Acquired Dyslexia

Abstract

Acquired dyslexia is the loss of a previously mastered ability to read, caused by brain injury such as stroke or neurodegeneration, and distinct from the developmental dyslexia present from childhood. MeSH classifies it as a disorder of reading. Within the dual-route model of reading aloud, its forms divide into central dyslexias, which damage the mapping from print to sound or meaning, and peripheral dyslexias, which disrupt the visual analysis of the letter string before word recognition. The central forms comprise surface, phonological, and deep dyslexia; the principal peripheral form is pure alexia, or letter-by-letter reading. Each pattern is a dissociation that tests theories of the normal reading system, and together they show that skilled reading is not one process but several separable components, each independently vulnerable to damage.

Keywords: acquired dyslexia, alexia, dual-route model

What Acquired Dyslexia Is

Acquired dyslexia — equivalently alexia — is the partial or complete loss of a previously competent ability to read, produced by injury to the brain regions that support reading: most often a stroke, but also tumor, head injury, or a neurodegenerative disease. According to MeSH it is classified as a disorder, and specifically as a disorder of reading: it is filed under dyslexia and, at a higher level, under the neurocognitive disorders. This places it correctly as an acquired impairment of a skilled cognitive function, and it separates it sharply from developmental dyslexia, the difficulty in learning to read that is present from childhood in the absence of any acquired lesion. The two share a name and a symptom — impaired reading — but not a cause: one is the breakdown of a system that was built and then damaged, the other a system that never developed typically.

The scientific interest of acquired dyslexia lies in the fact that brain damage does not abolish reading uniformly. Instead it produces distinct patterns of spared and impaired reading, and each pattern is a dissociation that reveals a component of the normal reading system by knocking it out selectively. John Marshall and Freda Newcombe’s (1973) analysis of the paralexias — the systematic misreadings their patients produced — founded the modern approach: rather than treating alexia as a single deficit of reduced reading ability, they showed that the kinds of errors a patient makes sort them into separable syndromes, each implicating a different reading mechanism.

Two organizing distinctions frame the whole field. The first separates the central dyslexias, in which the damage falls on the core machinery that converts a recognized letter string into sound or meaning, from the peripheral dyslexias, in which the damage falls earlier, on the visual analysis that delivers a letter string to that machinery in the first place. The second, developed below, is the theoretical model — the dual-route account of reading aloud — within which these syndromes are explained as selective damage to one or another processing route.

Types of Acquired Dyslexia

In the Medical Subject Headings hierarchy, Dyslexia, Acquired sits beneath the broader descriptor Dyslexia and carries one narrower descriptor of its own:

Two caveats are needed to read this table correctly. First, MeSH is an indexing classification, not a cognitive theory: it records the single subtype (pure alexia) that has been minted as its own descriptor for cataloguing the literature, and it is silent on the several other syndromes — surface, phonological, and deep dyslexia — that cognitive neuropsychology treats as full members of the family. The clinically and theoretically recognized types of acquired dyslexia are therefore broader than this single MeSH child, and the sections below describe them in full. Second, the central/peripheral split that structures those sections is orthogonal to the MeSH tree: pure alexia is a peripheral dyslexia, whereas the central dyslexias impair a different, later stage of reading, so the two groups are not points on one severity scale but damage to different parts of the system.

The Dual-Route Framework

The syndromes of acquired dyslexia are explained within a model of how the intact brain reads a word aloud, and the dominant such model is the dual-route cascaded (DRC) model of Max Coltheart and colleagues (2001). It holds that a skilled reader converts print to speech by two routes working in parallel. The lexical route recognizes a word as a familiar whole and retrieves its stored pronunciation (and, via meaning, its sense) directly from a mental lexicon; it is the only route that can pronounce an irregular word such as pint or yacht, whose spelling disobeys the usual rules. The sublexical route applies grapheme-to-phoneme correspondence rules, sounding a letter string out piece by piece; it is the only route that can pronounce a nonword such as blicket, which has no lexical entry. For a regular word both routes converge on the same pronunciation; for irregular words and nonwords they diverge, and it is exactly at that divergence that the acquired dyslexias reveal themselves.

An influential alternative is the connectionist “triangle” model of David Plaut, James McClelland, Mark Seidenberg, and Karalyn Patterson (1996). It replaces the two discrete routes with a single learning network that maps among three kinds of representation — orthography (spelling), phonology (sound), and semantics (meaning) — and it derives the regular/irregular and word/nonword contrasts from the statistics of the mapping rather than from separate rule and lexicon modules. On this account there is a division of labour between a direct spelling-to-sound pathway and a semantically mediated one, and damage shifts the balance between them. The two frameworks make different theoretical commitments, but both predict the same central fact the syndromes below illustrate: reading aloud can fail in dissociable ways because it depends on more than one underlying pathway.

The Central Dyslexias

The central dyslexias impair the conversion of a correctly perceived letter string into sound or meaning, and three classic patterns divide the space according to which route is damaged.

Surface dyslexia is the pattern in which the lexical route is compromised and reading leans on the intact sublexical route. Regular words and nonwords are read well, but irregular words are regularized — pint is read to rhyme with mint, sew with few — because a rule-based sounding-out cannot honour their exceptional spellings. Anna Woollams, Matthew Lambon Ralph, David Plaut, and Karalyn Patterson’s (2007) “SD-squared” account tied surface dyslexia tightly to semantic dementia: as the degeneration erodes word meaning, the semantically supported lexical readings decay first and exactly for the low-frequency irregular words the model predicts, making surface dyslexia the reading signature of a semantic breakdown.

Phonological dyslexia is the mirror pattern: the sublexical route is compromised and reading leans on the intact lexical route. Familiar real words — regular or irregular — are read well, but nonwords are read very poorly, because sounding out an unfamiliar string is precisely what the damaged route did. Steven Rapcsak and colleagues (2009) showed that phonological dyslexia, and its writing counterpart phonological dysgraphia, reflect damage to a broader phonological system supporting the wider language network, not a reading-specific module, locating the deficit in the machinery that manipulates speech sounds.

Deep dyslexia is the most severe central pattern and the one that founded the field. Like phonological dyslexia it abolishes nonword reading, but it adds a striking symptom: semantic errors, or semantic paralexias, in which a printed word is read aloud as a word related in meaning — dog read as “cat,” forest as “trees.” It also shows strong effects of imageability and part of speech (concrete nouns easiest, function words hardest). Marshall and Newcombe (1973) first described it, and Branch Coslett and Eleanor Saffran’s work on preserved reading argued that some deep-dyslexic reading is mediated by a right-hemisphere system with a semantic but not a phonological competence. Jill Crisp and Matthew Lambon Ralph (2006) argued that deep and phonological dyslexia are not two disorders but two regions of a single continuum, graded by the joint severity of phonological and semantic impairment — deep dyslexia being the end where both are damaged enough for semantic errors to emerge.

The Peripheral Dyslexias

The peripheral dyslexias strike before the central machinery, at the visual analysis that must group a string of letters into a perceptual word. Their principal form is pure alexia — also called alexia without agraphia, letter-by-letter reading, or, in Elizabeth Warrington and Tim Shallice’s (1980) term, word-form dyslexia. Its defining sign is that reading, though possible, becomes slow and serial: the patient identifies a word by naming or sounding out its letters one at a time, so reading time rises steeply with the number of letters — the word-length effect that the Worked Example below quantifies. Writing is spared, producing the classic paradox of a patient who can write a sentence and then cannot read back what they have written.

The lesion is a disconnection of visual information from the reading system, centred on the left ventral occipitotemporal cortex — the region Stanislas Dehaene and Laurent Cohen (2011) describe as a Visual Word Form Area specialized for recognizing letter strings as visual wholes. Whether that region is truly a reading-specific module is debated: Cathy Price and Joseph Devlin’s (2011) interactive account argues that it is not a store of word forms but a general-purpose relay whose reading role emerges from its connections to language regions. Either way, damaging it removes the fast, parallel recognition of the whole word and forces the slow letter-by-letter strategy. The loss is not always total: Branch Coslett and Eleanor Saffran’s (1989) study of preserved reading in pure alexia found that some patients retain an implicit ability to categorize or match words they cannot explicitly identify, evidence that visual information still reaches the reading system by a residual path.

Pure alexia is also less “pure” than its name suggests. Marlene Behrmann, Jonathan Nelson, and Eve Sekuler (1998) showed that letter-by-letter reading is modulated by visual variables such as the complexity and confusability of the letters, arguing that the deficit is partly perceptual rather than a clean loss of an abstract word-form store. Other peripheral dyslexias localize the disruption differently: neglect dyslexia produces errors confined to one side of the word (usually the left), and attentional dyslexia allows single words to be read while letters or words crowd and intrude on one another in an array — each a different failure of the spatial and attentional delivery of the letter string to the recognition system.

Worked Example

The peripheral signature of pure alexia can be made quantitative. Because the patient reads letter by letter, single-word reading time grows linearly with word length: RT = a + b × N, where N is the number of letters, a is a fixed baseline, and b is the per-letter slope. In a skilled reader the whole word is recognized in parallel, so the slope b is essentially flat (a few milliseconds per letter); in pure alexia it is grossly elevated. A slope above roughly 100 ms per letter is the diagnostic marker of letter-by-letter reading.

Consider a control reader and a patient with pure alexia, each timed reading words of 3, 6, and 9 letters.

The control reader’s reading time barely moves — 545 ms for a three-letter word, 635 ms for a nine-letter word, a rise of only 90 ms across six extra letters — because the whole word is grasped at once and length is nearly irrelevant. The patient’s time climbs from 1650 ms to 3750 ms across the same span, an extra 2100 ms, because each of the six additional letters costs another serial identification step of about 350 ms. The slope, not the overall speed, is what diagnoses the syndrome: a reader can be globally slow for many reasons, but a per-letter cost of hundreds of milliseconds specifically identifies the serial, letter-by-letter reading of pure alexia. The interactive demonstration above recomputes the two lines as the slope and baseline are varied, showing how the word-length function separates whole-word from letter-by-letter reading.

Discussion

Acquired dyslexia is one of the founding case studies of cognitive neuropsychology, and its lesson is methodological as much as clinical. Because brain damage fractures reading along its natural seams rather than degrading it uniformly, the pattern of a patient’s errors is evidence about the architecture of the normal system — the logic Marshall and Newcombe (1973) introduced when they read the paralexias as a map of reading’s components. Surface and phonological dyslexia form a double dissociation that motivated the two routes of the dual-route model (Coltheart et al., 2001); deep dyslexia’s semantic errors implicate the meaning system in reading aloud; and pure alexia isolates the visual word-recognition stage that precedes them all. The disorder is, in effect, a natural experiment run repeatedly on the reading system, and the theories of Coltheart et al. (2001) and Plaut et al. (1996) are in large part attempts to account for exactly which combinations of sparing and loss occur.

Two cautions temper the neatness of the taxonomy. First, the categories are idealizations: real patients often show mixed profiles, and Crisp and Lambon Ralph (2006) argued that deep and phonological dyslexia are better seen as regions of a graded continuum than as discrete types — a warning that the syndrome names label tendencies, not natural kinds with sharp borders. Second, even a “peripheral” visual syndrome such as pure alexia is not perceptually pure (Behrmann et al., 1998), and even a “central” syndrome recruits systems well beyond reading (Rapcsak et al., 2009); the central/peripheral split is a useful first cut, not a clean anatomical partition.

The construct binds outward across the field. It is a disorder of a written-language skill that depends on the visual perception of letter strings and, through the sublexical route, on the same sound system studied in speech perception; its deep-dyslexic semantic errors implicate the semantic memory that stores word meaning; and pure alexia is a reading-specific member of the broader family of visual agnosia. Few disorders show so cleanly that a single everyday skill is assembled from several separately breakable parts.

Current Directions

One active line reframes the acquired dyslexias under the primary systems hypothesis, which holds that reading has no dedicated modules of its own but is built on general language systems — phonology, semantics, and vision — so that a reading deficit is always the reading-visible face of a broader impairment. Anna Woollams, Ajay Halai, and Matthew Lambon Ralph’s (2018) lesion-mapping study tested this directly, relating patients’ reading profiles to the integrity of the phonological and semantic networks rather than to reading-specific damage, and connecting the surface and phonological patterns to the same systems whose computational bases Lambon Ralph and colleagues (2017) set out for semantic cognition. On this view the classical syndromes are not disorders of a reading box but predictable shadows cast by damage to the language systems reading borrows.

A second front is rehabilitation. Zoe Woodhead and colleagues’ (2018) randomized trial of iReadMore — a tablet-based word-reading therapy tested with and without transcranial brain stimulation in central alexia — showed measurable, trained-item gains, moving the field from describing the syndromes toward treating them. Alongside the group studies, detailed single-case work continues to refine the phenomenology: Kristian Hansen and Randi Starrfelt’s (2019) combined first-person and neuropsychological account of pure alexia is a reminder that the lived experience of letter-by-letter reading still has something to teach a mature theory. Together these lines are pushing acquired dyslexia from a taxonomy of dissociations toward an integrated account that spans the underlying language systems, their neural bases, and their treatment.

Common Misconceptions

“Acquired dyslexia is the same thing as childhood dyslexia.”

They share a symptom but not a cause. Acquired dyslexia is the loss of a reading ability that was previously mastered, caused by brain injury; developmental dyslexia is a difficulty in ever acquiring fluent reading, present from childhood with no such lesion (Marshall & Newcombe, 1973).

“Losing the ability to read means losing language.”

In pure alexia writing and speech are preserved; the patient can write a sentence yet be unable to read it back. The impairment is specific to recognizing words visually, not to language as a whole (Warrington & Shallice, 1980).

“There is one disorder called alexia.”

There are several dissociable syndromes — surface, phonological, and deep dyslexia centrally, and pure alexia and neglect and attentional dyslexia peripherally — each defined by a different pattern of spared and impaired reading (Coltheart et al., 2001).

“A semantic error in reading means the patient cannot see the word.”

In deep dyslexia a word is read aloud as one close in meaning — dog as “cat” — which shows the visual word was processed far enough to reach its meaning; the failure is in retrieving its exact pronunciation, not in perceiving it (Crisp & Lambon Ralph, 2006).

Glossary

Acquired dyslexia.

The loss of a previously competent ability to read, caused by brain injury; also called alexia, and distinct from developmental dyslexia.

Attentional dyslexia.

A peripheral dyslexia in which single letters or words are read in isolation but crowd and intrude on one another when presented among others.

Deep dyslexia.

A central dyslexia marked by semantic errors in reading aloud, an inability to read nonwords, and strong imageability effects; the severe end of the phonological-deep continuum.

Dual-route model.

A theory of reading aloud in which a lexical route retrieves known words as wholes and a sublexical route sounds letter strings out by rule.

Lexical route.

The reading pathway that recognizes a familiar word as a whole and retrieves its stored pronunciation and meaning; the only route that can read irregular words.

Neglect dyslexia.

A peripheral dyslexia in which reading errors are confined to one side of the word, usually the left, reflecting a spatial attention deficit.

Paralexia.

A reading error in which one word is substituted for another; its type (visual, semantic, or morphological) helps classify the underlying dyslexia.

Phonological dyslexia.

A central dyslexia in which familiar real words are read well but unfamiliar nonwords cannot be sounded out, reflecting damage to the sublexical route.

Pure alexia.

A peripheral dyslexia, also called alexia without agraphia or letter-by-letter reading, in which words are read slowly letter by letter while writing is spared.

Regularization.

Reading an irregular word as if it obeyed the usual spelling-sound rules (pint to rhyme with mint); the hallmark error of surface dyslexia.

Sublexical route.

The reading pathway that converts graphemes to phonemes by rule, sounding a letter string out piece by piece; the only route that can read nonwords.

Surface dyslexia.

A central dyslexia in which regular words and nonwords are read well but irregular words are regularized, reflecting damage to the lexical route.

Visual Word Form Area.

A region of left ventral occipitotemporal cortex whose damage produces pure alexia; proposed as a hub for recognizing letter strings as visual wholes.

Word-length effect.

The steep, roughly linear rise in single-word reading time with the number of letters that is the behavioural signature of letter-by-letter reading.

Key Researchers

Marlene Behrmann (University of Pittsburgh). Showed that letter-by-letter reading in pure alexia is modulated by visual and lexical variables, challenging a purely perceptual account of the syndrome. ORCID - Wikipedia - Wikidata - Google Scholar - Faculty

Max Coltheart (Macquarie University). Architect of the dual-route cascaded model of reading aloud, the framework within which the acquired dyslexias are classified. ORCID - Wikipedia - Wikidata - Google Scholar - Faculty

John C. Marshall (1939–2007). With Freda Newcombe, wrote the 1973 paper that founded the modern deep and surface taxonomy of acquired dyslexia by reading the paralexias as evidence about the reading system. Wikipedia - Wikidata - Google Scholar

Karalyn Patterson (MRC Cognition and Brain Sciences Unit, University of Cambridge). Co-developer of connectionist models of reading and of the surface-dyslexia account linking it to semantic breakdown. Wikipedia - Wikidata - Faculty

Matthew A. Lambon Ralph (MRC Cognition and Brain Sciences Unit, University of Cambridge). Developed the primary-systems account of acquired dyslexia and the phonological-deep dyslexia continuum. ORCID - Wikidata - Google Scholar

Eleanor M. Saffran (1938–2002). Cognitive neuropsychologist whose case studies of deep dyslexia and pure alexia, including the right-hemisphere reading hypothesis, shaped the field. Wikipedia - Wikidata

Frequently Asked Questions

What is acquired dyslexia?
It is the loss of a previously mastered ability to read, caused by damage to the brain from a stroke, injury, tumor, or neurodegenerative disease. Also called alexia, it is a disorder of reading and is separate from the developmental dyslexia present from childhood (Marshall & Newcombe, 1973).

How is it different from developmental dyslexia?
Acquired dyslexia is the breakdown of a reading system that was built and then damaged; developmental dyslexia is a difficulty in learning to read fluently that is present from childhood with no acquired lesion. They share the symptom of impaired reading but not the cause (Marshall & Newcombe, 1973).

What are the main types of acquired dyslexia?
Cognitive neuropsychology recognizes central dyslexias — surface, phonological, and deep dyslexia — which damage the mapping from print to sound or meaning, and peripheral dyslexias, chiefly pure alexia, which disrupt the visual analysis of the letter string (Coltheart et al., 2001).

What is surface dyslexia?
It is a central dyslexia in which regular words and nonwords are read well but irregular words are regularized — pint read to rhyme with mint — because the lexical route is damaged and reading relies on rule-based sounding out. It is the reading signature of semantic dementia (Woollams et al., 2007).

What is deep dyslexia?
It is a severe central dyslexia in which the reader cannot read nonwords and makes semantic errors, reading a printed word aloud as one close in meaning, such as forest for trees. It shows strong effects of how imageable a word is (Crisp & Lambon Ralph, 2006).

What is pure alexia?
Pure alexia, or letter-by-letter reading, is a peripheral dyslexia in which reading becomes slow and serial while writing is preserved. Reading time rises steeply with word length, and the lesion involves the left ventral occipitotemporal cortex (Dehaene & Cohen, 2011).

Can someone with pure alexia still write?
Yes. Pure alexia is also called alexia without agraphia precisely because writing is spared: a patient may write a sentence fluently and then be unable to read it back, because the deficit lies in recognizing words visually, not in producing language (Warrington & Shallice, 1980).

Can acquired dyslexia be treated?
Rehabilitation is an active field. Tablet-based word-reading therapies such as iReadMore, tested with and without brain stimulation, produce measurable gains on trained words in central alexia, moving the field from describing the syndromes toward treating them (Woodhead et al., 2018).

Support Organizations

Organizations that provide information, assessment guidance, and support for acquired reading disorders and the strokes and dementias that cause them.

References