Learning Disabilities

Abstract

Learning disabilities are neurodevelopmental disorders in which a specific academic skill — reading, mathematics, or written expression — develops far below the level expected from a person's age, schooling, and general ability, despite adequate instruction and intact sensory function. The disorder is *specific* and *unexpected*: it does not follow from low intelligence, poor teaching, or lack of effort, but reflects an inefficiency in the particular brain systems that support learning that skill. This article traces the concept from its origin in the mid-twentieth century, through the collapse of the IQ–achievement discrepancy criterion and its replacement by response to intervention, to the multiple-deficit model that now frames causation. It covers the major types — dyslexia, dyscalculia, and disorders of written expression — their epidemiology and comorbidity, the genetic and neural bases, and evidence-based assessment and intervention.

Keywords: dyslexia, dyscalculia, neurodevelopment

A learning disability is one of the most common reasons a child of clearly normal intelligence fails at school, and one of the most misunderstood. The child can reason, converse, and solve problems as well as any classmate, yet cannot learn to read, or cannot learn arithmetic, at anything like the expected rate. That gap between ability and achievement — the sense that the difficulty is *unexpected* — is the intuition the field was built to capture (Kirk coined the term in 1963; see Grigorenko et al., 2020).

The account below moves from what a learning disability is, through the MeSH subtypes and the core academic features, to the epidemiology and the striking overlap between reading and mathematics disability, the shift in how the disorders are identified, the genetic and neural causes, and the assessment and instruction that follow. The recurring theme is that a learning disability is a dimensional, multiply-determined inefficiency in a specific skill network, not a categorical defect and not a mark of low ability.

Figure 1

The Defining Profile: Specific and Unexpected Underachievement

What Learning Disabilities Are

A learning disability is a disorder in one or more of the basic psychological processes involved in understanding or using language, spoken or written, that shows itself as an inability to learn a specific academic skill efficiently. The defining feature is *specificity*: the difficulty is confined to a particular domain — most often reading — while other abilities develop normally. The second defining feature is that the difficulty is *unexpected*, meaning it cannot be explained by intellectual disability, uncorrected vision or hearing problems, inadequate schooling, or a lack of opportunity to learn (Lyon, Shaywitz, & Shaywitz, 2003).

This is what separates a learning disability from generally weak achievement. A child who reasons poorly across the board and reads poorly too does not have a *specific* learning disability; the reading difficulty is in line with everything else. A child who reasons normally, speaks fluently, and does well in mathematics, yet cannot decode written words, does. The mismatch — strong where the skill is not required, weak precisely where it is — is the signature of the disorder (Fletcher, 2009). The empirical case for this distinction was first made by Rutter and Yule's Isle of Wight epidemiological studies, which showed that children with *specific reading retardation* formed a group distinguishable from those with general intellectual backwardness (Rutter & Yule, 1975).

The concept was named in 1963, when Samuel Kirk proposed “learning disabilities” as an umbrella term for children whose academic failure could not be attributed to intellectual disability or emotional disturbance, giving a scattered set of clinical observations a single organising label and, in time, a place in educational law (Grigorenko et al., 2020). Modern classifications refine rather than replace that insight. The DSM-5 folds the specific academic disorders into a single diagnosis, Specific Learning Disorder, with three coded specifiers — with impairment in reading, in mathematics, and in written expression — while MeSH and ICD retain the older family of separate learning-disorder terms.

Types of Learning Disabilities

MeSH places Learning Disabilities within the broader kind of communication and neurodevelopmental disorders and files two direct subtypes beneath it. It is worth stressing that a *classification* is an indexing device, not a theory of the disorder: the MeSH tree records how the literature is catalogued, and its two children sit alongside the clinically central types — dyslexia and dyscalculia — which the tree indexes elsewhere. The two axes are orthogonal, and the everyday clinical types below cross-cut the indexing hierarchy.

In clinical and research use the disorders are named by the skill affected. Dyslexia, or specific reading disability, is by far the most studied: a difficulty in accurate and fluent word recognition rooted in weak phonological processing (Peterson & Pennington, 2012). Developmental dyslexia is the childhood-onset form, distinguished from reading loss acquired after brain injury. Dyscalculia is the mathematical analogue — a difficulty in acquiring arithmetic and number sense out of proportion to general ability (Butterworth, Varma, & Laurillard, 2011). A disorder of written expression, sometimes called dysgraphia, affects spelling and the composition of text. Because these types overlap heavily, a single child may carry more than one.

Core Features

Each type of learning disability has a characteristic core deficit that sits beneath the visible academic failure. In dyslexia the core is *phonological*: children struggle to represent and manipulate the sound structure of words, which makes the mapping from letters to sounds — the engine of reading — slow and error-prone. This phonological deficit is measurable before reading is taught and predicts who will struggle, and it persists into adulthood even in people who have learned to compensate (Vellutino, Fletcher, Snowling, & Scanlon, 2004). Fluent word recognition, not comprehension, is the bottleneck; once decoding is laborious, everything downstream — reading speed, comprehension, vocabulary growth — is dragged down with it. Phonology is not the whole story, however: the *double-deficit hypothesis* holds that slow rapid automatised naming is a second, partly independent source of reading difficulty, so that children with both a phonological deficit and a naming-speed deficit are the most severely and persistently impaired (Wolf & Bowers, 1999). This is why the worked example below weights naming speed as a distinct contributor.

In dyscalculia the core deficit is in the sense of number itself: an imprecise representation of quantity that makes even simple magnitude comparisons and the learning of arithmetic facts effortful (Landerl, Bevan, & Butterworth, 2004). The child may count on fingers long after peers have memorised sums, and struggle to sense that eight is larger than six without counting. The deficit is domain-specific: reasoning, language, and reading can be entirely normal (Geary, 2011). This developmental form must be distinguished from acalculia, the *loss* of calculation ability after brain injury in a previously competent adult; dyscalculia is present from the start of formal schooling rather than acquired.

A further universal feature is a secondary cost. Because reading and mathematics are gateways to almost everything else in school, an untreated learning disability compounds over time: the child who reads slowly reads less, learns fewer words, and falls further behind — a widening gap Keith Stanovich named the *Matthew effect*, after the biblical passage in which the rich grow richer (Stanovich, 1986). This is why early identification matters so much: the disability itself is stable, but its consequences accumulate.

Epidemiology and Comorbidity

Learning disabilities are common. Depending on the exact cut-off used, reading disability affects roughly 5–10% of school-age children and mathematics disability a similar 3–7%; a large epidemiological study using standardised criteria put the prevalence of specific learning disorder in the general population near 6% (Moll, Kunze, Neuhoff, Bruder, & Schulte-Korne, 2014). Because the underlying skills are continuously distributed, the exact figure depends on where the diagnostic line is drawn — a point returned to below.

The most important epidemiological fact is *comorbidity*. Reading disability and mathematics disability co-occur far more often than would be expected if they were independent: a large twin study found that children with one disorder were several times more likely to have the other, and that the overlap is driven partly by shared cognitive risk factors such as working memory and processing speed (Willcutt et al., 2013). Learning disabilities also overlap heavily with attention-deficit/hyperactivity disorder and with developmental language disorder. This dense web of co-occurrence is one of the strongest arguments against single-cause models and one of the pillars of the multiple-deficit account.

Historically a sex difference was reported, with boys identified far more often than girls, but much of that gap reflects referral bias — boys' behaviour draws teachers' attention — rather than a true difference in prevalence; population studies that test every child find a much smaller male excess (Moll et al., 2014).

Identification: From Discrepancy to Response to Intervention

For decades a learning disability was identified by an *IQ–achievement discrepancy*: a child qualified if their achievement in a skill fell far enough below the level predicted by their IQ, typically by one to two standard deviations. The logic seemed to capture “unexpected” underachievement directly. In practice the criterion failed on several grounds. It required a child to fall far behind before the gap was large enough to count — a *wait-to-fail* model that delayed help until the most treatable years had passed. It was psychometrically unstable near the cut-off. And, most damaging, research showed that poor readers with and without an IQ discrepancy have the same phonological deficit, respond to the same instruction, and differ little in any way that matters — so the discrepancy was sorting children on a distinction without a difference (Fletcher, 2009).

The alternative, written into US law in 2004, is *response to intervention* (RtI): rather than waiting for a gap, schools give all children evidence-based instruction, monitor progress, and identify as learning-disabled those who fail to respond adequately even to intensified, high-quality teaching (Fuchs & Fuchs, 2006). RtI reframes the disability as *unexpected non-response* rather than unexpected underachievement, catches children earlier, and builds intervention into the identification process itself. Its own limitations — variability in how “adequate response” is defined, and the fact that it identifies *that* a child struggles without saying *why* — mean that modern practice often combines low-achievement cut-offs, response monitoring, and cognitive assessment rather than relying on any one.

Worked Example

The multiple-deficit model can be made concrete with a simple, deliberately illustrative risk calculation. Suppose reading disability emerges when a child's *cognitive risk load* — the accumulated severity of deficits across several contributing processes — crosses a threshold. Let three processes each contribute, on a 0–1 severity scale: phonological awareness (weight 0.45), rapid automatised naming (weight 0.35), and oral language / verbal working memory (weight 0.20). Define the composite load

D = 0.45·p + 0.35·r + 0.20·l

and turn it into the probability of reading disability with a logistic function centred at the threshold D = 0.5:

P = 1 / (1 + e^{−8(D − 0.5)})

Consider Child A, who has multiple mild-to-moderate deficits: a strong phonological weakness (p = 0.8), a moderate naming-speed weakness (r = 0.6), and a mild oral-language weakness (l = 0.4). The load is D = 0.45(0.8) + 0.35(0.6) + 0.20(0.4) = 0.36 + 0.21 + 0.08 = 0.65, giving P = 1 / (1 + e^−1.2) ≈ 0.77 — a 77% probability of reading disability.

Now consider Child B, who has an *equally severe* phonological deficit but nothing else: p = 0.8, r = 0.1, l = 0.1. The load is D = 0.36 + 0.035 + 0.02 = 0.415, giving P = 1 / (1 + e^0.68) ≈ 0.34 — only a 34% probability. The single risk factor, however strong, usually does not cross the threshold on its own. This is the central lesson of the multiple-deficit model: disorders arise from the *accumulation* of probabilistic risks, which is why no single cause has ever been found and why the same deficit produces a disorder in one child but not another (McGrath, Peterson, & Pennington, 2020).

Causes

Learning disabilities are substantially heritable. Twin and family studies put the heritability of reading disability at roughly 40–60%, with mathematics disability in a similar range, and the two share a large part of their genetic variance — a molecular echo of their behavioural comorbidity (Willcutt et al., 2013). No single gene causes dyslexia or dyscalculia. Instead many common variants of small effect, together with rarer variants, shift a child's position on the continuous distribution of reading or mathematical skill; candidate genes implicated in early studies act on neuronal migration and the development of cortical circuits, though replication has been mixed and the field has moved toward genome-wide approaches (Peterson & Pennington, 2015).

Crucially, the same probabilistic logic runs through the environment. Familial risk — having a parent or sibling with dyslexia — raises the odds severalfold, and a longitudinal meta-analysis shows that children at family risk have subtle oral-language weaknesses long before reading begins, marking the pathway from genetic liability through early language to later reading failure (Snowling & Melby-Lervag, 2016). This is the multiple-deficit model in its mature form: genetic and environmental risks are each individually neither necessary nor sufficient, and a disorder appears only where enough of them coincide (McGrath et al., 2020). The older single-deficit theories — that dyslexia is *the* phonological deficit, full stop — have given way to this probabilistic, multifactorial picture.

Neuroanatomy

Reading is not an evolved ability with a dedicated brain region; it is a cultural invention that recruits and reshapes circuits built for spoken language and vision. In skilled readers a left-hemisphere network does the work: an anterior region around the inferior frontal gyrus, a dorsal temporoparietal region that supports the effortful mapping of letters to sounds, and a ventral occipitotemporal region — the “visual word form area” — that comes to recognise familiar words rapidly and automatically. Functional imaging in dyslexia consistently shows *underactivation* of the two posterior left-hemisphere regions during reading, together with compensatory reliance on anterior and right-hemisphere areas (Shaywitz & Shaywitz, 2005).

The dyscalculic brain shows an analogous story in a different network: the intraparietal sulcus, which houses the brain's representation of numerical magnitude, is structurally and functionally atypical in children with mathematics disability (Butterworth et al., 2011). In both disorders the neural signature is one of *inefficiency* in a specific, skill-relevant network rather than gross damage — consistent with a developmental disorder of circuit tuning rather than a lesion, and consistent with the finding that targeted instruction can partly normalise the underactive circuits (Peterson & Pennington, 2012).

Assessment

A diagnostic assessment for a learning disability has three tasks. First, it must document *specific* underachievement: standardised, norm-referenced tests of the academic skill in question — word reading and fluency, spelling, calculation, written expression — establish that performance is well below age expectations, conventionally at or below the 7th–16th percentile. Second, it must establish that the difficulty is *unexpected* and *persistent*: it has continued despite adequate instruction and intervention, and is not better explained by intellectual disability, sensory impairment, or lack of schooling — the DSM-5 exclusionary criteria (Grigorenko et al., 2020).

Third, a thorough assessment probes the *cognitive processes* beneath the academic deficit — phonological awareness and rapid naming for reading, number sense for mathematics, working memory and processing speed across both — because these both explain the pattern and point to intervention targets (Hulme & Snowling, 2016). The move away from the discrepancy formula means that a low IQ no longer disqualifies a child from a learning-disability diagnosis, and a high one is no longer required to earn it; what matters is specific, unexpected, persistent underachievement in the skill itself.

Intervention

The best-evidenced interventions are direct, explicit, and skill-specific. For dyslexia, structured literacy — systematic, explicit teaching of the letter–sound correspondences, delivered intensively and early — produces reliable gains in decoding, and randomised trials show that well-designed oral-language and reading programmes causally improve literacy rather than merely correlating with it (Hulme & Snowling, 2016). The earlier such instruction begins, the better: intervening in the first years of school, before the Matthew effect has widened the gap, is far more effective than remediation later (Stanovich, 1986).

For dyscalculia, targeted instruction that builds number sense — making magnitude relations concrete and rehearsing arithmetic facts to fluency — is the analogue, though the evidence base is thinner than for reading (Butterworth et al., 2011). Across both, response to intervention is not only an identification tool but the treatment itself: tiered, increasingly intensive instruction with progress monitoring is the framework in which most children are now helped (Fuchs & Fuchs, 2006). Accommodations — extra time, audiobooks, calculators — support access to the wider curriculum but do not remediate the underlying skill, and are a complement to instruction, not a substitute for it.

Discussion

The history of learning-disabilities research is a case study in the maturing of a scientific concept. It began with a categorical intuition — some children have a hidden, specific block — operationalised by a discrepancy formula that turned out to sort children on a meaningless line. It passed through a single-deficit era in which each disorder was to be explained by one core cause. And it has arrived at a dimensional, multifactorial synthesis: skills are continuously distributed, disorders are the lower tail of that distribution, and the tail is reached by the accumulation of many probabilistic risks (McGrath et al., 2020). Fifty years of work have converged on this picture across reading, mathematics, and language (Grigorenko et al., 2020).

That synthesis has practical force. If disorders are dimensional, then the diagnostic threshold is a convention, not a natural boundary, and children just above the line need help too. If they are multifactorial, then no single test can diagnose and no single intervention can cure; assessment must be broad and instruction must be sustained. And if the same risk factors feed several disorders, then comorbidity is the rule, not the exception, and a child identified with one disability should be screened for the others.

Cognitive Implications

Learning disabilities are, for cognitive psychology, a natural experiment in the architecture of skilled performance. Because they dissociate one academic ability from general intelligence, they show that reading and arithmetic rest on specific, separable cognitive systems rather than on general brightness — dyslexia isolates the phonological-to-orthographic mapping, dyscalculia the magnitude system (Vellutino et al., 2004). The disorders thereby test and confirm componential models of reading and number that were built on the normal range.

The multiple-deficit model also carries a general lesson about developmental disorders. It reframes the search for “the cause” as a category error: a complex, continuously distributed cognitive skill has no single cause, and its disorders are emergent properties of many interacting risks. That framing, worked out most fully for learning disabilities, now shapes thinking about attention, language, and other neurodevelopmental conditions, and it dissolves the old opposition between the disorder and the normal range — the same factors that make one child a strong reader make another a weak one (Peterson & Pennington, 2015).

Current Directions

Current work is pursuing the multiple-deficit model into its genetic and predictive details. Genome-wide association studies of reading and mathematics, now reaching the large samples such analyses require, are beginning to identify the common variants of small effect the model predicts and to quantify the shared genetic architecture behind comorbidity (McGrath et al., 2020). A parallel effort uses early-language and pre-reading markers to build risk scores that flag children at family risk before failure begins, shifting the field from remediation toward prevention (Snowling & Melby-Lervag, 2016).

A second front is definitional. Because the skills are dimensional, researchers continue to debate where — and whether — to draw diagnostic lines, and a Delphi-style search for professional consensus on the very definition of dyslexia reflects how unsettled the boundaries remain even as the underlying science matures (Hulme & Snowling, 2016). The synthesis of fifty years is broad agreement on mechanism coexisting with genuine disagreement on classification (Grigorenko et al., 2020).

Common Misconceptions

“A learning disability means low intelligence.”

The opposite is definitional. A learning disability is diagnosed precisely when a skill deficit is unexpected given the child's ability; intelligence is normal or above by definition of the specific disorder (Lyon et al., 2003).

“Dyslexia is seeing letters backwards.”

Dyslexia is a language-based disorder of phonological processing, not a visual one. Reversing letters is common in all beginning readers and is not the core problem (Vellutino et al., 2004).

“Children grow out of it.”

The underlying deficit is lifelong; adults compensate but the phonological weakness persists. What improves is not the deficit but its consequences, and only with instruction (Shaywitz & Shaywitz, 2005).

“It is caused by poor teaching or laziness.”

Inadequate instruction is an exclusion criterion, not a cause; the disorder is defined by failure to learn despite adequate teaching, and is substantially heritable (Grigorenko et al., 2020).

Glossary

Acalculia.

The loss of calculation ability after brain injury in a previously competent adult, distinguished from developmental dyscalculia, which is present from the start of schooling.

Comorbidity.

The co-occurrence of two or more disorders in the same person more often than chance, as with reading and mathematics disability.

Discrepancy criterion.

The now-discredited identification rule requiring achievement to fall far below the level predicted by IQ.

Double-deficit hypothesis.

The view that phonological weakness and slow rapid automatised naming are two partly independent sources of reading difficulty, with the combination the most severe.

Dyscalculia.

A specific learning disability in mathematics, rooted in an imprecise sense of numerical magnitude.

Dysgraphia.

A specific disorder of written expression affecting spelling and the composition of text, out of proportion to general ability.

Dyslexia.

A specific learning disability in reading, rooted in weak phonological processing and slow word recognition.

Intelligence.

General reasoning ability; a learning disability is defined by a skill deficit that is unexpected given otherwise normal or superior intelligence.

Matthew effect.

The widening gap by which early reading difficulty compounds into ever-greater deficits in vocabulary and knowledge.

Multiple-deficit model.

The view that a learning disability emerges from the accumulation of several probabilistic risk factors past a threshold, not from a single cause.

Phonological awareness.

The ability to perceive and manipulate the sound structure of spoken words; its weakness is the core deficit in dyslexia.

Rapid automatised naming.

The speed of naming a series of familiar items; slow naming is a second risk factor for reading disability alongside phonological weakness.

Response to intervention.

An identification and treatment framework in which a learning disability is inferred from inadequate response to high-quality, intensifying instruction.

Specific learning disorder.

The DSM-5 diagnosis for a persistent, specific deficit in reading, mathematics, or written expression, coded by domain.

Working memory.

The system for briefly holding and manipulating information; its limits are a shared cognitive risk factor across reading and mathematics disability.

Key Researchers

Brian Butterworth

(living). An emeritus cognitive neuroscientist at University College London who advanced the defective-number-module account of developmental dyscalculia and its basis in the parietal magnitude system. ORCID

Charles Hulme

(living). A psychologist at the University of Oxford whose randomised controlled trials identified which language and reading interventions causally improve literacy, moving the field from correlation to cause. ORCID

Samuel A. Kirk

(1904–1996). The educator who coined the term “learning disabilities” in 1963, giving the field its organising name and shaping US special-education policy. Wikipedia

Karin Landerl

(living). A psychologist at the University of Graz whose cross-linguistic research maps how the structure of a writing system and the number system shape reading and arithmetic deficits. ORCID

Robin L. Peterson

(living). A pediatric neuropsychologist at Children's Hospital Colorado and the University of Colorado whose work develops and tests the multiple-deficit model of learning disabilities. ORCID

Michael Rutter

(1933–2021). The founder of modern child psychiatry, whose Isle of Wight epidemiological studies established specific reading retardation as distinct from general backwardness. Wikipedia

Gerd Schulte-Korne

(living). A child and adolescent psychiatrist at Ludwig-Maximilians-University Munich who led large-scale prevalence and genetics work on learning disorders, including the population estimate cited here. ORCID

Margaret J. Snowling

(living). A psychologist at the University of Oxford whose work established the oral-language and phonological basis of dyslexia and the familial-risk longitudinal design that traces the disorder from pre-reading language to later failure. ORCID

Frequently Asked Questions

What is a learning disability?

A learning disability is a neurodevelopmental disorder in which a specific academic skill — most often reading, mathematics, or written expression — develops far below the level expected from a person's age and general ability, despite adequate instruction and intact senses. The difficulty is specific to that skill and unexpected given the person's other strengths.

Is a learning disability the same as intellectual disability?

No. Intellectual disability involves broadly below-average reasoning across all domains. A specific learning disability is, by definition, an *unexpected* deficit in one skill against otherwise normal or superior ability, and low intelligence is an exclusion criterion for it (Lyon et al., 2003).

What are the main types?

Dyslexia (reading), dyscalculia (mathematics), and disorders of written expression or dysgraphia. DSM-5 groups them as one Specific Learning Disorder coded by which academic domain is impaired, while a person may have more than one at once.

What causes learning disabilities?

They are substantially heritable and multiply determined. The multiple-deficit model holds that no single gene or experience is responsible; a disorder emerges when several genetic and environmental risk factors of small effect accumulate past a threshold (McGrath et al., 2020).

Why did the IQ-discrepancy method fall out of favour?

It required children to fail badly before qualifying (“wait to fail”), was unstable near the cut-off, and — decisively — failed to identify a meaningfully different group: poor readers with and without an IQ discrepancy share the same deficit and respond to the same instruction (Fletcher, 2009).

What is response to intervention?

A framework in which all children receive evidence-based instruction, their progress is monitored, and those who fail to respond even to intensified high-quality teaching are identified as learning-disabled. It catches children earlier and folds intervention into identification (Fuchs & Fuchs, 2006).

Do reading and mathematics disability occur together?

Frequently. They co-occur far more often than chance because they share cognitive risk factors such as working memory and processing speed, and share much of their genetic variance (Willcutt et al., 2013).

Can learning disabilities be treated?

The underlying deficit is lifelong, but its impact can be greatly reduced by direct, explicit, skill-specific instruction delivered early — structured literacy for reading, number-sense building for mathematics — with accommodations supporting access to the wider curriculum (Hulme & Snowling, 2016).

Support Organizations

• International Dyslexia Association (IDA) — research-based information, provider referral, and standards for structured-literacy instruction.
• Learning Disabilities Association of America (LDA) — advocacy, family support, and public information on the full range of learning disabilities.
• National Center for Learning Disabilities (NCLD) — policy, research translation, and resources for parents and educators.
• Understood.org — practical guidance for families and adults navigating learning and attention differences.

References

Rutter, M., & Yule, W. (1975). The concept of specific reading retardation. Journal of Child Psychology and Psychiatry, 16(3), 181–197. https://doi.org/10.1111/j.1469-7610.1975.tb01269.x

Stanovich, K. E. (1986). Matthew effects in reading: Some consequences of individual differences in the acquisition of literacy. Reading Research Quarterly, 21(4), 360–407. https://doi.org/10.1598/RRQ.21.4.1

Wolf, M., & Bowers, P. G. (1999). The double-deficit hypothesis for the developmental dyslexias. Journal of Educational Psychology, 91(3), 415–438. https://doi.org/10.1037/0022-0663.91.3.415

Lyon, G. R., Shaywitz, S. E., & Shaywitz, B. A. (2003). A definition of dyslexia. Annals of Dyslexia, 53(1), 1–14. https://doi.org/10.1007/s11881-003-0001-9

Vellutino, F. R., Fletcher, J. M., Snowling, M. J., & Scanlon, D. M. (2004). Specific reading disability (dyslexia): What have we learned in the past four decades? Journal of Child Psychology and Psychiatry, 45(1), 2–40. https://doi.org/10.1046/j.0021-9630.2003.00305.x

Landerl, K., Bevan, A., & Butterworth, B. (2004). Developmental dyscalculia and basic numerical capacities: A study of 8–9-year-old students. Cognition, 93(2), 99–125. https://doi.org/10.1016/j.cognition.2003.11.004

Shaywitz, S. E., & Shaywitz, B. A. (2005). Dyslexia (specific reading disability). Biological Psychiatry, 57(11), 1301–1309. https://doi.org/10.1016/j.biopsych.2005.01.043

Fuchs, D., & Fuchs, L. S. (2006). Introduction to response to intervention: What, why, and how valid is it? Reading Research Quarterly, 41(1), 93–99. https://doi.org/10.1598/RRQ.41.1.4

Fletcher, J. M. (2009). Dyslexia: The evolution of a scientific concept. Journal of the International Neuropsychological Society, 15(4), 501–508. https://doi.org/10.1017/S1355617709090900

Geary, D. C. (2011). Consequences, characteristics, and causes of mathematical learning disabilities and persistent low achievement in mathematics. Journal of Developmental & Behavioral Pediatrics, 32(3), 250–263. https://doi.org/10.1097/DBP.0b013e318209edef

Butterworth, B., Varma, S., & Laurillard, D. (2011). Dyscalculia: From brain to education. Science, 332(6033), 1049–1053. https://doi.org/10.1126/science.1201536

Peterson, R. L., & Pennington, B. F. (2012). Developmental dyslexia. The Lancet, 379(9830), 1997–2007. https://doi.org/10.1016/S0140-6736(12)60198-6

Willcutt, E. G., Petrill, S. A., Wu, S., Boada, R., DeFries, J. C., Olson, R. K., & Pennington, B. F. (2013). Comorbidity between reading disability and math disability: Concurrent psychopathology, functional impairment, and neuropsychological functioning. Journal of Learning Disabilities, 46(6), 500–516. https://doi.org/10.1177/0022219413477476

Moll, K., Kunze, S., Neuhoff, N., Bruder, J., & Schulte-Korne, G. (2014). Specific learning disorder: Prevalence and gender differences. PLoS ONE, 9(7), e103537. https://doi.org/10.1371/journal.pone.0103537

Peterson, R. L., & Pennington, B. F. (2015). Developmental dyslexia. Annual Review of Clinical Psychology, 11, 283–307. https://doi.org/10.1146/annurev-clinpsy-032814-112842

Snowling, M. J., & Melby-Lervag, M. (2016). Oral language deficits in familial dyslexia: A meta-analysis and review. Psychological Bulletin, 142(5), 498–545. https://doi.org/10.1037/bul0000037

Hulme, C., & Snowling, M. J. (2016). Reading disorders and dyslexia. Current Opinion in Pediatrics, 28(6), 731–735. https://doi.org/10.1097/MOP.0000000000000411

McGrath, L. M., Peterson, R. L., & Pennington, B. F. (2020). The multiple deficit model: Progress, problems, and prospects. Scientific Studies of Reading, 24(1), 7–13. https://doi.org/10.1080/10888438.2019.1706180

Grigorenko, E. L., Compton, D. L., Fuchs, L. S., Wagner, R. K., Willcutt, E. G., & Fletcher, J. M. (2020). Understanding, educating, and supporting children with specific learning disabilities: 50 years of science and practice. American Psychologist, 75(1), 37–51. https://doi.org/10.1037/amp0000452