What Learning to Read Changes in the Mind
Learning to read changes language processing because it builds a durable connection between visible symbols, spoken language, and meaning. A fluent reader can recognize written words, recover their pronunciations, identify morphological patterns, and construct meanings from sentences. With experience, the connection also works in the other direction: the spelling of a familiar word can become relevant when that word is heard. Literacy therefore changes more than visual word recognition. It reorganizes some aspects of phonological awareness, lexical representation, visual specialization, and the coordination of spoken and written language.
This transformation is especially clear in alphabetic literacy. Speech already contains systematic phonological organization, but an inexperienced listener does not necessarily treat each word as a manipulable sequence of individual phonemes. Alphabetic instruction links graphemes to speech sounds and makes such segments available for deliberate attention. Literacy experience then establishes word-specific orthographic representations, which can be activated during the processing of spoken words. Brain-imaging studies reveal changes in left ventral occipitotemporal cortex and in networks connecting visual and language-related processing. These conclusions converge across behavioral, developmental, and adult-literacy research, while the size and causal interpretation of particular effects depend on the research design. Morais et al. (1979); Dehaene et al. (2010); Dehaene et al. (2015).
The central scientific question is how a learned cultural system becomes integrated with an existing language system. Reading ordinarily builds on oral or signed language, vocabulary, grammatical knowledge, and social communication. Once literacy develops, it creates new opportunities for language learning, explicit analysis, and access to knowledge. Its psychological effects are shaped by the writing system, instructional experience, language background, age of acquisition, and amount and type of reading practice.
Literacy, Reading, and Language Processing: Core Definitions
Literacy Is a Family of Learned Skills
Literacy includes the capacity to use written language for meaningful purposes. Decoding a novel printed word, recognizing a familiar word automatically, writing an intelligible message, understanding an extended text, and evaluating written information involve overlapping but distinguishable skills. Scientific studies often operationalize literacy narrowly, through word reading or letter knowledge, because these measures can be tested precisely. Broader participation in literate practices includes reading purposes, written genres, social resources, schooling, and access to printed or digital materials.
Reading acquisition is the developmental or educational process through which people learn to connect a writing system with language. Reading fluency means increasing accuracy, speed, and automaticity in recognizing meaningful text; it does not imply that comprehension is guaranteed. Reading comprehension also depends on vocabulary, syntax, background knowledge, inference, discourse integration, and attention. These latter topics have their own research questions, while the present article examines the transformation produced by acquiring literacy itself. Castles, Rastle, and Nation (2018); Hulme and Snowling (2013).
Phonology, Orthography, and Meaning
Phonology concerns the structured sound patterns of a language, including contrastive speech categories and their organization. Orthography concerns a conventional writing system and its spellings. Graphemes are functional written units; phonemes are contrastive sound units. Morphology concerns meaningful word components, such as the relationship between teach, teacher, and teaching. Semantics concerns meaning, and syntax concerns the organization of words into structured expressions. Reading coordinates these representations across modalities rather than simply adding pictures of words to memory.
For example, readers of English learn that the written sequence light corresponds to a familiar spoken word and meaning, and that the letters gh participate in different historical spelling patterns across words. A learner can sound out an unfamiliar word, infer a meaning from context, and eventually recognize its spelling rapidly. That change reflects learning at multiple levels: letter–sound relations, word-specific form, meaning, and distributional experience. Spelling patterns can also stabilize or complicate conscious judgments about speech sounds. Ehri (2014); Rastle (2019).
Literacy Is Acquired Through Instruction and Experience
Human children acquire a spoken or signed language through sustained participation in linguistic and social environments. A conventional writing system is a historically developed technology that generally requires explicit exposure, guidance, and extended practice. The acquisition of alphabetic mappings provides a powerful route into independent word learning, but different scripts organize the relationships among signs, syllables, phonemes, morphemes, and meaning differently. No single English-specific pathway describes every writing system. Castles, Rastle, and Nation (2018); Seymour, Aro, and Erskine (2003).
How Spoken Language Supports Reading Before Reading Changes Speech
The Foundations of Early Word Reading
Spoken language supplies phonological forms and meanings that readers learn to access through print. In alphabetic systems, knowledge of letter–sound correspondences, phonemic awareness, and rapid automatized naming are important early predictors of word reading. Oral vocabulary and grammatical knowledge become increasingly important when the reader must construct the meaning of longer phrases, sentences, and passages. A child can be an effective everyday speaker while still finding arbitrary grapheme–phoneme relationships difficult to learn; conversely, accurate decoding does not guarantee rich language comprehension. Hulme and Snowling (2013).
Phonological awareness refers to deliberate attention to the structure of spoken words. It includes noticing rhyme and syllables as well as identifying, blending, deleting, or rearranging phonemes. These operations differ in complexity. A young child may recognize that cat rhymes with hat before being able to say what remains after removing the first sound from cat. Phonemic awareness is a particularly consequential aspect of learning an alphabetic system because written symbols invite the learner to analyze a spoken word into units associated with letters or letter combinations.
The development is reciprocal. Initial phonological sensitivity can facilitate alphabetic learning, while letters and decoding instruction sharpen access to phonemic structure. The historical view that phonemic awareness must be completely established before any letter instruction misses this interaction. Research on phoneme awareness interventions and systematic phonics supports teaching sound analysis in conjunction with the alphabetic principle, although instructional efficacy and the psychology of literacy-induced changes are distinct questions. National Reading Panel (2000); Castles, Rastle, and Nation (2018).
The Alphabetic Principle and Flexible Decoding
The alphabetic principle is the understanding that written units represent systematic features of spoken words. A beginning reader learns the correspondences needed to translate print into a candidate pronunciation. Decoding permits the learner to identify words never seen in print, provided the mappings, spoken vocabulary, and linguistic context support recognition. It is a generative skill: knowledge acquired from several words can be applied to unfamiliar letter strings.
English requires sensitivity to recurring spelling patterns, context-dependent pronunciations, and word morphology. Letter–sound instruction starts the process, while encounters with actual words gradually build more specialized knowledge. This combination explains why a reader can initially decode a word laboriously and later recognize it as a familiar orthographic form. English vowel spellings such as those found in bread, bead, and break illustrate why learners must move beyond simple one-letter–one-sound assumptions. Castles, Rastle, and Nation (2018); Rastle (2019).
The Development of Phonemic Awareness
Evidence From Adults Without Alphabetic Schooling
An influential study by Morais and colleagues compared adults who had remained without literacy instruction with people from broadly similar environments who had learned rudimentary reading in adulthood. Participants without literacy struggled to add or delete a phoneme at the beginning of a novel spoken item, whereas participants with some alphabetic literacy performed these tasks more successfully. The result established an important distinction between using speech fluently and consciously analyzing speech as a sequence of discrete phonemes. Morais et al. (1979).
This does not mean that adults who cannot read lack phonological representations or cannot distinguish speech sounds. Ordinary speech perception, word learning, and fluent conversation require rich sensitivity to sound contrasts and temporal patterns. Laboratory phoneme-deletion tasks place an additional metalinguistic demand on the listener: isolate a particular analytical unit, retain an unfamiliar item, and mentally manipulate its structure. The effect concerns explicit access to phonemic units within a specific task, not the existence or intelligence of the underlying speech system.
Why Learning Letters Changes the Task
A writing system supplies stable external categories for speech analysis. When a learner repeatedly relates the letter b to a consonantal element and practices combining or segmenting sounds, the learner becomes practiced in representing speech in a way useful for reading and spelling. The orthographic category supports comparison across different tokens spoken by different people. It also makes phoneme identity available for conscious discussion: a learner can name the first sound, compare sound sequences, and rehearse segment combinations.
The resulting change is often described as an expansion of metalinguistic awareness. It is better understood as a collection of specific skills than as a uniform upgrade to all thinking. Phonemic tasks vary in vocabulary, attentional requirements, memory burden, and the degree to which a writing system maps cleanly onto phonological structure. The training route may vary across alphabetic and nonalphabetic scripts. Dehaene et al. (2015); Huettig and Mishra (2014).
How Written Words Become Stored and Recognized
From Phonological Decoding to Orthographic Knowledge
At the beginning, identifying an unfamiliar spelling often requires conscious assembly of sound from letters and familiar units. Repeated successful encounters allow the reader to retain a more precise orthographic representation. A known word can subsequently be recognized with much less overt recoding. The developmental change is not a simple switch from sounding out to visual memorization. Reliable sight recognition typically depends on systematic learning of how letter sequences relate to pronunciations and meanings. Ehri (2014).
The orthographic lexicon is the research term for learned word-specific knowledge of spelling. Its representations differ in strength and specificity. A learner may recognize one word in a familiar font yet confuse a minimally different spelling; a more experienced reader can discriminate close neighbors and retrieve meaning quickly across visual formats. Orthographic representations are integrated with the spoken lexicon, morphological relationships, and semantic knowledge.
Share's Self-Teaching Hypothesis
David Share proposed that successful phonological recoding of unfamiliar printed words enables independent orthographic learning. Each encounter offers a chance to establish an item-specific link among letter order, pronunciation, and a known or inferred meaning. This mechanism helps explain how children can expand their reading vocabulary far beyond the set of words taught individually by an instructor. In a direct test, second-grade readers encountered unfamiliar printed forms in brief texts and subsequently showed retention of their spellings, including when tested after a delay. Share (1995); Share (1999).
Self-teaching does not predict that every successful decoding episode guarantees durable word learning. Attention, repetition, contextual understanding, orthographic similarity, and the accuracy of the inferred pronunciation matter. Word learning also depends on how clearly a learner represents meaning and on the opportunities that follow the first encounter. The self-teaching hypothesis is influential because it specifies a plausible mechanism for learning from ordinary reading experience, not merely a list of precursor skills.
Orthographic Mapping and Lexical Quality
Ehri's orthographic mapping account explains how grapheme–phoneme links bind spellings to pronunciations and meanings in memory. Perfetti's lexical quality hypothesis emphasizes the precision and integration of orthographic form, phonological form, and meaning. Both accounts connect reading fluency to increasingly reliable lexical representations. They differ in emphasis: one foregrounds the acquisition process, while the other focuses on how representation quality constrains performance and comprehension. Ehri (2014); Perfetti (2007).
A word can be recognized fluently yet understood only vaguely. The quality of a lexical representation includes semantic flexibility and contextual appropriateness, so reading expertise grows through increasingly varied encounters with language. Conversely, proficient speakers who encounter a word's spelling for the first time may already know its pronunciation and meaning. Literacy adds a new route into that established linguistic knowledge.
How Orthography Changes the Processing of Spoken Words
Written Forms Can Become Active During Listening
Once spelling and sound are repeatedly linked, hearing a familiar word can activate knowledge about how it is written. This is a major demonstration that literacy affects language beyond the page. For instance, a heard word may be easier or harder to classify depending on whether its phonological pattern corresponds to a consistent or inconsistent set of possible spellings. Written-language knowledge has entered the cognitive system used for spoken-word decisions. Ziegler and Ferrand (1998).
This kind of influence is often called an orthographic effect in auditory word recognition. Such effects should be interpreted according to the precise experimental task. They can reflect activation of spelling representations during auditory perception, learned links between phonology and orthography, or strategic recruitment of orthography for decisions. The claim that spelling can influence spoken-word processing is well supported, while the degree to which that influence is obligatory across all listening contexts is more constrained.
Auditory Lexical Decision and Feedback Consistency
In an auditory lexical-decision experiment, a participant hears a sound sequence and decides whether it is a real word. Ziegler and Ferrand reported longer decision times and more errors for words with spoken rimes compatible with several spellings, relative to words with more consistent spelling patterns. Later experiments by Ziegler, Petrova, and Ferrand found robust feedback-consistency effects for auditory lexical decisions in French and English, alongside null results for corresponding visual conditions under their controls. The distinction demonstrates why mechanisms inferred from one task should not automatically be generalized to another. Ziegler and Ferrand (1998); Ziegler, Petrova, and Ferrand (2008).
The auditory signal contains no letters, so consistent influences of spelling-related properties indicate learning-induced cross-modal organization or task-level access to orthographic information. Researchers continue to distinguish online feedback between processing levels from long-term restructuring of sound representations through literacy. The two accounts may produce similar behavioral patterns; experiments must identify the processes that separate them.
Changes to Speech Processing Are Selective
Literacy does not overwrite spoken language or make orthography the dominant mechanism of ordinary comprehension. Listeners can understand speech without accessing a spelling, and spontaneous conversation can succeed in societies and settings without writing. The evidence instead supports the addition of learned interactions that matter most under identifiable conditions: explicit phonemic judgments, particular lexical decisions, speech-to-print associations, and neural responses during spoken-language tasks. Huettig and Mishra (2014); Dehaene et al. (2010).
How Literacy Changes the Brain
The Visual Word Form Area
A central finding in the neuroscience of reading concerns the left ventral occipitotemporal cortex, especially a region commonly called the visual word form area (VWFA). As readers gain expertise, this region becomes increasingly responsive to familiar letter strings and written words. Its position within a larger visual recognition network allows rapid visual information to interact with learned language representations. Calling it a word-form area does not imply that it was biologically dedicated to writing before writing existed. Its specialization emerges through experience and network connectivity. Dehaene et al. (2015).
Learning to read exploits the flexibility of existing visual systems. Earlier visual pathways already support recognition of complex patterns and invariant object identities. Printed language imposes novel discrimination requirements, including attention to the identity and order of symbols whose differences may be small. A letter's orientation can distinguish different graphemes, even where mirror invariance is beneficial for object recognition. Reading experience therefore contributes to new forms of visual categorization and stable letter-string recognition.
Longitudinal Evidence in Children
In a longitudinal fMRI investigation, Dehaene-Lambertz, Monzalvo, and Dehaene followed ten young children through their first year of school. As letter knowledge and reading improved, selective neural responses to written words and digits emerged in appropriate ventral visual regions. Retrospective analyses suggested that growing word selectivity developed in territory previously only weakly specialized for other visual categories. This within-child design offers more direct evidence about developmental change than a one-time contrast between different readers. The sample was small, however, and schooling includes multiple simultaneous experiences. Dehaene-Lambertz, Monzalvo, and Dehaene (2018).
The researchers also observed changes in the relative lateralization of face-responsive regions. Such findings motivate a neural recycling account, in which reading recruits cortical territory inside a network with preexisting functional biases. Recycling describes experience-dependent specialization, not wholesale replacement of other capacities. Interpretations must respect the spatial scale of fMRI and the developmental variability of individuals.
Comparing Adult Literacy Experiences
Dehaene and colleagues studied adults differing in literacy history, including those who had not learned to read, people who acquired literacy in adulthood, and adults literate since childhood. In their fMRI results, literacy was associated with greater responses to written forms in left fusiform cortex, changes in visual processing, increased phonological responses to speech in planum temporale, and top-down orthographic activation from spoken inputs. Adult-acquired literacy showed many of these associations. This pattern supports the conclusion that the adult brain retains substantial capacity for literacy-related plasticity. Dehaene et al. (2010).
Group comparisons provide converging observations but cannot isolate schooling, social opportunity, individual learning history, and selection effects by themselves. Adults with distinct literacy backgrounds often differ in more ways than reading skill. The significance of these experiments lies in their carefully specified neural predictions and their consistency with subsequent learning studies, rather than in a claim that every between-group difference must be caused by literacy alone.
Structural and Connectivity Changes
Carreiras and colleagues examined adults who had learned to read after childhood and compared them with adults without that educational opportunity. They reported anatomical differences in white- and gray-matter features implicated in communication among reading-related regions. Structural correlates suggest sustained learning may influence the organization of distributed neural networks, although cross-sectional comparisons leave alternative explanations. Carreiras et al. (2009).
A separate longitudinal adult-literacy study by Skeide and colleagues followed adults receiving reading instruction in India. After approximately six months, changes in functional connectivity between occipital visual systems and subcortical structures, including thalamic and midbrain regions, were related to decoding gains. The result directly implicates experience-dependent changes across distributed visual pathways. It does not mean that the study identified a universal neural signature applicable to every script, learner, or instructional program. Skeide et al. (2017).
Reading and Spoken-Language Networks Interact
Neural reading systems must link print to phonology and meaning. Literacy-related activity in ventral visual regions interacts with temporal, parietal, and frontal networks supporting language and attention. Some pathways are especially relevant to assembling pronunciations from print, while others participate in rapid access to familiar spelling–meaning relations. These are functional tendencies within interacting networks rather than isolated one-purpose modules. The broader organization of spoken language continues to supply linguistic structure and meaning. Dehaene et al. (2015); Rastle (2019).
The neuroscience therefore supports a reciprocal view: language helps establish print representations, and literacy changes the systems used for certain forms of language processing. The actual direction of influence depends on whether a person is decoding, recognizing, listening, spelling, or constructing meaning.
Reading Development Across Writing Systems
Alphabetic Depth and the English Case
Writing systems differ in the consistency of their mappings between symbols and speech. Orthographic depth is a term for how transparent or variable these mappings are. In relatively transparent alphabetic orthographies, frequent grapheme–phoneme correspondences can support accurate decoding comparatively early. English presents more context-sensitive and variable sound–spelling relationships. A multinational study comparing foundation reading in thirteen European orthographies found substantial differences in early word and nonword reading, consistent with a role for orthographic depth and syllable complexity. These are group-level patterns, not fixed abilities of speakers of different languages. Seymour, Aro, and Erskine (2003).
A deep orthography can make simple decoding less predictable while preserving regularities at other linguistic levels. English spellings often carry morphological information: related words retain common written patterns even when pronunciation shifts. This can make a written word family useful for connecting meaning and form. It also complicates the assumption that the purpose of writing is to encode phonemes alone. Rastle (2019).
Morphology and the Development of Skilled Reading
Morphemes are recurring units contributing to meaning or grammatical function. Written words often reveal morphological families, affixes, and relationships that help organize a reader's lexical knowledge. In English, the spelling of a base may remain relatively stable across inflectional and derivational forms. Morphological information therefore offers another bridge between word recognition and comprehension.
Rastle's review emphasizes that skilled readers show rapid sensitivity to morphological structure, while the developmental path to that automatic analysis remains an active research question. Explicit knowledge of word parts in schoolchildren is not equivalent to the fast, apparently automatic morphological decomposition seen in skilled adult reading experiments. The distinction illustrates how literacy can make a structure available for conscious reflection long before it becomes an efficient component of online recognition. Rastle (2019).
Nonalphabetic and Mixed Systems
Not all writing systems encode language at the phoneme level. Some emphasize syllabic, morphemic, or mixed correspondences, often with substantial phonological information. Reading development will accordingly depend on the statistical structure of the script, the spoken language, and the mapping the learner must master. Results from English phoneme-deletion tasks or European alphabetic systems should be generalized carefully to readers of Chinese characters, Japanese scripts, or other orthographies.
A crosslinguistic perspective reveals both broad convergence and meaningful variation: learned visual symbols become linked with linguistic knowledge, but their functional units and learning demands are not identical. In their 2026 synthesis of bilingual reading research, Kovelman and Sun emphasize that the developmental reading brain is shaped by crosslinguistic transfer, language experience, and orthographic differences. Their review offers current evidence about variation in bilingual populations; it should not be treated as evidence that bilingualism automatically improves every aspect of reading or executive function. Kovelman and Sun (2026).
The Relationship Between Literacy, Vocabulary, and Comprehension
Print Opens Additional Paths to Language Knowledge
A reader encounters words, syntactic structures, topics, and discourse conventions that may be less frequent in everyday conversation. With growing fluency, text can become an independent source of vocabulary learning and conceptual knowledge. The resulting exposure may improve lexical familiarity and explicit access to linguistic forms. These consequences depend on how much and what kind of material people read; basic decoding alone does not guarantee sustained engagement with meaningful text.
Written and spoken vocabulary interact. Familiar spoken words make early printed words easier to identify, while new words encountered in text can become available in later conversation. Reading also exposes learners to punctuation, conventional spelling, text organization, and genres of explanation. Language comprehension draws on knowledge acquired across all these experiences. Perfetti (2007); Castles, Rastle, and Nation (2018).
Automatic Recognition Frees Capacity for Meaning
Early decoding consumes attention. When accurate word identification becomes more automatic, the reader can allocate more processing resources to following propositions, making inferences, and integrating meaning across sentences. The advantage is conditional: a fluent reader who lacks relevant vocabulary or background knowledge may still misunderstand a text. A reader who can explain a passage orally may nevertheless struggle when visual word identification remains slow.
These dissociations motivate research that separates decoding, listening comprehension, vocabulary, and text-level performance. The psychological effects of learning to read emerge both from changes in the identification of written words and from new opportunities to learn through print. The mechanisms of constructing discourse-level meaning warrant their own specialized treatment. Hulme and Snowling (2013); Castles, Rastle, and Nation (2018).
What Literacy Does to Metalinguistic Awareness and Memory
Explicit Language Analysis
Literacy supports deliberate judgments about written and spoken form. A reader can compare rhymes, analyze phoneme positions, identify affixes, spell unfamiliar forms, and treat sentences as objects of reflection. Some of these activities depend on learned orthographic conventions, while others expand preexisting capacities to notice and reason about language. The most robust literacy-linked findings concern specific forms of phonological awareness and orthographic knowledge.
An increase in performance on an explicit language task must be interpreted in relation to its instructions and familiarity. People may solve spoken tasks through verbal reasoning, orthographic imagery, rehearsal, or strategies learned in school. The study of literacy therefore needs tasks that distinguish perceptual processing from conscious manipulation and verbal knowledge from knowledge of spelling. Morais et al. (1979); Huettig and Mishra (2014).
Memory and Cognitive Performance
Written language changes access to information: a note, document, or searchable text can store a message outside biological memory and be consulted repeatedly. Literacy also creates opportunities for vocabulary growth and practiced manipulation of linguistic sequences. Experimental differences between literacy groups have been observed in some memory, visual, and metalinguistic tasks, but such differences may involve schooling, occupational practice, and familiarity with test conventions. A broad claim that learning to read universally increases memory capacity or general intelligence would exceed the evidence discussed here. Huettig and Mishra (2014).
The strongest inference is task-specific: alphabetic literacy supports deliberate phoneme manipulation, written-word learning, and print-mediated knowledge acquisition. Claims about global cognitive transformation require careful operational definitions and comparison groups. General memory systems remain a separate research domain, even when memory helps explain how word forms become learned.
Theories of the Reading–Language Relationship
Neural Recycling and Cultural Learning
The neural recycling hypothesis proposes that recently invented cultural skills recruit and reshape preexisting cortical circuits. Writing became widespread far more recently than the evolution of visual recognition, so the brain uses adaptable systems rather than an inherited region devoted exclusively to print. Evidence of emerging ventral visual specialization during reading instruction, together with adult-learning effects, makes this a compelling explanatory framework. It remains a theory about the mechanisms and constraints of neural organization, and particular predictions should be judged against longitudinal and intervention evidence. Dehaene et al. (2015); Dehaene-Lambertz, Monzalvo, and Dehaene (2018).
Orthographic Mapping, Self-Teaching, and Lexical Quality
Three influential accounts address complementary questions. Orthographic mapping describes how specific spelling–sound bonds can be stored. The self-teaching hypothesis addresses how independent decoding events establish novel written-word knowledge. Lexical quality emphasizes how precise and integrated knowledge of spelling, sound, and meaning influences fluent identification and comprehension. These accounts are useful at different explanatory levels and are not interchangeable labels for one single process. Ehri (2014); Share (1995); Perfetti (2007).
Interactive Written and Spoken Language Processing
Interactive theories allow activation to flow among orthography, phonology, and semantic representations. They predict that reading and listening can recruit overlapping lexical information, while different tasks give different weights to each representation. Auditory orthographic-consistency effects are important evidence for cross-modal coupling. Yet models must specify whether the measured effect reflects an automatic feedback mechanism, long-term restructuring, or a conscious decision strategy. Theoretical sophistication depends on predictions that separate these possibilities rather than assigning every response-time difference to the same mechanism. Ziegler, Petrova, and Ferrand (2008).
How Researchers Establish the Effects of Literacy
Cross-Sectional Comparisons
Researchers compare groups with different reading histories using phonemic tasks, spoken lexical decisions, visual recognition, structural imaging, and functional neuroimaging. These studies can reveal strong associations and motivate mechanistic hypotheses. They are especially informative when groups are matched on relevant aspects of social and language background. Their main interpretive limitation is that literacy status often correlates with education, occupational experience, household conditions, and prior exposure to formal testing.
Longitudinal Studies and Intervention Designs
Longitudinal studies measure the same learners repeatedly as reading develops. They help establish the time course of skill acquisition and associated behavioral or neural changes. Interventions and controlled training can test whether instruction changes a targeted capability, with the quality of the control condition determining how narrowly the causal effect can be attributed. Child neuroimaging and adult literacy-training studies are particularly valuable for evaluating developmental plasticity. Dehaene-Lambertz, Monzalvo, and Dehaene (2018); Skeide et al. (2017).
Behavioral Experiments
Phoneme deletion, segmentation, and blending tasks measure explicit access to speech-sound structure. Lexical decision tests how rapidly or accurately participants classify letter strings or auditory items. Pseudoword learning experiments examine how new spellings are retained after decoding opportunities. Each paradigm isolates a component of processing at the cost of introducing task-specific demands. Reliable inferences require carefully matched stimuli, appropriate baselines, and alternative explanations.
Eye Tracking, Electrophysiology, and Neuroimaging
Eye tracking records fixations and saccades during visual reading and can clarify the timing of information uptake, but it cannot by itself identify a unique mental representation. Event-related potentials allow millisecond-level investigation of responses to print and speech. fMRI identifies hemodynamic changes correlated with regional processing and network organization; its temporal and spatial interpretation is indirect. Structural MRI and connectivity methods investigate organization across longer learning periods. Convergence across methods provides stronger evidence than any single measure. Dehaene et al. (2010); Dehaene et al. (2015).
Crosslinguistic Designs and Measurement Equivalence
Researchers comparing scripts must account for phonological grain size, spelling consistency, morphology, instructional methods, and cultural context. Two tasks given the same English-language name may require different operations across scripts. A strong crosslinguistic study therefore establishes that the measures capture comparable skills and reports both shared patterns and context-specific differences. Research on bilingual and biliterate learners also separates proficiency and literacy in each language from the fact of using multiple languages. Seymour, Aro, and Erskine (2003); Kovelman and Sun (2026).
What the Evidence Establishes and What Remains Debated
Established Findings
Alphabetic literacy facilitates conscious access to individual phonemic segments. Reading experience builds detailed orthographic representations linked to pronunciations and meanings. The learning process produces reliable experience-related specialization for written words in ventral visual systems. Reading and speech processing interact, and spelling knowledge can affect at least some spoken-word tasks. Targeted phonological decoding and phoneme training can support early alphabetic reading. These claims have converging evidence from experiments, reviews, and learning studies. Morais et al. (1979); Dehaene et al. (2010); Castles, Rastle, and Nation (2018).
Evidence Requiring Greater Qualification
The exact size and universality of spelling effects during natural speech comprehension remain dependent on task and model. Structural brain differences associated with literacy may partly reflect prior experience and sample characteristics. Some developmental findings are based on small longitudinal neuroimaging samples. Effects on broad cognition, reasoning, and memory are difficult to separate from schooling, socioeconomic environment, and written-language practice. The size and form of literacy-induced changes may differ across scripts and histories of learning. Huettig and Mishra (2014); Dehaene et al. (2015).
The Difference Between Mechanism and Educational Prescription
Evidence that decoding events support orthographic learning helps explain why systematic grapheme–phoneme instruction is consequential for novice readers of alphabetic scripts. It does not establish one universally optimal sequence for every learner, language, or school system. Classroom interventions combine teacher expertise, materials, practice, feedback, vocabulary, and comprehension instruction. Recommendations must draw on intervention research as well as cognitive theory. The National Reading Panel provides historically important synthesis of phonemic awareness and phonics studies, while modern reading-science reviews explain how those early skills develop into skilled word reading and comprehension. National Reading Panel (2000); Castles, Rastle, and Nation (2018).
Individual Differences, Age, and Educational Context
Children Learn at Different Rates
Learners vary in the quality of phonological representations, language proficiency, attention, visual familiarity, opportunities for practice, and how readily they map print onto known words. They also encounter different writing systems and educational environments. The same observed reading speed may reflect distinct combinations of familiarity, word complexity, fluency, and comprehension. These differences should be examined developmentally rather than converted into diagnostic labels. Hulme and Snowling (2013).
Adults Can Develop New Literacy-Related Networks
Research on adult literacy shows that neural reorganization remains possible after childhood. Training is linked with changes in visual and subcortical connectivity, and some adult readers exhibit neural patterns also associated with childhood-acquired literacy. The learning experience of an adult will nevertheless differ from that of a child because vocabulary, world knowledge, habitual attention, and established language use already have long histories. Adult literacy should be studied as a genuine learning trajectory with its own resources and barriers. Dehaene et al. (2010); Skeide et al. (2017).
Biliteracy and Multilingual Learners
A multilingual learner may know a concept in one language before learning the written word in another. Orthographic regularity, phonological inventories, vocabulary, and exposure can influence which learning skills transfer and which require separate practice. Bilingual literacy research shows that interpreting a reading measure requires information about proficiency and experience in the relevant languages. Multilingualism should not be used as a proxy for reading difficulty or as proof of an automatic cognitive advantage. Kovelman and Sun (2026).
Literacy Is Distinct From a Clinical Diagnosis
Slow reading, weak phonemic awareness, limited exposure to print, and unfamiliarity with a language can arise for different reasons. A scientific account of normal literacy acquisition does not determine whether a person has dyslexia, developmental language disorder, or another diagnosable condition. Such evaluations require developmental histories, appropriate measures, linguistic context, and clinical or educational expertise. This article describes shared reading mechanisms and learning-related plasticity; condition-specific diagnosis and intervention belong to specialized clinical sources.
Practical Implications for Language Learning and Research
For educators, the mechanisms imply that linking sounds and letters, giving meaningful opportunities to decode unfamiliar words, strengthening vocabulary, and sustaining contact with connected text play complementary roles. Progress is better characterized by changes in accuracy, automaticity, and comprehension than by a single test score. Practicing words only as visual shapes leaves out the powerful systematic connections that support generalization in alphabetic systems. Yet decoding activities should always lead toward understanding and meaningful language use. Castles, Rastle, and Nation (2018); Ehri (2014).
For researchers, literacy illustrates that perceptual and linguistic systems remain plastic throughout learning. It also warns against treating performance on culturally familiar laboratory tasks as a transparent measure of universal mental capacity. Researchers studying speech, phonological awareness, visual recognition, or brain lateralization should consider literacy history as a relevant participant characteristic. Orthographic experience can influence results even when the stimulus is spoken rather than written. Huettig and Mishra (2014); Ziegler, Petrova, and Ferrand (2008).
For learners, the scientific message is that literacy grows through repeated successful interactions with a writing system, spoken-language knowledge, and meaningful text. Word recognition becomes more efficient as representations become detailed and interconnected. Reading is an acquired form of language experience that subsequently creates new forms of language experience.
Frequently Asked Questions
Does learning to read change how people hear spoken words?
It can. Orthographic knowledge influences performance in some spoken-word recognition tasks, and literacy is associated with changes in neural responses to speech. The effects depend on the task and should not be interpreted as a change in all aspects of everyday hearing. Ziegler and Ferrand (1998); Dehaene et al. (2010).
Can someone speak a language fluently without knowing how to read it?
Yes. Rich spoken or signed language competence can develop without literacy in a conventional writing system. Literacy adds learned routes for representing and analyzing language. Difficulty with phoneme-deletion tasks among adults without alphabetic training does not establish a lack of ordinary speech perception or vocabulary. Morais et al. (1979).
Why does learning letters help phonemic awareness?
Letters supply stable representations that can be systematically connected to speech segments. Instruction and practice support deliberately identifying, blending, and manipulating phonemes. Early phonological abilities also make letter learning easier, so the relationship develops in both directions. Hulme and Snowling (2013).
What changes in the brain when reading is learned?
Visual regions, especially the left ventral occipitotemporal system associated with written word forms, become more responsive to print. Connections with phonological and language systems are also affected. Longitudinal child and adult studies support experience-dependent neural plasticity. Dehaene-Lambertz, Monzalvo, and Dehaene (2018); Skeide et al. (2017).
Does reading automatically improve intelligence or memory?
Evidence does not justify a universal increase in general intelligence or memory capacity attributable to reading alone. Literacy creates powerful tools for information access and supports particular language-related skills, while broader associations can reflect education, cultural practice, and social opportunity. Huettig and Mishra (2014).
Is learning to read the same process in every language?
The general challenge of connecting visual symbols to language is shared, but scripts differ in grain size, consistency, morphology, and educational practice. Findings from English alphabetic reading cannot be transferred without examination to every writing system. Seymour, Aro, and Erskine (2003); Kovelman and Sun (2026).
Can adults learn to read and show measurable neural changes?
Yes. Adult literacy studies document changes associated with new reading skills, including functional and connectivity changes. This demonstrates continued plasticity while leaving room for substantial individual and educational differences. Dehaene et al. (2010); Skeide et al. (2017).
Conclusion
Learning to read changes language processing by establishing a practiced system of relationships among print, sound, morphology, and meaning. The strongest evidence documents increased explicit phonemic awareness, the growth of word-specific orthographic representations, measurable interactions between spelling and spoken-word processing, and functional reorganization of neural systems supporting written language. Reading development also expands access to linguistic forms and knowledge available through text. Its broader cognitive consequences depend on the learning environment, linguistic history, and method used to measure change. Literacy is a powerful example of how cultural experience becomes integrated with human language and brain organization.
Related Articles
Psycholinguistics: How Language Is Processed in the Mind — Experimental concepts and methods for studying how language is represented and processed.
Word Recognition: How the Mind Identifies Spoken and Written Words — Processes involved in identifying familiar words across speech and print.
Speech Perception: How the Mind Turns Sounds Into Words — The perceptual basis of spoken-language recognition.
Language Processing in the Brain: Cognitive Networks and Neural Systems — The wider neural systems supporting language operations.
Semantic Memory: Facts, Meaning, and Knowledge — How stored meanings and conceptual knowledge contribute to language use.
Reading Comprehension: How Language, Context, and Inference Create Meaning — related psycholinguistic findings and mechanisms.
References
Carreiras, M., Seghier, M. L., Baquero, S., Estévez, A., Lozano, A., Devlin, J. T., and Price, C. J. (2009). An anatomical signature for literacy. Nature, 461, 983–986. https://doi.org/10.1038/nature08461
Castles, A., Rastle, K., and Nation, K. (2018). Ending the reading wars: Reading acquisition from novice to expert. Psychological Science in the Public Interest, 19(1), 5–51. https://doi.org/10.1177/1529100618772271
Dehaene, S., Cohen, L., Morais, J., and Kolinsky, R. (2015). Illiterate to literate: Behavioural and cerebral changes induced by reading acquisition. Nature Reviews Neuroscience, 16, 234–244. https://doi.org/10.1038/nrn3924
Dehaene, S., Pegado, F., Braga, L. W., Ventura, P., Nunes Filho, G., Jobert, A., Dehaene-Lambertz, G., Kolinsky, R., Morais, J., and Cohen, L. (2010). How learning to read changes the cortical networks for vision and language. Science, 330(6009), 1359–1364. https://doi.org/10.1126/science.1194140
Dehaene-Lambertz, G., Monzalvo, K., and Dehaene, S. (2018). The emergence of the visual word form: Longitudinal evolution of category-specific ventral visual areas during reading acquisition. PLOS Biology, 16(3), e2004103. https://doi.org/10.1371/journal.pbio.2004103
Ehri, L. C. (2014). Orthographic mapping in the acquisition of sight word reading, spelling memory, and vocabulary learning. Scientific Studies of Reading, 18(1), 5–21. https://doi.org/10.1080/10888438.2013.819356
Huettig, F., and Mishra, R. K. (2014). How literacy acquisition affects the illiterate mind: A critical examination of theories and evidence. Language and Linguistics Compass, 8(10), 401–427. https://doi.org/10.1111/lnc3.12092
Hulme, C., and Snowling, M. J. (2013). Learning to read: What we know and what we need to understand better. Child Development Perspectives, 7(1), 1–5. https://doi.org/10.1111/cdep.12005
Kovelman, I., and Sun, X. (2026). The bilingual reading brain: A developmental and crosslinguistic perspective. Annual Review of Developmental Psychology, 8, advance online publication. https://doi.org/10.1146/annurev-devpsych-010923-114059
Morais, J., Cary, L., Alegria, J., and Bertelson, P. (1979). Does awareness of speech as a sequence of phones arise spontaneously? Cognition, 7(4), 323–331. https://doi.org/10.1016/0010-0277(79)90020-9
National Reading Panel. (2000). Report of the National Reading Panel: Teaching children to read. National Institute of Child Health and Human Development. https://www.nichd.nih.gov/publications/pubs/nrp/findings
Perfetti, C. A. (2007). Reading ability: Lexical quality to comprehension. Scientific Studies of Reading, 11(4), 357–383. https://doi.org/10.1080/10888430701530730
Rastle, K. (2019a). EPS mid-career prize lecture 2017: Writing systems, reading, and language. Quarterly Journal of Experimental Psychology, 72(4), 677–692. https://doi.org/10.1177/1747021819829696
Rastle, K. (2019b). The place of morphology in learning to read in English. Cortex, 116, 45–54. https://doi.org/10.1016/j.cortex.2018.02.008
Seymour, P. H. K., Aro, M., and Erskine, J. M. (2003). Foundation literacy acquisition in European orthographies. British Journal of Psychology, 94(2), 143–174. https://doi.org/10.1348/000712603321661859
Share, D. L. (1995). Phonological recoding and self-teaching: Sine qua non of reading acquisition. Cognition, 55(2), 151–218. https://doi.org/10.1016/0010-0277(94)00645-2
Share, D. L. (1999). Phonological recoding and orthographic learning: A direct test of the self-teaching hypothesis. Journal of Experimental Child Psychology, 72(2), 95–129. https://doi.org/10.1006/jecp.1998.2481
Skeide, M. A., Kumar, U., Mishra, R. K., Tripathi, V. N., Guleria, A., Singh, J. P., Eisner, F., and Huettig, F. (2017). Learning to read alters cortico-subcortical cross-talk in the visual system of illiterates. Science Advances, 3(5), e1602612. https://doi.org/10.1126/sciadv.1602612
Ziegler, J. C., and Ferrand, L. (1998). Orthography shapes the perception of speech: The consistency effect in auditory word recognition. Psychonomic Bulletin and Review, 5, 683–689. https://doi.org/10.3758/BF03208845
Ziegler, J. C., Petrova, A., and Ferrand, L. (2008). Feedback consistency effects in visual and auditory word recognition: Where do we stand after more than a decade? Journal of Experimental Psychology: Learning, Memory, and Cognition, 34(3), 643–661. https://doi.org/10.1037/0278-7393.34.3.643