punctuation, currencies, arrows, mathematics and related symbols. See https://superdyslexia.com/coverage.html for boundaries.

Creator: Karl Lillrud
Official site: https://superdyslexia.com/


### research.md

# Super Dyslexia evidence registry

Canonical page: https://superdyslexia.com/research.html

Record count: 82

## E001: International Dyslexia Association 2025 Definition of Dyslexia

Domain: definition
Year: 2025
Evidence type: scientific consensus definition
Confidence: high

Claim: Dyslexia involves persistent difficulty with word reading and or spelling accuracy, speed, or both. Presentation varies with orthography and reflects interacting genetic, neurobiological and environmental influences.

Design implication: Do not model dyslexia as a single visual defect. Build a multidimensional and orthography aware system.

Source: https://eida.org/definition-of-dyslexia/

## E002: Does font improve reading in dyslexic children?

Domain: font
Year: 2026
Evidence type: meta analysis
Confidence: high

Claim: Specialist dyslexia fonts do not show a reliable overall advantage over standard fonts for reading speed or accuracy.

Design implication: Do not make a fixed specialist font the core hypothesis. Optimize measurable typographic variables and personalize them.

Source: https://pubmed.ncbi.nlm.nih.gov/42536336/

## E003: Visual attention span deficit in developmental dyslexia: A meta-analysis

Domain: visual_attention
Year: 2023
Evidence type: meta analysis
Confidence: high

Claim: Visual attention span performance is on average poorer in dyslexic groups, with substantial moderation by task, stimulus type, writing system and comparison group.

Design implication: Support reduced simultaneous visual load and test personalization rather than assuming one universal visual span.

Source: https://pubmed.ncbi.nlm.nih.gov/37683312/

## E004: Visual attention span capacity in developmental dyslexia: A meta-analysis

Domain: visual_attention
Year: 2023
Evidence type: meta analysis
Confidence: high

Claim: Dyslexic readers show lower mean visual attention span scores and greater individual variability than typical readers.

Design implication: Treat reader variability as a first class design requirement. Adaptive settings are preferable to a single preset.

Source: https://pubmed.ncbi.nlm.nih.gov/36867955/

## E005: Is orthographic knowledge a strength or a weakness in individuals with dyslexia?

Domain: orthographic_processing
Year: 2021
Evidence type: meta analysis
Confidence: high

Claim: Orthographic knowledge is substantially weaker in dyslexic readers than chronological age peers, while the difference is much smaller against reading level matched controls.

Design implication: Preserve familiar letter skeletons and word forms. Excessively novel glyphs may add orthographic learning burden.

Source: https://pubmed.ncbi.nlm.nih.gov/33712993/

## E006: Phonological skills and their role in learning to read

Domain: phonology
Year: 2012
Evidence type: meta analytic review
Confidence: high

Claim: Phonemic awareness shows a large group difference between dyslexic and age matched typical readers and remains a strong correlate of word reading.

Design implication: Typography cannot replace phonological support. A reading system should optionally integrate text to speech, syllable or phoneme support and explicit decoding aids.

Source: https://pubmed.ncbi.nlm.nih.gov/22250824/

## E007: Reading disabilities in children: A selective meta-analysis of the cognitive literature

Domain: cognition
Year: 2015
Evidence type: meta analysis
Confidence: high

Claim: Children with reading disabilities show group differences across rapid naming, phonological awareness, working memory, short term memory, visual spatial memory and executive processing.

Design implication: Reduce unnecessary memory and navigation demands. Reading support should minimize simultaneous task load.

Source: https://pubmed.ncbi.nlm.nih.gov/25725389/

## E008: Working memory, short-term memory, and reading disabilities

Domain: working_memory
Year: 2009
Evidence type: meta analysis
Confidence: high

Claim: Children with reading disabilities show moderate disadvantages in short term and working memory, especially tasks with verbal and phonological storage demands.

Design implication: Prefer short visual chunks, stable context, visible progress and low memory dependent navigation.

Source: https://pubmed.ncbi.nlm.nih.gov/19255286/

## E009: Do tasks make a difference? Executive function and reading difficulties

Domain: executive_function
Year: 2010
Evidence type: meta analysis
Confidence: moderate

Claim: Executive function differences occur in children with reading difficulties, but effects vary substantially with task demands and definitions.

Design implication: Avoid interfaces that require frequent switching, inhibition or remembering multiple reading controls at once.

Source: https://pubmed.ncbi.nlm.nih.gov/20306629/

## E010: Auditory processing deficit in individuals with dyslexia

Domain: auditory_processing
Year: 2020
Evidence type: meta analysis
Confidence: moderate

Claim: Mismatch negativity studies show small to moderate group differences in auditory speech processing in children and adults with dyslexia.

Design implication: Optional audio support may help some readers but should not be assumed to address a universal auditory deficit.

Source: https://pubmed.ncbi.nlm.nih.gov/32610180/

## E011: Coherence motion perception in developmental dyslexia

Domain: visual_motion
Year: 2010
Evidence type: meta analysis
Confidence: moderate

Claim: Motion coherence perception differs on average between dyslexic and control groups, but the relationship to continuous reading ability is much smaller.

Design implication: Do not use magnocellular or motion theories as the primary font design foundation.

Source: https://pubmed.ncbi.nlm.nih.gov/20957687/

## E012: Is a procedural learning deficit a causal risk factor for developmental language disorder or dyslexia?

Domain: procedural_learning
Year: 2021
Evidence type: meta analytic review
Confidence: moderate

Claim: Evidence does not adequately support a generalized procedural learning deficit as a causal explanation of dyslexia.

Design implication: Do not build typography around a generalized procedural learning theory.

Source: https://pubmed.ncbi.nlm.nih.gov/34166019/

## E013: Is excessive visual crowding causally linked to developmental dyslexia?

Domain: crowding
Year: 2019
Evidence type: multi experiment longitudinal and intervention study
Confidence: moderate_high

Claim: Excessive visual crowding was observed in dyslexic children at unattended locations, spacing improved accuracy, and prereading crowding predicted later poor reading.

Design implication: Provide adjustable spacing and reduce adjacent visual clutter, while preserving clear word boundaries.

Source: https://pubmed.ncbi.nlm.nih.gov/31077708/

## E014: Crowding effects on word identification in developmental dyslexia

Domain: crowding
Year: 2002
Evidence type: controlled experiments
Confidence: moderate

Claim: Dyslexic readers were more sensitive to surrounding visual material and moderate interletter spacing improved vocal reaction time in the dyslexic group.

Design implication: Spacing is a testable control variable, not a universal fixed value.

Source: https://pubmed.ncbi.nlm.nih.gov/12056688/

## E015: Extra-large letter spacing improves reading in dyslexia

Domain: spacing
Year: 2012
Evidence type: controlled reading experiment
Confidence: moderate

Claim: Extra large letter spacing improved reading performance in Italian and French dyslexic children in the tested conditions.

Design implication: Include wider spacing as a candidate setting, but test it against coupled word spacing and individual response.

Source: https://pubmed.ncbi.nlm.nih.gov/22665803/

## E016: Inter-letter spacing, inter-word spacing, and font with dyslexia-friendly features

Domain: spacing
Year: 2020
Evidence type: controlled factorial experiment
Confidence: high

Claim: Specialist letterforms did not improve reading. Increased interletter spacing without adequate interword spacing reduced reading speed.

Design implication: Optimize letter spacing and word spacing jointly. Never increase one independently by default.

Source: https://pubmed.ncbi.nlm.nih.gov/32172467/

## E017: Interword and interletter spacing effects during reading revisited

Domain: spacing
Year: 2016
Evidence type: controlled eye movement experiments
Confidence: high

Claim: The effect of letter spacing depends on the font default and interacts with interword spacing.

Design implication: Store spacing in relative font metrics and evaluate word segmentation, not only raw tracking values.

Source: https://pubmed.ncbi.nlm.nih.gov/27936854/

## E018: A Special Font for People with Dyslexia: Does it Work and, if so, why?

Domain: spacing
Year: 2016
Evidence type: controlled experiment
Confidence: high

Claim: A reported Dyslexie speed advantage disappeared when Arial was matched for within word and between word spacing.

Design implication: Separate glyph shape effects from spacing effects in every experiment.

Source: https://pubmed.ncbi.nlm.nih.gov/27194598/

## E019: Dyslexie font does not benefit reading in children with or without dyslexia

Domain: font
Year: 2017
Evidence type: controlled experiments
Confidence: high

Claim: Children with dyslexia did not read Dyslexie faster or more accurately than Arial, and most preferred Arial.

Design implication: Do not equate subjective specialist appearance with measurable benefit. Familiarity matters.

Source: https://pubmed.ncbi.nlm.nih.gov/29204931/

## E020: Does a specialist typeface affect how fluently children process letters, words, and passages?

Domain: font
Year: 2022
Evidence type: controlled experiment
Confidence: moderate_high

Claim: When spacing was controlled, Dyslexie did not improve word or passage reading versus Calibri, although rapid letter naming was somewhat more fluent.

Design implication: Distinctive glyphs may help letter identification without necessarily improving passage reading. Test both levels separately.

Source: https://pubmed.ncbi.nlm.nih.gov/36054673/

## E021: The effect of print size on reading speed in dyslexia

Domain: font_size
Year: 2006
Evidence type: controlled psychophysical experiment
Confidence: moderate

Claim: Dyslexic children required larger critical print sizes to reach maximum reading speed than reading ability matched controls in this study.

Design implication: Include print size or x height in personalization. Do not assume conventional point size is equivalent across fonts.

Source: https://pubmed.ncbi.nlm.nih.gov/16601822/

## E022: Effect of different font sizes and of spaces between words on eye movement performance

Domain: eye_movement
Year: 2018
Evidence type: eye tracking experiment
Confidence: moderate

Claim: Larger and more spaced text reduced fixation duration in both dyslexic and non dyslexic children in the tested material.

Design implication: Use fixation duration as an optimization signal, not as a diagnostic cause. Test size and spacing together.

Source: https://pubmed.ncbi.nlm.nih.gov/30291918/

## E023: AI-enabled eye-movement and emerging multimodal frameworks for precision dyslexia screening

Domain: eye_movement
Year: 2026
Evidence type: systematic review
Confidence: moderate_high

Claim: Dyslexic readers commonly show longer fixation durations, more regressions and lower saccadic efficiency, but machine learning studies are heterogeneous and often at risk of overfitting.

Design implication: Use gaze features to evaluate adaptation and reader load. Never infer a clinical diagnosis from a single reading session.

Source: https://pubmed.ncbi.nlm.nih.gov/42404570/

## E024: Evaluation of ocular movements in patients with dyslexia

Domain: eye_movement
Year: 2015
Evidence type: eye movement study
Confidence: moderate

Claim: Reading related saccade and regression differences were present, while basic tracking saccades were not significantly different, supporting the interpretation that many reading eye movement abnormalities may be secondary to linguistic processing difficulty.

Design implication: Do not design a font around the assumption of primary oculomotor dysfunction.

Source: https://pubmed.ncbi.nlm.nih.gov/25804764/

## E025: The effect of coloured overlays and lenses on reading

Domain: color
Year: 2016
Evidence type: systematic review
Confidence: high

Claim: Higher quality evidence does not support colored overlays or lenses as a general treatment for reading difficulty or dyslexia.

Design implication: Do not prescribe colored backgrounds as a default dyslexia feature. Allow user preference without therapeutic claims.

Source: https://pubmed.ncbi.nlm.nih.gov/27580753/

## E026: Orthographic depth and developmental dyslexia

Domain: orthography
Year: 2021
Evidence type: systematic review and meta analysis
Confidence: high

Claim: Orthographic depth moderates some manifestations of dyslexia, while reading speed remains a reliable discriminator across European orthographies.

Design implication: Font and reading algorithms must be language and writing system aware. English results cannot simply be universalized.

Source: https://pubmed.ncbi.nlm.nih.gov/33982221/

## E027: Culture Shapes the Neural Functional Alteration and Its Development in Children With Dyslexia

Domain: cross_linguistic_neuroscience
Year: 2026
Evidence type: meta analysis
Confidence: high

Claim: Across Chinese, English and German, dyslexia showed shared hypoactivation in left fusiform and inferior parietal regions, alongside language specific differences.

Design implication: There are shared reading network constraints, but optimization needs script specific models and test corpora.

Source: https://pubmed.ncbi.nlm.nih.gov/42549619/

## E028: Functional Connectivity Alterations in Developmental Dyslexia

Domain: neuroscience
Year: 2026
Evidence type: functional connectivity meta analysis
Confidence: moderate_high

Claim: Reduced functional connectivity between left inferior frontal and left fusiform regions was found across task and resting state studies.

Design implication: The product should optimize fluent visual word recognition and language integration, but neural findings do not specify a single glyph shape.

Source: https://pubmed.ncbi.nlm.nih.gov/41229151/

## E029: Neuroimaging meta-analyses of phonological awareness deficits in children with developmental dyslexia

Domain: neuroscience
Year: 2026
Evidence type: neuroimaging meta analysis
Confidence: moderate_high

Claim: Children with dyslexia showed reduced activation in left inferior temporal and occipital regions during phonological awareness tasks.

Design implication: Visual and phonological processes interact. Reading support should be multimodal rather than purely typographic.

Source: https://pubmed.ncbi.nlm.nih.gov/41784900/

## E030: Reading intervention and neuroplasticity

Domain: neuroplasticity
Year: 2022
Evidence type: systematic review and meta analysis
Confidence: high

Claim: Reading intervention studies report brain changes across distributed reading and domain general systems, but the quantitative meta analysis did not find a single consistent activation change.

Design implication: Avoid simplistic normalize the dyslexic brain claims. Measure behavioral outcomes directly.

Source: https://pubmed.ncbi.nlm.nih.gov/34856223/

## E031: A meta-analysis of morphological interventions

Domain: morphology
Year: 2010
Evidence type: meta analysis
Confidence: high

Claim: Morphological instruction produced a small positive overall effect on literacy outcomes in learners with literacy difficulties, with effects across phonological awareness, morphology, vocabulary, comprehension and spelling.

Design implication: An advanced reading system can optionally expose meaningful morpheme structure, especially for complex words, but should test cognitive load.

Source: https://pubmed.ncbi.nlm.nih.gov/20799003/

## E032: A Meta-Analysis of Phonemic Awareness Instruction

Domain: intervention
Year: 2022
Evidence type: meta analysis
Confidence: high

Claim: Phonemic awareness instruction has positive effects for children suspected of reading disability, especially segmentation and composite outcomes.

Design implication: If the product extends beyond typography, phoneme and grapheme support has a stronger evidence basis than a specialist font alone.

Source: https://pubmed.ncbi.nlm.nih.gov/35858272/

## E033: Prevalence of Developmental Dyslexia in Primary School Children

Domain: population
Year: 2022
Evidence type: systematic review and meta analysis
Confidence: high

Claim: Pooled prevalence of developmental dyslexia in primary school children was about 7.1 percent, with no significant pooled prevalence difference across alphabetic versus logographic scripts or shallow versus deep orthographies.

Design implication: Design for a large and globally diverse user population, not a niche visual style.

Source: https://pubmed.ncbi.nlm.nih.gov/35204003/

## E034: Reduced perceptual span in dyslexic children: a gaze-contingent eye-tracking study

Domain: perceptual_span
Year: 2026
Evidence type: gaze contingent eye tracking study
Confidence: moderate_high

Claim: Children with dyslexia showed greater foveal load and reduced parafoveal processing in a moving window reading task. Window size affected their reading speed less than controls, consistent with a smaller effective perceptual span.

Design implication: Do not assume that showing more upcoming text is always useful. Test whether reducing visible peripheral text or visual clutter improves efficiency for a specific reader.

Source: https://pubmed.ncbi.nlm.nih.gov/41370866/

## E035: The perceptual span in dyslexic reading and visual search

Domain: perceptual_span
Year: 2024
Evidence type: moving window eye tracking study
Confidence: moderate

Claim: Evidence for a reduced perceptual span in dyslexic adult reading was equivocal, and there was no evidence for a smaller perceptual span in visual search, although search accuracy and saccade amplitudes differed.

Design implication: Perceptual span should be measured per reader rather than assumed from diagnosis.

Source: https://pubmed.ncbi.nlm.nih.gov/39155549/

## E036: Parafoveal Processing and Transposed-Letter Effects in Developmental Dyslexic Reading

Domain: parafoveal_processing
Year: 2024
Evidence type: boundary paradigm eye tracking study
Confidence: moderate_high

Claim: Dyslexic children obtained parafoveal preview benefit and encoded letter identity separately from letter position, but later measures suggested reduced sensitivity to initial letter transpositions.

Design implication: Preserve strong initial letter identity cues and avoid typography that makes first letter position harder to discriminate.

Source: https://pubmed.ncbi.nlm.nih.gov/39627982/

## E037: Parafoveal processing and transposed-letter effects in dyslexic reading

Domain: parafoveal_processing
Year: 2022
Evidence type: boundary paradigm eye tracking study
Confidence: moderate_high

Claim: Adults with dyslexia showed parafoveal orthographic preview benefits, but early positional encoding of initial letters differed from skilled readers.

Design implication: Initial character clarity and stable letter position cues deserve explicit testing in adaptive glyph design.

Source: https://pubmed.ncbi.nlm.nih.gov/35818161/

## E038: Word length and word frequency affect eye movements in dyslexic children reading in a regular German orthography

Domain: lexical_processing
Year: 2010
Evidence type: eye movement experiment
Confidence: moderate_high

Claim: In German dyslexic children, fixation durations and saccade counts increased with word length and frequency difficulty, and regressions increased for low frequency words.

Design implication: Calibration texts must manipulate word length and lexical frequency independently because typography effects may change with lexical difficulty.

Source: https://pubmed.ncbi.nlm.nih.gov/20225005/

## E039: Eye fixation patterns among dyslexic and normal readers: effects of word length and word frequency

Domain: lexical_processing
Year: 1995
Evidence type: eye movement experiment
Confidence: moderate_high

Claim: Long and low frequency words increased gaze duration, reinspection time, and fixation count in dyslexic and younger reading level controls. The pattern supported lexical processing difficulty rather than a unique oculomotor deficit.

Design implication: Use lexical difficulty as a covariate when interpreting eye movement differences.

Source: https://pubmed.ncbi.nlm.nih.gov/7490575/

## E040: Eye Movements of French Dyslexic Adults While Reading Texts

Domain: lexical_processing
Year: 2025
Evidence type: eye tracking study
Confidence: moderate

Claim: French adults with dyslexia made more errors and more progressive and regressive fixations, with worse performance for meaningless text, suggesting greater reliance on semantic context during difficult reading.

Design implication: Test reading support under both supportive semantic context and low context material because an aid can appear effective only when context compensates for decoding difficulty.

Source: https://pubmed.ncbi.nlm.nih.gov/40722285/

## E041: An In-Depth Investigation of Eye Movement Profile of Dyslexic Readers Using a Standardized Text-Reading Aloud Task in French

Domain: eye_voice_span
Year: 2025
Evidence type: eye tracking and oral reading study
Confidence: moderate

Claim: Adults with dyslexia showed persistent word length effects and a reduced and unstable eye voice span during oral reading, consistent with continued reliance on sublexical decoding.

Design implication: For oral reading support, measure eye voice coordination and not only gaze metrics.

Source: https://pubmed.ncbi.nlm.nih.gov/41594959/

## E042: An experimental eye-tracking study of text adaptation for readers with dyslexia

Domain: text_adaptation
Year: 2021
Evidence type: eye tracking experiment
Confidence: moderate_high

Claim: Visual support and lexical simplification facilitated sentence processing. Benefits from lexical simplification were larger for readers with lower print exposure and smaller vocabularies.

Design implication: Adaptive reading should be able to change lexical difficulty and add optional semantic support, but should personalize by vocabulary and print exposure.

Source: https://pubmed.ncbi.nlm.nih.gov/33580863/

## E043: Individual differences in the perceptual span during reading

Domain: perceptual_span
Year: 2015
Evidence type: moving window eye tracking study
Confidence: moderate_high

Claim: In general readers, perceptual span size was associated with language ability rather than oculomotor processing speed.

Design implication: Treat perceptual span partly as a consequence of language processing efficiency, not merely a visual capacity.

Source: https://pubmed.ncbi.nlm.nih.gov/26059082/

## E044: Eye movements and the perceptual span in beginning and skilled readers

Domain: perceptual_span
Year: 1986
Evidence type: developmental eye tracking experiment
Confidence: moderate

Claim: Perceptual span changes with reading proficiency and text difficulty, and a smaller span was not sufficient to explain slower beginning reading.

Design implication: Do not infer that a small span is itself the primary cause of poor reading. Optimize foveal processing and linguistic accessibility too.

Source: https://pubmed.ncbi.nlm.nih.gov/3701249/

## E045: Language-universal and script-specific factors in the recognition of letters in visual crowding

Domain: script_specific_processing
Year: 2024
Evidence type: visual crowding experiment
Confidence: moderate

Claim: In Hebrew readers, crowded letter recognition reflected both general lexical effects and script specific transitional probability effects tied to right to left reading direction.

Design implication: Confusability, contextual letter probability, and reading direction must be modeled separately for each script.

Source: https://pubmed.ncbi.nlm.nih.gov/38070387/

## E046: Effects of word length and word frequency among dyslexic, ADHD-I and typical readers

Domain: comorbidity
Year: 2023
Evidence type: comparative eye tracking study
Confidence: moderate

Claim: Children with dyslexia and children with inattentive ADHD showed different cognitive, linguistic, word length, and word frequency patterns, supporting the need to distinguish overlapping conditions.

Design implication: Do not treat attentional difficulty as synonymous with dyslexia. Calibration should identify performance patterns without diagnosing comorbidity.

Source: https://pubmed.ncbi.nlm.nih.gov/37009492/

## E047: The effect of inter-letter spacing on reading performance and eye movements in typically reading and dyslexic children

Domain: spacing
Year: 2022
Evidence type: eye tracking reading experiment
Confidence: high

Claim: Increased interletter spacing did not improve reading speed, accuracy, or comprehension in dyslexic or typical children, although condensed spacing increased fixation duration and wider spacing shortened fixations in dyslexic children.

Design implication: A gaze improvement alone is insufficient evidence of better reading. Spacing must be judged against speed, accuracy, and comprehension together.

Source: https://www.sciencedirect.com/science/article/pii/S0959475221001353

## E048: Influence of increased letter spacing and font type on the reading ability of dyslexic children

Domain: spacing
Year: 2018
Evidence type: font and spacing experiment
Confidence: moderate

Claim: Dyslexie did not outperform other fonts, while increased letter spacing improved performance and could remove disadvantages seen with less readable fonts.

Design implication: Treat font form and spacing as separable factors. Spacing can sometimes compensate for poor font choices but should not justify a poor base design.

Source: https://pubmed.ncbi.nlm.nih.gov/30094714/

## E049: Good fonts for dyslexia

Domain: font
Year: 2013
Evidence type: within subject eye tracking experiment
Confidence: moderate

Claim: In one study of adults with dyslexia, sans serif, monospaced, and roman styles were associated with better performance than serif, proportional, and italic alternatives. These findings are not sufficient to establish universal font rules.

Design implication: Use sans serif and upright styles as useful starting priors, not hard constraints. Re test within each reader and language.

Source: https://doi.org/10.1145/2513383.2513447

## E050: Serifs and font legibility

Domain: font
Year: 2005
Evidence type: controlled font feature experiment
Confidence: moderate_high

Claim: When serif size was isolated experimentally, serif presence produced little or no meaningful difference in continuous reading speed in normally sighted readers.

Design implication: Do not encode serif versus sans serif as a universal causal rule. Specific glyph geometry matters more than the category label alone.

Source: https://pmc.ncbi.nlm.nih.gov/articles/PMC4612630/

## E051: Does print size matter for reading? A review of findings from vision science and typography

Domain: font_size
Year: 2011
Evidence type: vision science review
Confidence: high

Claim: Reading speed typically rises with print size until a critical print size and then plateaus across a broad fluent range. Angular x height is a more meaningful visual measure than nominal point size.

Design implication: Store and calibrate visual angle and x height, not only CSS pixels or points.

Source: https://pubmed.ncbi.nlm.nih.gov/21828237/

## E052: Text-background polarity affects performance irrespective of ambient illumination and colour contrast

Domain: display
Year: 2007
Evidence type: controlled display experiment
Confidence: moderate_high

Claim: Proofreading performance was better with dark text on a light background than the reverse under the tested conditions, while color contrast could not substitute for luminance contrast.

Design implication: Maintain strong luminance contrast. Positive polarity is a useful default prior but should remain user configurable.

Source: https://pubmed.ncbi.nlm.nih.gov/17510822/

## E053: The advantage of positive text-background polarity is due to high display luminance

Domain: display
Year: 2009
Evidence type: controlled display experiment
Confidence: moderate_high

Claim: When overall display luminance was equated, the usual positive polarity advantage disappeared, suggesting that luminance is an important mechanism behind polarity effects.

Design implication: Model actual luminance and contrast rather than treating dark mode versus light mode as a categorical rule.

Source: https://pubmed.ncbi.nlm.nih.gov/19562598/

## E054: Effects of ambient illumination, contrast polarity, and letter size on text legibility under glance-like reading

Domain: display
Year: 2017
Evidence type: legibility threshold experiment
Confidence: moderate

Claim: Negative polarity under dark ambient illumination produced worse glance like word legibility, with effects interacting with letter size and illumination.

Design implication: Reading presets should account for ambient conditions and effective size, not just background color.

Source: https://pubmed.ncbi.nlm.nih.gov/28166901/

## E055: The Relation between Physiological Parameters and Colour Modifications in Text Background and Overlay during Reading

Domain: color
Year: 2021
Evidence type: multimodal physiological reading experiment
Confidence: low_moderate

Claim: A small study found some color conditions associated with shorter reading duration in dyslexic children, but the design does not override higher level systematic review evidence showing no general treatment effect for colored overlays.

Design implication: Color can be personalized for comfort or tested empirically, but must not be prescribed as a dyslexia treatment.

Source: https://pubmed.ncbi.nlm.nih.gov/33922926/

## E056: Designing a Variable Typeface for Dyslexia-Friendly Reading

Domain: font_variability
Year: 2026
Evidence type: variable typeface design and empirical testing study
Confidence: moderate

Claim: A recent variable typeface study reported that not all conventional dyslexia font features improved performance and argued for adjustable simple to complex letterforms and individual adaptation.

Design implication: Build variability into the font architecture rather than shipping one immutable dyslexia style.

Source: https://hrcak.srce.hr/en/clanak/502779

## E057: Linguistic correlates of Chinese children with developmental dyslexia

Domain: cross_linguistic
Year: 2026
Evidence type: systematic review and meta analysis
Confidence: high

Claim: In Chinese children with developmental dyslexia, large group differences were reported in morphological awareness, phonological awareness, orthographic awareness, and rapid automatized naming, with moderate short term memory differences.

Design implication: A Latin alphabet font model cannot be generalized to Chinese. Morphology, character structure, and orthographic awareness require script specific logic.

Source: https://pubmed.ncbi.nlm.nih.gov/41678943/

## E058: The Percentages of Cognitive Skills Deficits among Chinese Children with Developmental Dyslexia

Domain: cross_linguistic
Year: 2022
Evidence type: systematic review and meta analysis
Confidence: high

Claim: Among Chinese children with developmental dyslexia, no single cognitive deficit characterized all readers. Rapid automatized naming deficits had the highest pooled prevalence among measured skills at about 44 percent.

Design implication: Never assume a universal cognitive subtype. The engine should infer useful settings from observed reading response, not diagnosis labels.

Source: https://pubmed.ncbi.nlm.nih.gov/35624935/

## E059: A meta-analysis of reading-level match dyslexia studies in consistent alphabetic orthographies

Domain: orthography
Year: 2019
Evidence type: meta analysis
Confidence: high

Claim: In highly consistent alphabetic orthographies, reading level matched dyslexic children did not consistently differ in rapid naming, phonological memory, or auditory temporal processing, while phoneme manipulation differences remained.

Design implication: Orthographic consistency changes which deficits are most discriminative. Parameter priors must be language aware.

Source: https://pubmed.ncbi.nlm.nih.gov/31664608/

## E060: Rapid automatized naming: what it is, what it is not, and why it matters

Domain: rapid_naming
Year: 2025
Evidence type: scholarly review
Confidence: moderate_high

Claim: Rapid automatized naming is an important indicator of reading fluency difficulty, but it reflects coordinated processes rather than a single elementary deficit.

Design implication: Use naming fluency to characterize reading automaticity, not as a direct typography prescription.

Source: https://pubmed.ncbi.nlm.nih.gov/39078510/

## E061: Does whole-word shape play a role in visual word recognition?

Domain: word_shape
Year: 2002
Evidence type: lexical decision experiments
Confidence: moderate

Claim: Distorting overall word outline affected lower frequency words more strongly, but results did not support a simple whole word shape theory of recognition.

Design implication: Preserve local letter identity and orthographic coding rather than designing primarily around global word silhouettes.

Source: https://pubmed.ncbi.nlm.nih.gov/12201337/

## E062: The effect of font width on eye movements during reading

Domain: font_width
Year: 2021
Evidence type: eye tracking typography experiment
Confidence: moderate_high

Claim: Large changes in font width altered fixation and saccade patterns even when overall reading speed remained similar, indicating eye movement tradeoffs rather than a simple wider is better rule.

Design implication: Font width should be optimized jointly with spacing and line length. Eye movement changes alone are not proof of improved reading.

Source: https://pubmed.ncbi.nlm.nih.gov/34225106/

## E063: To boldly go where no text has gone before: The effects of boldface letters on eye movements in reading

Domain: font_emphasis
Year: 2025
Evidence type: eye tracking reading experiment
Confidence: moderate_high

Claim: Bolding the first half of every word, as used by Bionic Reading style systems, did not improve reading and produced costs compared with normal unbolded text.

Design implication: Do not assume that partial bolding guides optimal fixation or word recognition. Any emphasis algorithm needs direct evidence.

Source: https://pubmed.ncbi.nlm.nih.gov/40240742/

## E064: Reading on LCD vs e-Ink displays: effects on fatigue and visual strain

Domain: display_quality
Year: 2012
Evidence type: extended reading experiment
Confidence: moderate_high

Claim: Extended reading on modern LCD and electronic ink displays produced similar subjective and objective fatigue when image quality was comparable, suggesting display quality matters more than technology label alone.

Design implication: Optimize resolution, rendering, contrast, and visual size instead of assuming electronic ink is inherently superior.

Source: https://pubmed.ncbi.nlm.nih.gov/22762257/

## E065: Effects of VDT resolution on visual fatigue and readability

Domain: display_quality
Year: 1989
Evidence type: reading fatigue and eye movement experiment
Confidence: moderate

Claim: Higher display resolution particularly improved readability for very small characters, while differences were minimal for sufficiently large characters.

Design implication: Rendering quality becomes especially important near or below critical print size. Small text should trigger stronger anti aliasing and minimum size safeguards.

Source: https://pubmed.ncbi.nlm.nih.gov/2776740/

## E066: Investigating Effects of Typographic Variables on Webpage Reading Through Eye Movements

Domain: web_typography
Year: 2019
Evidence type: eye tracking clinical trial
Confidence: moderate_high

Claim: Multiple realistic webpage typographic variables affected eye movements and reading performance across children and adults with and without dyslexia, with many effects shared across groups.

Design implication: Many good readability changes are universal rather than dyslexia exclusive. Build a high quality reading system first, then personalize where group or individual effects diverge.

Source: https://pubmed.ncbi.nlm.nih.gov/31481744/

## E067: Visual and Oculomotor Function in Developmental Dyslexia: A Systematic Review and Meta-Analysis

Domain: visual_oculomotor
Year: 2026
Evidence type: systematic review and meta analysis
Confidence: high

Claim: Across 26 studies, dyslexic groups showed several binocular and oculomotor differences, including more fixations, longer fixation durations, more regressions and reduced saccade amplitude, while visual acuity and refractive error did not significantly differ.

Design implication: Use eye movement and visual effort measures as validation signals, but do not design the font around a claim that dyslexia is a primary acuity or refractive disorder.

Source: https://pubmed.ncbi.nlm.nih.gov/41801664/

## E068: Are letter identity and letter order encoding functionally separable? Evidence from masked priming with visual similarity and letter transpositions

Domain: visual_word_recognition
Year: 2026
Evidence type: masked priming experiments
Confidence: moderate_high

Claim: Recent visual word recognition experiments continue to show that letter identity and letter order are constrained by visual similarity during early processing.

Design implication: Protect distinctive local letter identity cues and avoid unnecessary visual similarity among confusable glyphs.

Source: https://pubmed.ncbi.nlm.nih.gov/42649345/

## E069: You read best what you read most: An eye tracking study

Domain: font_familiarity
Year: 2021
Evidence type: eye tracking study
Confidence: moderate

Claim: Typeface familiarity influenced reading speed, supporting the value of familiar letter structures rather than novelty for its own sake.

Design implication: Preserve familiar humanist Latin skeletons and make differentiation local rather than globally unusual.

Source: https://pubmed.ncbi.nlm.nih.gov/33828793/

## E070: Investigating letter recognition in the brain by varying typeface: an event-related potential study

Domain: letter_recognition
Year: 2014
Evidence type: ERP experiment
Confidence: moderate

Claim: Typeface fluency affected both early and later stages of letter processing even when the task required abstract letter identity.

Design implication: Avoid unnecessarily disfluent or decorative letter construction in a reading typeface.

Source: https://pubmed.ncbi.nlm.nih.gov/24880492/

## E071: Visual letter similarity effects during sentence reading: Evidence from the boundary technique

Domain: letter_similarity
Year: 2018
Evidence type: boundary paradigm eye tracking study
Confidence: moderate_high

Claim: Visual letter similarity in parafoveal preview measurably affected early word processing during sentence reading.

Design implication: Treat visual confusability as a genuine variable and preserve high identity contrast in critical glyph pairs.

Source: https://pubmed.ncbi.nlm.nih.gov/30119047/

## E072: The role of letter features in visual-word recognition: Evidence from a delayed segment technique

Domain: letter_features
Year: 2016
Evidence type: lexical decision experiments
Confidence: moderate

Claim: Different components of letters contribute unequally to word identification, with midsegments and junction structure carrying important information.

Design implication: Do not concentrate all differentiation in decorative terminals. Preserve informative internal structure, junctions and midsegments.

Source: https://pubmed.ncbi.nlm.nih.gov/27289422/

## E073: The Influence of Typography on Algorithms that Predict the Speed and Comfort of Reading

Domain: typographic_rhythm
Year: 2020
Evidence type: typography and visual comfort study
Confidence: moderate

Claim: Font design, weight, x height relative to line separation, expansion and compression can alter spatial regularity measures associated with likely reading speed and visual comfort.

Design implication: Avoid overly periodic vertical stroke rhythms and treat weight, x height and line spacing as interacting variables.

Source: https://pubmed.ncbi.nlm.nih.gov/32178260/

## E074: High and low luminance letters, acuity reserve, and font effects on reading speed

Domain: display_legibility
Year: 1998
Evidence type: controlled reading experiment
Confidence: moderate

Claim: At high luminance there was no reading rate difference between the serif and sans serif fonts tested, while low luminance favoured the sans serif font in that experiment.

Design implication: Screen and luminance conditions can interact with font design. Protect simple outlines and sufficient acuity reserve at small sizes.

Source: https://pubmed.ncbi.nlm.nih.gov/12116700/

## E075: Efficacy of coloured overlays and lenses for treating reading difficulty: an overview of systematic reviews

Domain: color
Year: 2018
Evidence type: overview of systematic reviews
Confidence: high

Claim: Across systematic reviews there is not a reliable evidence base for recommending coloured overlays or lenses for reading difficulty or discomfort.

Design implication: Keep color as a comfort and brand preference, not a dyslexia treatment claim.

Source: https://pubmed.ncbi.nlm.nih.gov/29633383/

## E076: Alphabetic letter identification: effects of perceivability, similarity, and bias

Domain: letter_identification
Year: 2011
Evidence type: review plus experiments
Confidence: moderate_high

Claim: Letter identification performance reflects perceivability, visual similarity and response bias, providing a framework for evaluating confusable characters.

Design implication: Treat confusability pairs as measurable recognition problems rather than aesthetic preferences.

Source: https://pubmed.ncbi.nlm.nih.gov/22036587/

## E077: Increasing inter-word spacing reduces migration errors and improves reading comprehension in students with dyslexia

Domain: spacing
Year: 2024
Evidence type: controlled reading experiment
Confidence: moderate

Claim: Increasing inter-word spacing improved standardized reading comprehension in students with dyslexia and reduced reported letter and word migration errors in the tested texts.

Design implication: Increase word boundary salience cautiously in the Reading cut while avoiding exaggerated global tracking.

Source: https://pubmed.ncbi.nlm.nih.gov/39139062/

## E078: Closed letter counters impair recognition

Domain: letter_recognition
Year: 2022
Evidence type: controlled psychophysical experiment
Confidence: high

Claim: Closed apertures impaired identification of letters including a, c, e, r, s, t and f, including parafoveal conditions and crowding.

Design implication: Protect open apertures and counters, especially in the Reading cut and heavier weights.

Source: https://pubmed.ncbi.nlm.nih.gov/35217404/

## E079: Smaller visual angles show greater benefit of letter boldness than larger visual angles

Domain: font_weight
Year: 2019
Evidence type: controlled recognition experiment
Confidence: high

Claim: Thin strokes and extreme boldness reduced recognition, while moderate additional boldness provided more benefit at smaller visual angles.

Design implication: Use Regular as the main reading weight, offer Medium for small visual sizes, and avoid extreme body-text weights.

Source: https://pubmed.ncbi.nlm.nih.gov/31421483/

## E080: The dispute about sans serif versus serif fonts: An interaction between the variables of serif and stroke contrast

Domain: stroke_contrast
Year: 2022
Evidence type: controlled word recognition experiment
Confidence: moderate_high

Claim: Typeface performance depended on the interaction between serif presence and stroke contrast rather than a simple serif versus sans-serif category rule.

Design implication: Keep the approved low-contrast sans structure, rather than relying on a categorical sans-serif claim.

Source: https://pubmed.ncbi.nlm.nih.gov/35661978/

## E081: Fonts of wider letter shapes improve letter recognition in parafovea and periphery

Domain: font_width
Year: 2022
Evidence type: controlled parafoveal and peripheral recognition experiments
Confidence: high

Claim: Wider letter shapes improved target letter recognition and reduced neighbouring-letter misreadings relative to narrower forms in parafoveal and peripheral presentation.

Design implication: Use a subtle width increase in the Reading cut, not a globally expanded display design.

Source: https://pubmed.ncbi.nlm.nih.gov/34704915/

## E082: Applying cognitive and perceptual science to typeface choices

Domain: typography_review
Year: 2026
Evidence type: peer reviewed review
Confidence: high

Claim: Typeface legibility depends on reading scenario. The review highlights simple designs, open counters, non-condensed widths, adequate spacing and avoidance of extreme weights, while rejecting the idea of one universally most legible typeface.

Design implication: Use an optical family with a beautiful Display cut and a Reading cut rather than forcing one geometry into all contexts.

Source: https://pubmed.ncbi.nlm.nih.gov/40817624/


### methodology.md

# Super Dyslexia methodology and decision ledger

Canonical page: https://superdyslexia.com/methodology.html

Last reviewed: 6 October 2026

Decision count: 24

Research informed does not mean scientifically proven best. Comparative human validation remains a separate stage.

## D01: Do not invent a strange dyslexia alphabet

Grade: A
Scope: dyslexia specific

Finding: A 2026 meta-analysis of 15 studies and 688 dyslexic readers found no reliable overall reading advantage for specialist dyslexia fonts as a category.

Decision: Keep familiar, socially neutral letter skeletons. Do not use heavy bottoms, arbitrary rotation, irregular baselines, or novelty merely to signal dyslexia.

Font change: Both Harmonia and High Speed use familiar, socially neutral skeletons. High Speed increases local distinction without heavy bottoms, irregular baselines or decorative disability cues.

Evidence records: E002, E019, E020

## D02: Preserve familiar orthographic forms

Grade: A/B
Scope: dyslexia specific plus design inference

Finding: Orthographic knowledge is an important component of reading, and unusual specialist letterforms have not reliably improved passage reading.

Decision: Increase distinction locally instead of redesigning every glyph.

Font change: Harmonia prioritizes familiar forms. High Speed provides stronger alternative silhouettes as a separate explicit family rather than hidden contextual alternates.

Evidence records: E005, E020, E061

## D03: Protect open apertures and counters

Grade: A
Scope: general visual recognition

Finding: Controlled experiments show more identification errors when apertures are closed.

Decision: Keep c, e, r, s, f and t open, including in heavier weights.

Font change: Both Reading families protect open internal white space. High Speed further emphasizes selected openings only where recognition benefits from clearer silhouette separation.

Evidence records: E078

## D04: Use moderate weight, never extremes

Grade: A
Scope: general visual recognition

Finding: Very thin and extremely bold strokes reduce recognition; moderate additional boldness helps more at smaller visual angles.

Decision: Regular is the main reading weight. Medium is recommended for small screens or small visual sizes.

Font change: Light, Regular, Medium, SemiBold and Bold are included. Regular and Medium are the preferred starting points for sustained body reading; readers can compare weights in the Type Studio.

Evidence records: E079

## D05: Use low stroke contrast

Grade: B
Scope: general typography

Finding: Controlled research shows performance depends on stroke contrast as well as serif presence.

Decision: Keep the approved low-contrast sans structure.

Font change: The family remains visually clean and consistent rather than introducing high-contrast display strokes.

Evidence records: E080, E050

## D06: Widen reading forms subtly, not globally

Grade: B
Scope: general visual recognition

Finding: Wider letters improved parafoveal and peripheral letter recognition, while other work shows width also changes eye-movement strategy.

Decision: Create a separate Reading cut with restrained width increases.

Font change: Each family has a dedicated Reading cut with calmer text proportions and a tighter Display cut for larger typography. The cuts are explicit font files rather than synthetic browser transforms.

Evidence records: E081, E062

## D07: Treat letter spacing and word spacing as one system

Grade: A
Scope: dyslexia specific

Finding: Increasing letter spacing without adequate word spacing can slow reading, and spacing can explain apparent specialist-font advantages.

Decision: Avoid extreme tracking. Preserve strong word boundaries.

Font change: Built in spacing remains moderate. The website Type Studio exposes additional letter spacing for reader controlled comparison rather than forcing exaggerated tracking into every setting.

Evidence records: E016, E017, E018, E047

## D08: Increase word boundary salience cautiously

Grade: B
Scope: dyslexia specific

Finding: A 2024 study reported improved comprehension and fewer migration errors with increased inter-word spacing in students with dyslexia.

Decision: Increase word spacing modestly while retaining normal word texture.

Font change: Word boundaries are kept clearly visible while letter spacing remains restrained. Word spacing is independently adjustable in the Type Studio.

Evidence records: E077, E016

## D09: Crowding matters, but more spacing is not a cure

Grade: A
Scope: dyslexia specific

Finding: Crowding and spacing studies are mixed; some readers benefit, while other studies show no general speed or comprehension benefit from wider spacing.

Decision: Keep spacing moderate and make future personalization possible.

Font change: Reading cut is a strong default, not an extreme dyslexia preset.

Evidence records: E013, E014, E015, E047, E003

## D10: Keep I, l, 1 and i separable

Grade: B
Scope: recognition focused design inference

Finding: Distinctive forms can improve isolated letter identification even when passage benefits are not guaranteed.

Decision: Strengthen high-risk local distinctions without making the alphabet unfamiliar.

Font change: Harmonia keeps calm familiar I, l, 1 and i forms. High Speed increases local silhouette differences while preserving the overall visual language.

Evidence records: E020, E082

## D11: Offer stronger a and o separation through explicit family choice

Grade: B/C
Scope: personalization hypothesis

Finding: A more differentiated a may improve local recognition for some readers, but unfamiliar orthographic forms can impose a familiarity cost.

Decision: Keep Harmonia familiar, and expose stronger a and o silhouette separation in High Speed as an explicit family choice rather than a hidden alternate.

Font change: The stronger recognition treatment is exposed as the High Speed family. It is a visible family choice, not a hidden OpenType switch.

Evidence records: E005, E020, E082

## D12: Differentiate O and 0 through structure, not decoration

Grade: B
Scope: general recognition

Finding: Alphanumeric confusability matters, but slash-zero can collide conceptually with Ø in supported languages.

Decision: Use proportion and construction to separate O and 0. Do not rely on an unusual default slash, dot or decorative mark merely to signal accessibility.

Font change: Harmonia differentiates O and 0 through proportion and construction. High Speed increases the distinction structurally while preserving a professional texture.

Evidence records: E082

## D13: Do not distort b d p q around reversal folklore

Grade: A/B
Scope: dyslexia theory guardrail

Finding: The evidence does not support treating letter reversal as the universal cause of dyslexia, and specialist letter-shape novelty does not reliably improve passage reading.

Decision: Keep directional forms conventional and distinct through normal typographic cues.

Font change: No heavy bottoms, rotation or dramatic asymmetry is forced on b d p q.

Evidence records: E002, E019, E020

## D14: Protect rn and vv from accidental merging

Grade: B
Scope: crowding inference

Finding: Local crowding and letter proximity can impair identification even when more global spacing is not beneficial.

Decision: Use pair-specific protection rather than global tracking.

Font change: Potential rn and m collisions are handled through width, shoulder construction and spacing, with stronger separation in High Speed.

Evidence records: E013, E078

## D15: Use enough effective visual size

Grade: A
Scope: dyslexia specific plus vision science

Finding: Dyslexic children can require larger critical print sizes, and reading speed depends strongly on visual angle near the critical threshold.

Decision: Do not compensate for undersized text with gimmicks.

Font change: Reading guidance prioritizes adequate size; Medium weight is available for smaller text.

Evidence records: E021, E022, E051

## D16: Protect word-initial clarity

Grade: B
Scope: dyslexia specific emerging evidence

Finding: Parafoveal studies suggest altered early letter-position processing, and recent gaze-contingent work reports greater foveal load in dyslexic children.

Decision: Keep word-initial glyphs simple and recognizable.

Font change: No decorative entry strokes or contextual novelty at word beginnings.

Evidence records: E034, E036, E037, E043

## D17: Use an optical family instead of one fixed geometry

Grade: A/B
Scope: evidence synthesis

Finding: Legibility varies by reading scenario; no single typeface geometry is universally optimal.

Decision: Separate aesthetic large-text needs from continuous-reading needs.

Font change: Harmonia and High Speed each provide dedicated Reading and Display cuts across five weights with upright and italic styles.

Evidence records: E082, E056

## D18: Do not use mandatory ligatures

Grade: B
Scope: conservative design inference

Finding: The product prioritizes explicit grapheme identity and local character recognition.

Decision: Avoid automatically merging visible letters in the Reading cut.

Font change: Standard reading CSS disables optional ligatures for body text.

Evidence records: E020, E061

## D19: Punctuation must remain visible

Grade: B
Scope: general reading design

Finding: Reading is not only letter recognition; segmentation and navigation cues should remain visible at text sizes.

Decision: Keep punctuation optically visible and rhythmically consistent at reading sizes without novelty enlargement.

Font change: Punctuation is kept optically visible and rhythmically consistent in both Reading cuts without novelty enlargement.

Evidence records: E082

## D20: Publish only audited diacritic and language coverage

Grade: A technical
Scope: multilingual engineering

Finding: A font cannot claim multilingual support if stacked marks clip or combine incorrectly.

Decision: Treat diacritics as first class reading information and limit language claims to the audited character repertoire.

Font change: The published release contains broad precomposed Latin coverage plus combining mark glyphs. Language support claims are limited to the audited repertoire and do not imply complete script shaping for unsupported writing systems.

Evidence records: E026, E027

## D21: Do not prescribe color as dyslexia treatment

Grade: A
Scope: dyslexia specific

Finding: Systematic-review evidence does not support colored overlays as a general dyslexia treatment.

Decision: Offer theme choice for comfort, not therapeutic claims.

Font change: Website supports light and dark themes; the font itself remains color neutral.

Evidence records: E025, E055

## D22: No Bionic-style prefix bolding

Grade: A
Scope: direct experimental evidence

Finding: Controlled eye-movement research found no reading benefit and costs from bolding the first part of every word.

Decision: Do not build prefix bolding into the font or reading system.

Font change: Bold is reserved for normal semantic hierarchy.

Evidence records: E063

## D23: Treat eye movements as outcomes, not causes

Grade: A
Scope: dyslexia specific

Finding: Dyslexic readers often show different reading eye movements, but basic oculomotor function can be normal and lexical difficulty strongly modulates gaze.

Decision: Use gaze as a validation signal, not as a mythology for glyph design.

Font change: Human validation protocol measures gaze optionally alongside accuracy, comprehension and effort.

Evidence records: E023, E024, E039, E047

## D24: Make personalization granular instead of strange by default

Grade: C
Scope: product hypothesis

Finding: Dyslexia and visual-attention profiles are heterogeneous across readers and languages.

Decision: Keep strong defaults while exposing meaningful controls that correspond to real shipping font files and normal typographic variables.

Font change: Personalization is primarily explicit: Harmonia or High Speed family, Reading or Display cut, weight, size, letter spacing, word spacing and line height. The website makes these controls directly testable.

Evidence records: E004, E035, E056, E058


### coverage.md

# Super Dyslexia language and character coverage

Last reviewed: 6 October 2026

Super Dyslexia V4 contains two complementary families by Karl Lillrud: Harmonia and High Speed. Each family provides Reading and Display cuts across Light, Regular, Medium, SemiBold and Bold, with upright and italic styles.

Every shipping face contains 931 glyphs and 929 Unicode mappings.

## Current original coverage

Latin, extended Latin, Vietnamese, modern Greek, modern Cyrillic, combining marks, punctuation, currencies, arrows, mathematics and related symbols.

## Script boundary

Arabic, Hebrew, Indic, Southeast Asian, CJK, Korean and other writing systems require dedicated native script modules. They are not represented by placeholder or borrowed glyphs merely to claim universal support.

## Family choices

Harmonia prioritizes calm familiarity and premium contemporary texture.

High Speed increases local character distinction while preserving a professional visual system.

Official project: https://superdyslexia.com/
Creator: Karl Lillrud, https://www.karllillrud.com/


### about.md

# About Super Dyslexia

Canonical page: https://superdyslexia.com/about.html

Super Dyslexia by Karl Lillrud is a contemporary typeface project with a public evidence trail. The project exposes 82 evidence records and 24 design decisions.

It is not a medical device, diagnostic tool or treatment, and it does not claim universal reading benefit.

Creator information and contact: https://www.karllillrud.com/


### accessibility.md

# Designed toward WCAG 2.2 Level AA, without pretending validation is automatic.

This build includes accessibility focused code and content patterns, but formal conformance can only be claimed after the deployed production site is tested as a whole.

## Implemented in this build

* Semantic landmarks, one primary heading per static page and a keyboard skip link.
* Visible keyboard focus with high contrast focus indicators.
* Improved minimum interactive target sizes.
* Text alternatives and accessible labels where controls would otherwise be ambiguous.
* Reduced motion support for reveal effects and the reading friction experience.
* Screen reader safe static text in the animated reading experience.
* Color contrast improvements for small orange text and primary calls to action.
* Native dialog elements for modal workflows and live status regions for form feedback.
## Known validation requirement

Before a public conformance statement is made, the live site should be tested with automated tools plus manual keyboard navigation, zoom, reflow, screen reader testing and representative user testing. Any connected checkout, authentication or third party experience must be included in that review.

## Accessibility feedback

If a user encounters an accessibility barrier, the public contact route is Karl Lillrud’s contact page . Include the page, task, browser and assistive technology when possible.


Canonical HTML: https://superdyslexia.com/accessibility.html


### ai.md

# Human readable first, agent readable by design.

Super Dyslexia exposes its important claims, research and routes in ordinary semantic HTML, then adds machine friendly representations for retrieval systems and browser agents.

## Authoritative resources

/research.html contains the complete crawlable evidence registry. /methodology.html contains all 24 design decisions. /llms.txt provides a concise optional agent map. /ai/site-profile.json provides a structured convenience summary.

## Search and crawler policy

The default robots policy permits normal search crawling and explicitly permits major AI search and model crawlers while excluding internal deployment folders. Raw font binaries are not intended as indexed search results. Operators can change training crawler policy independently without blocking search crawlers.

## Browser agents

The personal access and educator application forms include progressive WebMCP declarative annotations. Browsers that do not support the proposed API simply ignore those attributes and the normal forms continue to work.

## No fake GEO signal

The package does not add invented ratings, fabricated citations, hidden keyword text, mass generated doorway pages or unsupported medical claims. The strategy is to make original evidence, authorship, method, source links and task routes easy to discover and verify.

## Machine interfaces


Canonical HTML: https://superdyslexia.com/ai.html


## Current machine readable release surfaces

- https://superdyslexia.com/data/font_coverage.json
- https://superdyslexia.com/ai/content-index.json
- https://superdyslexia.com/ai/actions.json
- https://superdyslexia.com/coverage.html


### license.md

# Use the font under the route that matches your purpose.

The website separates personal, professional and educational use so a user or AI agent can identify the correct route before requesting access.

## Free personal use

Personal, private and noncommercial use after email verification. Redistribution and professional use are excluded.

## Professional license

Commercial and organizational use under version 2026.10. The installable files are delivered after verified payment.

## Educator application

Schools and educators can apply for a separate educator license based on intended educational use.

## Research data

The evidence registry and methodology are public for inspection. Publication does not grant rights to redistribute font files.


Canonical HTML: https://superdyslexia.com/license.html