Language acquisition and learning to read in English and Chinese

The study of spoken and written language acquisition is an inherently interdisciplinary area, at the interface between developmental psychology, theoretical linguistics, phonology, and cognitive psychology. It is also an area where cross-cultural comparative studies across Chinese and English speakers are highly relevant. Additionally, the development of children with atypical speaking and writing abilities can further illuminate the developmental issues involved in spoken and written language acquisition.

1. Spoken language acquisition

a. Social

The acquisition of spoken language is a remarkable feat that children achieve at a young age and with little explicit training. Observation of this fact has prompted many questions from psychologists, philosophers, and linguistics: are humans "hard wired" (biologically determined) to acquire language? Or is language just a particular type of learning that comes from being in the right sort of environment? This debate has often been called the 'nature-nurture' debate and continues to be an area of discussion and controversy in contemporary work on language acquisition.

i. Nature-Nurture Debate

Skinner, B. (1957).  Verbal Behavior. New York: Appleton-Century-Crofts.

Here Skinner offers a classic behaviorist account of language acquisition, which he terms verbal behavior. According to this model, language learning is a consequence of being raised in an environment with exposure to correct language usage, where correct utterances were rewarded and errors were punished. For example, Skinner claimed that children are rewarded for talking when their utterances are met with nods or other active signs of approval. As children grow older parents have increasingly more strict requirements on what counts as sufficient for communication. In this way, they shape the linguistic behavior of their children. On this model, language learning is dependent upon having the right reward structure in the environment, with little to none of the linguistic structure coming from the child.

Chomsky, N. (1959).  "Review of Verbal Behavior by B.F. Skinner." Language, 35, 26-58.

Chomsky's review of Skinner's Verbal Behavior has long been seen as a devastating objection to Skinner's view, and as a key text in the 'cognitive revolution' that initiated the study of cognitive science after several decades in which behaviorism had been the preponderant force. Chomsky offers a line of arguments as to why Skinner's account of language acquisition is insufficient. The central thesis of the critique is that the notions of 'stimulus,' 'response' and 'reward' have no coherent application to language learning. Language learning will require descriptions of the processing that occurs in the learner's mind, an idea antithetical to the central tenets of behaviorism. In concluding, he offers suggestions as to how such an account of language might be formulated.

Chomsky, N. (1975).  Reflections on Language. New York: Pantheon.

In this, and many other of his writings on language, Chomsky argued for the child's innate disposition towards language acquisition. Here he posits the idea of a language acquisition device – a type of mental organ that children are born with which makes the acquisition of language possible. This device is meant to have its own features of organization that make it distinct from other cognitive systems and the types of learning they afford. This device is often described as containing a universal grammar which can be set to the parameters of the particular language in which the child is born. Thus, contrary to Skinner, Chomsky defends the view that the bulk of the impetus for language learning comes from the language learner himself, not the environment.

Pinker, S. (1994).  The Language Instinct: How the Mind Creates Language. New York: Harper Collins.

Pinker offers a more recent defense of a Chomskyan view of language acquisition. His view is distinct from Chomsky's in that he offers a natural selection argument to explain the origins of natural language. Language, on Pinker's view, is an instinct that humans have evolved because it has afforded a series of advantages (coordination of behavior, expansion of cognitive resources, etc). Pinker offers further elaboration of Chomsky's ideas of a universal grammar and highlights a wide variety of evidence in support of this idea.

Elman, J., Bates, E., Johnson, M., Karmiloff-Smith, A., Parisi, D., & Plunkett, K. (1996).   Rethinking Innateness: A Connectionist Perspective on Development. Cambridge, MA: MIT Press/Bradford Books.

Elman et al.'s book offers a challenge to the understanding of language acquisition as an innate ability, as forwarded by Chomsky and Pinker. The authors press for a clear definition of innateness, and argue that current formulations of nativism ignore crucial facts about the interactions that happen between the mind and its surrounding environment. They acknowledge the existence of certain internal constraints on the possibilities for language learning, but emphasize the brain's plasticity regarding particular features of language acquisition. In particular, they deny two claims: 1) that language learning is a wholly distinct form of learning, and 2) that there are no significant explanatory factors for language acquisition in the environment. Further, the authors draw on a host of evidence that challenges assumptions about the nature and function of a language acquisition device.

ii. Critical Period

The general view that there is an early period in life during which we are especially prepared to acquire language is called the critical period hypothesis. The first and most famous formulation of this hypothesis comes from Lenneberg.

Lenneberg, E. (1953).   "Cognition in ethnolinguistics." Language, 29, 463-471.

Lenneberg, E. (1964).  "A biological perspective of language." In E. Lenneberg (ed), New Directions in the Study of Language, Cambridge, MA: MIT Press (pp. 65-88).

Lenneberg was an ethnobiologist who studied language in comparison to the development of other species-typical behaviors in other animals. His formulation of the critical period hypothesis argues for a biological understanding of language acquisition. He viewed language as the key species-typical behavior for humans. And, as with many such behaviors, he argued that there exists a period of brain development that is ideally suited for acquiring them and that once this period is over acquisition of the behavior is difficult if not impossible. Language must be acquired before cerebral dominance is established, so that language can be localized in the left hemisphere. At the time of his work, cerebral dominance was thought to be achieved by puberty, and thus Lenneberg saw puberty as marking the closing of the critical period of language acquisition.

iii. Feral Children

Children who have grown up in isolation or in the wild are often referred to as feral children. Although no one ever wishes such a fate for a child, discovery of such cases can help us to understand what abilities, particularly linguistic ones, require immersion in a language community.

Lane, H. (1976).   The wild boy of Aveyron. Cambridge, MA: Harvard University Press.

In this book, Lane tells the story of Victor, a boy found in the woods of rural France in 1800. Victor was 12 or 13 at the time of his capture, and was cared for by Jean-Marc Itard, who wanted to teach Victor to speak and socialize normally. Itard was able to do much to help Victor, but Victor never learned to speak. He was only capable of mastering a few small phrases. Cases like this have been taken to support the existence of a critical period during which a child must receive exposure to language in order to develop full-blown linguistic abilities.

Fromkin, V., Krashen, S., Curtiss, S., Rigler, D., & Rigler, M. (1974).   "The development of language in Genie: A case of language acquisition beyond the critical period." Brain and Language, 1, 81-107.

Genie is perhaps the most famous case of childhood isolation. Genie, as she is called, was discovered at age 13 and had been locked into a dark room for most of her life. Genie had little exposure to language or any outside stimulation during her isolated childhood. When she was rescued, Genie was undernourished, with no language or social skills. A team of researchers worked with Genie over the next several years to help her improve in all of these areas. Though she improved in many areas with their help, her difficulties with language persisted. Genie was able to learn many words, and even developed the ability to communicate with those around her. However, she never mastered English syntax, and would constantly confuse word order, tense, etc. The researchers thus suggest a refinement of the critical period hypothesis. Some aspects of language learning are possible after puberty, but full language learning might not be.

iv. Motherese

Motherese is the name given to the type of language adults typically use when speaking to young children. This speech is characterized by shorter utterances involving concrete objects with intonational exaggeration. This manner of speaking has been shown to help children in the acquisition of their native language.

Furrow, D., Nelson, K., & Benedict, H. (1979).   "Mothers' speech to children and syntactic development: Some simple relationships." Journal of Child Language, 6, 423-442.

In this study Furrow et al. investigated the relationship between children's linguistic environment and their acquisition of language. The researchers took speech samples from the parents of newborns and compared features of those utterances to the child's later language abilities at one and a half and two years of age. They found that many of the features of the mother's speech were correlated with the child's later speech. The more characteristics of motherese the mother's language displayed, the more likely the children were to be above-average in their own language abilities. The researchers concluded that motherese does in fact aid language development.

Gleitman, L., Newport, E., & Gleitman, H. (1984).   "The current status of the motherese hypothesis." Journal of Child Language, 11, 43-79.

Through a re-examination of the Furrow et al. and other data on the benefits of motherese, Gleitman et al. concluded that motherese does have an impact on child language acquisition, but that most of the benefits of motherese are seen during language development from one and a half to two years of age, after which the effects of motherese are less prominent. Further, they discovered that a mother's inclusion of some complex speech during the second year of a child's life can actually help a child's language growth more than continued motherese.

Cooper, R., Abraham, J., Berman, S., & Staska, M. (1997).   "The development of infants' preference for motherese." Infant Behavior & Development, 20, 477- 488.

Cooper et al. investigated the extent to which young infants prefer motherese to typical adult-directed speech. By comparing motherese to adult-directed speech, the researchers were able to show not only that children prefer the motherese, but further, that they have the ability to recognize speaker-specific characteristics and show a preference for that which most closely resembles their mother's.

Masataka, N. (1992).   "Motherese in a signed language." Infant Behavior & Development, 15, 453-460.

Matasaka investigated the possibility of motherese in sign language through an examination of Japanese sign language. Matasaka concluded that mothers of deaf infants do slow down their signs and exaggerate their movements. It is therefore concluded that there is an analogue of motherese in sign language.

v. Pidgins/Creoles

Often times groups from different language communities need to talk to one another. In such situations, pidgin languages emerge. Pidgins are rudimentary languages that evolve when many different language groups must co-exist and communicate. Creoles are the languages developed by the children who grow up in pidgin language environments. The creole language involves sophistication of the pidgin – tense, clause structure, etc. are brought in to supplement the earlier more basic pidgin. Creolization is thus often seen as evidence of children's innate linguistic abilities.

Bickerton, D. (1983).   "Creole languages." In S. Wang (ed) The Emergence of Language: Development and Evolution, New York: W.H. Freeman (pp. 59-69).

Bickerton, D. (1984).  "The language bioprogram hypothesis." Behavioral and Brain Sciences, 7, 173-221.

Bickerton is known for introducing the language biogram hypothesis. This hypothesis states that children have an innate grammar available to them that they can draw upon if the language input in their environment is insufficient. In this way, it acts as a backup system, ensuring that the child has the ability to develop a system for communicating with others. Bickerton's evidence for this hypothesis comes from pidgin and creole languages. In his discussion of creoles Bickerton argues for a refinement of Chomsky's description of the child's language acquisition device. Instead of having switches that must be engaged by the child's native language (as Chomsky argued), Bickerton claims that children are capable of selecting basic language parameters for themselves when no such cues are available.

Goldin-Meadow, S., & Mylander, C. (1990).   "Beyond the input given: The child's role in the acquisition of language." Language, 66, 323-355.

In studies of congenitally deaf children, Goldin-Meadow and Mylander further the language biogram hypothesis by showing children's ability to provide structure to their linguistic environment even when they are not exposed to the sounds of spoken language. The researchers showed that deaf children who had not been exposed to any formal sign language often create their own form of gestural language – called homesign – that bears a close resemblance to the stage language development of young speaking children. For example, they go through stages of single- and multi-word utterances with increasing amounts of syntactic and morphological structure.

Kegl, J., Senghas, A., & Coppola, M. (1999).   "Creation through contact: Sign language emergence and sign langauge change in Nicaragua." In M. DeGraff (Ed) Language Creation and Language Change: Creolization, Diachrony, and Development, Cambridge, MA: MIT Press (pp. 179-237).

vi. Whorfian Hypothesis

Whorf, B. (1956).   Language, Thought, and Reality. Cambridge, MA: MIT Press.

Whorf was an anthropologist who studied the relation between language and thought. Stated simply, his view was that language determined thought. This involved two particular claims. First is the claim of linguistic determinism. This states that language determines various nonlinguistic cognitive processes, such as perception. The second claim is of linguistic relativity. This states cognitive processes are determined in different ways for different languages. One of Whorf's famous examples is of an Eskimo group that possesses 15 distinct words for snow. Having this many words for snow, Whorf argued, allowed them to perceive differences in types of snow that aren't available to speakers of other languages without such distinctions.

Whorf's views have received a great deal of criticism over the years. For a particular critique of the snow example just mentioned, see: Martin, L. (1986). "Eskimo words for snow: A case study in the genesis and decay of an anthropological example." American Anthropologist, 88, 418-423.

Lucy, J. (1992).   Language Diversity and Thought: A Reformulation of the Linguistic Relativity Hypothesis. Cambridge: Cambridge University Press.

Lucy's books offers a review of the Whorfian hypothesis of linguistic relativity and how it has evolved over time. Further, Lucy examines the various empirical findings that were thought to bolster and refute Whorf's claims. The author also includes suggestions of lines of empirical investigation over which this issue could continue to be fruitfully explored.

Hurford, J. (1987).  Language and Number: Cambridge: Cambridge University Press.

One of the central areas in which debates about the possibility of linguistic determinism persists is that of a language's representations of numbers and its impact on mathematical achievement. Specifically, many researches have examined the relation between languages that have very regular ways of expressing numerical relationships – such as Asian languages – and languages where the numerals have much less regularity, such as English. Hurford offers a thorough discussion of these differences and the differences in the ease with which children learn to count across language systems.

Miller, K., Smith, C., Zhu, G., and Zhang, H. (1995).  "Preschool origins of cross-national differences in mathematical competence: The role of number- naming systems." Psychological Science, 6, 56-60.

Miller et al. looked at differences in counting between Chinese- and English-speaking children. The predominant difference between the two groups was in their counting ability. The base-10 structure of number names is more obvious in Chinese, affording its speakers more ease in learning to count. The authors conclude that although the differences between the two groups were slight at a young age, the early difficulties in comprehending counting may cause problems at a later age because of mathematics' dependence on manipulation of symbols.

b. Cognitive

There are many cognitive achievements a young child must make in order to learn a spoken language. These achievements correspond, roughly, to the major areas of linguistic study. Children must understand the nature of communication (pragmatics), the sounds of language (phonology), the meaning of words used in language (semantics) and how to structure utterances (syntax). Due to developmental constraints, children can often perceive distinctions in spoken language before they are capable of producing such distinctions themselves. Thus, each entry below distinguishes between production and comprehension.

Prelinguistic Communication

Children learn a lot about how communication works prior to their ability to understand or use a particular written language. This exposure lays the foundation for children's later development of the formal aspects of spoken language (syntax, semantics, phonology, etc.).

Production

Bruner, J. (1975).   "The ontogenesis of speech acts." Journal of Child Language, 2, 1-19.

Bruner examined parental interactions with their prelinguistic children. His major finding was that parents communicate with their children by treating the child's actions as initiations of communication. Parents often respond verbally to their child's behavior (e.g., saying "Good girl!" when they take a bite, or "What's wrong" when they take a bite). This observation led Bruner to conclude that the acquisition of linguistic concepts begins with the child's attention to his or her actions as communicative gestures.

Harding, C. (1982).   "Development of the intention to communicate." Human Development, 25, 140-151.

Harding's work on communication in prelinguistic infants was some of the first to set principles for determining when the behavior of the infant was intended to communicate. Three criteria for intentional behavior are discussed: 1) waiting, 2) persistence, and 3) development of alternative plans. Harding concludes that the ability to communicate intentions is an ability that develops gradually, but is operative before the child begins to speak (from approximately 8 months of age).

Bates, E., Camaioni, L., & Volterra, V. (1975).  "The acquisition of performatives prior to speech. Merrill-Palmer Quarterly, 21, 205-226.

Before children are capable of using spoken language they communicate with gestures. But certain gestures are more intentional than others. There are criteria for counting a nonspeech act as communicative. In this study, Bates and colleagues focused on two types of communicative acts: assertions and requests. Both types of communicative acts are present in children prior to verbal communication. Young children often use an object to get their parent's attention (assertion) or use an adult to help them obtain an object (request). In studying the development of these actions over time, they concluded that children begin to understand how intentionality works, and then begin to use their gestures to communicate with others.

Comprehension

DeCasper, A., & Fifer, W. (1980).   "Of human bonding: Newborns prefer their mothers' voices." Science, 208, 1174-1176.

This study by DeCasper and Fifer was the first to show that very young newborns could differentiate their mother's voice from other female voices. This was observed by measurement of the infants' non-nutritive sucking (which increases when the infant is experiencing something novel). The researchers conclude that infant's ability to distinguish their mother's voice indicates a high degree of attentiveness on the part of the infant, and may reflect a form of important bonding between the infant and mother.

DeCasper, A., & Spence, M. (1986).  "Prenatal maternal speech influences newborns' perception of speech sounds." Infant Behavior and Development, 9, 133-150.

DeCasper and Spence tested the child's ability to attend to their mothers' speech in their prenatal environment. They did this by having mothers recite a passage at regular intervals throughout their pregnancy. After the child was born they compared the infant's response to the rehearsed versus novel passages. Infants showed a strong preference for the passage that had been recited during pregnancy, suggesting the ability to attend to prosodic features of speech even before they are born.

Robin, C. & Aslin, R. (1989).  "The language environment of the young infant: Implications for early perceptual development." Canadian Journal of Experimental Psychology, 43, 247-265.

In this article Robin and Aslin review recent work on preverbal infants' use of auditory information during language processing. They conclude that there is considerable evidence indicating that infants are actively attuned to sounds in their environment, particularly those that have the features of their native language. The ability to be attuned to their own language at such an early age leads the authors to speculate that the infants' prenatal experiences of their mother's voice may increase the salience of utterances in their native language.

Phonology

Children go through stages of vocalization in an effort to learn the sound systems of their language. This requires that the child be capable of speech perception – distinguishing the sounds of the language around them, and further the motor abilities to produce sounds themselves.

Production

Oller, D. (1980).  "The emergence of the sounds of speech in infancy." In G. Yeni-Komshian, J. Kavanaugh, & C. Ferguson (Eds) Child Phonology: Vol. 1, Production . New York: Academic Press (pp. 93-112).

In this article, Oller presents a description of the stages of vocalization that young children go through in their first year of life. At birth, crying is the only vocalization infants can make. By the age of two months babies begin cooing. Babbling typically begins around 6 to 7 months of age. In the first stage of babbling – called reduplicated babbling – children repeat the same consonant-vowel segments over and over (e.g., gagagaga). Variegated babbling – which involves several different syllables concatenated together – begins around the first year (e.g., dabugigo). Soon after children begin to utter their first words.

Reich, P. (1986).   Language Development. Englewood Cliffs, NJ: Prentice Hall.

In this book, Reich discusses a stage of language development in which children are situated between babbling and uttering their first words. Sometime around the first year of life many children begin to use idiomorphs – invented words or sounds to stand for objects. For example, a young child might call their bottle "gagu". These personalized words represent children's first attempt at communicating meaning.

Comprehension

Eimas, P., Siqueland, E., Jusczyk, P., & Vigorito, J. (1971).   "Speech perception in infants." Science, 171, 303-306.

Eimas et al. showed that infants directly after birth have the ability to make very fine distinctions along acoustical dimensions. For example, they are able to detect differences in voice onset time for many phonemes that adults cannot discriminate between. However, by the end of 4 months there is serious degeneration in this ability. This research, and subsequent findings in this area, has been used as an example of a universal readiness for language, which over time, is structured to accord with the child's particular linguistic environment.

Lasky, R., Syrdal-Lasky, A., & Klein, R. (1975).  "VOT discrimination by four to six and a half month old infants from Spanish environments." Journal of Experimental Child Psychology, 20, 215-225.

The original study by Eimas et al. was conducted with English-speaking children. Here, Lasky and colleagues replicate the decreased sensitivity to acoustical variation over time in Spanish-speaking infants.

Streeter, L. (1976).   "Language perception of two-month-old infants shows effects of both innate mechanisms and experience." Nature, 259, 39-41.

The ability of infants to perceive phonemic distinctions that adults are incapable of has raised questions as to whether this ability is innate, or formed through experience. To investigate this question, Streeter examined discrimination abilities of two-month-old infants in Kikuyu-speaking environments (Kikuyu is spoken in Kenya). Streeter concluded that some distinctions – such as that between voiced and voiceless sounds is the product of an innate mechanism. However, other discriminations that children are capable of at this young age come from exposure to a particular language.

Mandel, D., Jusczyk, P., & Pisoni, D. (1995).  "Infants' recognition of the sound patterns of their own names." Psychological Science, 6, 314-317.

Mandel and colleagues focused not on children's innate abilities for acoustic differentiation, but on their ability to learn to recognize their own names. Their results show that – by 4.5 months – children are capable of distinguishing their own names from other names. The researchers conclude that children are attuned to the sounds in their own linguistic environments from an early age.

Saffran, J., Aslin, R., & Newport, E. (1996).   "Statistical learning by 8-month-old infants." Science, 274, 1926-1928.

Saffran et al. also focused on what associations children can learn to make through exposure to certain language sounds. By presenting children with a variety of speech sounds, the researchers showed that 8-month-old children were able to segment a seemingly fluid stream of speech into word-like segments through attention to the statistical properties of neighboring speech sounds. This phenomenon occurred after only 2 minutes of exposure to the speech stream, indicating that even very young children have at their disposal sophisticated mechanisms for parsing linguistic sounds in their surrounding environment.

Marcus, J., Vijayan, S., Rao, S., & Vishton, P. (1999).   "Rule learning by seven-month-old infants." Science, 283, 77-80.

In response to Saffran et al.'s findings that children can attend to the statistical properties of speech in their environments, Marcus and colleagues conducted a study to show that children were also sensitive to, and using, rules to parse speech streams into words. They designed speech-segmenting tasks which could not be solved by merely attending to statistical regularities or counting. 7-month-old infants were capable of recognizing correct segmentation of speech in these cases. Thus, the researchers concluded that young children can extract rules from their environment, not only statistical regularities.

Nazzi, T., Bertoncini, J., & Mehler, J. (1998).   "Language discrimination by newborns: Toward an understanding of the role of rhythm." Journal of Experimental Psychology: Human Perception and Performance, 24, 756-766.

Nazzi et al. conducted a study showing that French infants can distinguish between some foreign languages. The infants were able to distinguish between languages that differ in prosodic features (e.g., Japanese and English) but had more trouble distinguishing languages with similar timing (e.g., Spanish and Italian). The researchers conclude that children parse and classify linguistic sounds largely on the basis of rhythmic information.

Mattock, K., & Burnham, D., (2006).   "Chinese and English infants' tone perception: Evidence for perceptual reorganization." Infancy, 10, 241-265.

Mattock & Burhnam examined infants' ability to distinguish various speech and nonspeech tones. Tonal discrimination is important, they reason, because over half the world's population speaks a tonal language. Their results indicate that children from different language backgrounds are all capable of tonal discriminations at 6-months, but that by 9 months there is a difference in discriminatory abilities between children in tonal language communities (e.g., Chinese) and non-tonal language communities (e.g., English). The researchers conclude that the child's ability to perceive tones is, at least to some degree, reorganized as a product of his or her language environment.

Lexical/Semantic

Children typically utter their first word near the end of their first year of life. The process of acquiring words accelerates gradually thereafter, until around the age of two when their utterances begin to include multiple words and their vocabulary expands dramatically. Even more quickly than they learn to utter words themselves, children are learning to understand the words and phrases of those around them.

Production

Carey, S. (1978).   "The child as word learner." In M. Halle, J. Bresnan, & G. Miller
(Eds), Linguistic Theory and Psychological Reality, Cambridge: MA: MIT Press (pp. 264-293).

Young children show remarkable ease at acquiring new words. In this article, Carey terms this ability "fast mapping" in reference to the fact that young children, around the age of 2 or 3, develop the ability to grasp the meaning of a new word with minimal exposure and without explicit training. She estimates that by the age of 6 the average English-speaking child knows 14,000 words.

Bloom, P. (2000).   "Fast mapping and the course of word learning." In How Children Learn the Meanings of Words (chapter 2). Cambridge, MA: MIT Press.

Some researchers have attributed the child's emergent ability to engage in fast mapping to the existence of a "word spurt" – a period of approximately two years in which children dramatically accelerate in their rate of word learning. Bloom disputes the idea of a word spurt. He agrees that children engage in fast mapping, but states that this is an ability that is maintained across the lifespan. If comprehension and production are both taken into account, he argues, then the increase in rate of word acquisition looks much more gradual.

Nelson, K. (1973).   "Structure and strategy in learning to talk." Monographs of the Society for Research in Child Development, 38.

Nelson investigated which words children acquire first through a longitudinal study with 1-2 year olds. There is an assumption that most early word learning is dedicated to the names of objects. He found that, for most children the first few words they learn are names for objects that are salient to them. However, children learn and use words from many grammatical categories from an early age. The first 50 words of each child contained words from several grammatical and concept domains.

Rodgon, M. (1976).   Single-word Usage, Cognitive Development, and the Beginnings of Combinatorial Speech. Cambridge: Cambridge University Press.

Children's linguistic communication begins with the utterance of single words. Young children often use these words to convey more than is contained in the word itself. A single word that is used to express a larger thought is called a holophrase. For example, an utterance of "dog" by a young child might refer to the child's awareness that there is a dog across the street. Rodgon coined this term and examined the use of holophrases by young children. Rodgon discovered that holophrases are used to stand for complete assertions and that they can be used to express a variety of semantic relations.

Subrahmanyam, K., & Chen, H., (2006).   "A crosslinguistic study of children's noun learning: The case of object and substance words." First Language, 26, 141-160.

The researchers compared children's learning of object and substance names across English- and Mandarin-speaking preschoolers. The English language makes a distinction between count (objects – e.g., dogs) and mass (substances – e.g., water) nouns, whereas Mandarin does not. Subrahmanyam and Chen were interested in whether differences in language impact the features that children attend to when naming novel items. They found that all children, both English- and Mandarin-speaking, attended to shape when naming novel items at the age of 3, regardless of whether the item was an object or substance. As the children grew older, Mandarin-speakers focused on the item's material, whereas English-speakers continued to focus on the item's shape.

Hirsh-Pasek, K., & Golinkoff, R. (2006).   Action Meets Word: How Children Learn Verbs. Oxford: Oxford University Press.

In this book, Hirsh-Pasek and Golinkoff explore how children learn the meanings of verbs. Much of early language acquisition has focused on children's acquisition of names for objects and other nouns; this book represents a growing interest in children's acquisition of words from other grammatical categories. The book includes entries from a variety of researchers and investigates many interesting topics, including verb acquisition across different languages and the role that understanding intentionality plays in learning words for actions.

Comprehension

Rosch, E., Mervis, C., Gray, W., Johnson, D., & Boyes-Braem, P. (1976).   "Basic objects in natural categories." Cognitive Psychology, 8, 382-439.
Parents give children basic level category words

Rosch et al. have shown that most parental utterances to young children include basic category names. Basic category names are those that carry the most information and are most easily differentiated from one another. The researchers present evidence suggesting that these are the easiest objects for formulating visual images, which may explain why words from this category are named earliest by young children.

Markman, E., & Hutchison, J. (1984).  "Children's sensitivity to constraints on word meaning: Taxonomic versus thematic relations." Cognitive Psychology, 16, 1-27.

When a child hears a word, how does he or she decide which of the possible things in the environment the word refers to? Further, how does the child learn the precise meaning of the word? Markman and Hutchison investigated these questions through studying naming abilities in two- to five-year-olds. They found that children have a tendency to assume that the words they are given are names for categories, not individuals. For example, children assume (correctly, in this case) that the word dog picks out all four-legged furry animals that bark. The researchers conclude that this, and likely other constraints, help children to select the meanings of novel words to which they are exposed.

Markman, E., & Wachtel, G. (1988).   "Children's use of mutual exclusivity to constrain the meanings of words." Cognitive Psychology, 20, 121-157.

Markman and Wachtel discovered that children constrain the possible meanings of a new word, in part, through use of a mutual exclusivity bias. That is, young children assume that there is only one name for any given object. Thus, new words are automatically assumed to refer to something for which the child does not already have a name. If the entire object is new to the child, the child will assume the new word refers to the whole object. If the object already has a name the child will assume that the new word refers to a part of the object. The researchers conclude that the mutual exclusivity bias motivates children to learn words for things besides objects, such as attributes, substances, and parts.

Syntactic

Children's utterances increase in length and sophistication over time as they acquire the rules of the language around them. Children not only need to learn the vocabulary of their spoken language, but also the various grammatical classes of words and how to structure utterances. As the research described here indicates, children develop these abilities quite naturally, despite the complexity of many grammatical rules.

Production

Brown, R. (1973).  A First Language: The Early Stages. Cambridge, MA: Harvard University Press.

Brown investigated the early stages of language acquisition in young children through a longitudinal analysis of three children. In studying their utterances, Brown developed a metric for assessing the syntactic sophistication of child language. Mere word counts are insufficient, as certain words (plurals, tensed verb forms, etc.) display more sophistication than others. Brown's metric is known as mean length of utterance, or MLU. This is used to count the average number of morphemes (meaningful sound units) in a child's utterance. Further research on children's MLU indicates that this number increases steadily from one to four years of age as children develop the ability to use the syntax of their native language.

Scarbrough, H. (1990).   "Index of productive syntax." Applied Psycholinguistics, 11, 1-22.

Brown's MLU measurements work for measuring syntactic complexity in children up until the age of four, but the metric is less useful for assessing children's utterances after this age. In response to this limitation, Scarbrough developed a method of assessing syntactic development in the later preschool years. This measure is called the index of productive syntax (IPSyn). It counts the emergence of new forms of noun and verb phrases and sentence structures, including questions and negation. IPSyn scores of preschool children have been shown to be a reliable predictor of later reading ability.

Berko, J. (1958).   "The child's learning of English morphology." Word, 14, 150-177.

Berko tested young children's ability to generalize their knowledge of English morphology to novel words. In so doing, she was looking to see whether children had acquired morphological rules, or were simply memorizing the morphological forms of particular words. 4- to 7-year-olds were shown a novel object and told "This is a Wug". The experimenter would then get a second object of the same kind and say Now there is another one. There are two of them. There are two _____. If children could complete the sentence with the correct plural then it would indicate that they had indeed learned a morphological rule that they could apply to novel situations. This is what Berko found. She thus concluded that children have a rule system for morphology.

Braine, M. (1976).  "Children's first word combinations." Monographs of the Society for Research in Child Development , 41, 104.

In this study Braine examined the syntactic patterns in two-word utterances of children learning English, Samoan, Finnish, Hebrew or Swedish. This observation reveals that, across languages, each child had learned a variety of different syntactic patterns that allowed them different ways of mapping meaning onto the surface structure of their utterances. Braine concludes that children's early multi-word utterances are deceptively simple, and actually reveal a great deal of syntactic sophistication.

Bowerman, M. (1973).   "Structural relationship in children's utterances: Syntactic or semantic?" In T. Moore (Ed) Cognitive Development and the Acquisition of Language, New York: Academic Press (pp. 197-213).

Bowerman coined the term "semantic bootstrapping" for the process by which children learn syntactic categories. On this view, children use their knowledge of semantic relations as a way of determining syntactic ones.

Klima, E., & Belugi, U. (1966).   "Syntactic regularities in the speech of children." In J. Lyons & R. Wales (Eds) Psycholinguistics Papers, Edinburgh: Edinburgh University Press (pp. 183-208).

Klima and Belugi investigated the emergence of negation and question-formation in young children's utterances. Their results indicate that both of these abilities emerge in stages. In the case of negation, children begin by appending the negative to the front end of sentences (e.g., No want to go), and only later begin to integrate the negative into the structure of the sentence (e.g., I do not want to go). In the case of questions, children begin distinguishing questions from assertions merely by intonation (e.g.,The dog likes milk?). At a later stage they learn to invert the subject and auxiliary verb (e.g., Does the dog like milk?).

Limber, J. (1973).   "The genesis of complex sentences." In T. Moore (Ed) Cognitive Development and the Acquisition of Language. New York: Academic Press (pp. 169-185).

In this paper, Limber describes the child's acquisition of relative clauses. A child's first relative clauses are typically object relatives. For example, you did what I just did, is an object relative clause. It is not until later that children begin to utter subject relative clauses, such as the dog who was running is tired. Limber speculates that difficulty with subject relative clauses may be a reflection of processing limitations.

Comprehension

Golinkoff, R., Hirsh-Pasek, K., Cauley, K., & Gordon, L. (1987).   "The eyes have it: Lexical and syntactic comprehension in a new paradigm." Journal of Child Language, 14, 23-45.

Golinkoff and colleagues introduced a new method for assessing language comprehension, and used it to show that long before they are capable of producing complex utterances, children can understand such utterances. 2-year-olds were shown two side-by-side video screens depicting similar actions, only one of which was actually being described in a tape-recorded utterance. The researchers measured which of the two screens the young child watched directly following the utterance. Visual fixation indicated that children were able to understand fairly sophisticated utterances from a young age, thus re-emphasizing the need to distinguish between production and comprehension in assessment of young children's linguistic abilities.

Shatz, M. (1978).   "On the development of communicative understandings: An early strategy for interpreting and responding to messages." Cognitive Psychology, 10, 271-301.

Shatz's work with young children and language acquisition shows that children bootstrap their way into communicative comprehension. Children follow simple heuristic processes to interpret the language around them – they use their understanding of individual words to decipher the meaning of sentences. Further, by giving young children ambiguous sentences in a variety of contexts, Shatz was able to how children utilize information in the environment as a way of deciphering meaning. He thus concludes that children, from a young age, are capable of monitoring their context as a way understanding those around them.

Bever, T. (1970).   "The cognitive basis for linguistic structures." In J. Hayes (Ed) Cognition and the Development of Language, New York: Wiley (pp. 279-362).

Bever studied language comprehension in young children. He discovered that they develop certain heuristics for interpreting sentences, which actually causes them to go through a period of difficulty interpreting passive sentences. Young children, age 2 to 3, perform above chance when interpreting passives. But around 4 or 5, children develop a rule for decoding utterances. They typically interpret noun plus verb plus noun as agent plus action plus object. This method of interpretation works well for active sentences, but not passive ones. Thus 4 to 5 year olds do worse interpreting passives than younger children. Bever concludes that this sort of overregularization is a direct consequence of children's attunement to the rules and patters of the language around them, and is a necessary step in the acquisition of spoken language.

Discourse/Pragmatics

In learning the various aspects of language (phonology, syntax, vocabulary), the goal is for children to use that language to communicate with others. Thus it is of interest to ask about the nature of children's conversational skills and how they evolve throughout their development of language across the preschool years.

Production

Shatz, M. & Gelman, R. (1973).   "The development of communication skills: Modifications in the speech of young children as a function of listener." Monographs of the Society for Research in Child Development, 38 (Serial No. 152).

One way in which children can display sensitivity to the requirements of communication is by modifying their speech across environments. In this set of studies, Shatz and Gleman show that even four-year-olds can make modifications to their speech under certain conditions. As their results show, four-year-olds slow their speech down and use shorter utterances when speaking to younger children than they do when speaking to children their age or adults. These findings indicate that children are able to take some factors regarding the linguistic position of their interlocuters into account from an early age.

Peterson, C. & Dodsworth, P. (1991).   "A longitudinal analysis of children's cohesion and noun specification and narratives." Journal of Child Language, 18, 397-415.

Peterson and Dodsworth examined children's ability to tell stories about their persona experiences across the two- to four-year-old age range. Perhaps unexpectedly, children's narratives became more structured and comprehensible with advances in age and mean length of utterance. Interestingly, throughout the age range studied children had particular difficulty introducing novel nouns and often used nouns ambiguously.

Pratt, M., & McKenzie-Keating, S. (1985).   "Organizing stories: Effects of development and task difficulty on referential cohesion in narrative." Developmental Psychology, 21, 350-356.

Pratt and McKenzie-Keating studied children's ability to include correct and unambiguous references in their tellings of stories. Through examination of four- and six-year-olds they found that referential error rates decreased with age, but that rates of referential ambiguity remained constant. These results suggest that even as children's story telling abilities improve they have difficulty evaluating the language they use to convey information.

Comprehension

Snow, C. (1977).   "The role of social interaction in language acquisition." In J. Berko Gleason & N. Bernstein Ratner (Eds) Psycholinguistics, 2nd edition, Fort Worth: Harcourt (pp. 453-481).

Snow's research on parent-child interactions when children are very young indicates that parent behavior may help children develop an understanding of the nature and rules of communication. Particularly, Snow found that parents often treat children's unintentional (and later, intentional) noises as turns in a conversation. She speculates that these interactions may help children understand language as a social activity that facilitates communication.

Bloom, L., Rocissano, L., & Hood, L. (1976).   "Adult-child discourse: Developmental interactions between information processing and linguistic knowledge." Cognitive Psychology, 8, 521-552.

Bloom et al. analyzed adult-child interactions, assessing them in terms of their content and contextual relations to investigate how children use information from adult input to form their own responses. They found that, from 21 to 36 months of age, children had consistently more speech that was adjacent to parent speech (directly preceded by) than nonadjacent. Further, the number of utterances that were contingent upon the content of parent speech increased over time, although it was more often after questions than non-questions.

c. Bilingualism / second-language acquisition

How children acquire language has long been a central area of study in cognitive science. Only more recently has the field undertaken the serious study of the acquisition of a second language. Many important questions are raised by these investigations. Is there a transfer of knowledge and skill from one language to another? Does it matter when you learn a second language? Do bilinguals have trouble keeping the two languages separate? Are both languages processed in the same way?

Kroll, J., & De Groot, A. (2005).   Handbook of Bilingualism: Psycholinguistic Approaches. Oxford: Oxford University Press.

Kroll and De Groot's handbook provides a comprehensive assessment of the current issues and findings regarding the study of bilingualism. The book focuses on the psycholinguistic approach, but also incorporates findings from cognitive neuroscience and computational modeling.

Bhatia, T. & Ritchie, W. (1999).   "The bilingual child: Some issues and perspectives." In W. Ritchie & T. Bhatia (eds) Handbook of Child Language Acquisition, San Diego, CA: Academic Press (pp. 569-643).

In this review of child bilingualism, Bhatia and Ritchie introduce the distinction between sequential and simultaneous bilingualism. Sequential bilingualism occurs when a child learns one language prior to the other; in simultaneous language acquisition both languages are learned concurrently. The child's form of bilingualism has an influence on his or her proficiency in the second language.

De Houwer, A. (1995).   "Bilingual language acquisition." In P. Fletcher & B. MacWhinney (eds) Handbook of Child Language, Cambridge: Blackwell (pp. 219-250).

De Houwer addresses the issue of the developmental timecourse of language acquisition in monolingual versus bilingual children. He concludes that, when the bilingualism is simultaneous, the developmental sequence is very similar for the two types of language learners. There is a small time delay in bilingual children's acquisition of certain syntactical and grammatical abilities, but this distance is overcome by the time the child begins school.

Meisel, J. (1993).  "Simultaneous first language acquisition: A window on early grammatical development." D.E.L.T.A., 9, 353-385.

Meisel investigated children's ability to acquire the grammar of two languages in cases of simultaneous bilingualism. She concludes that simultaneous bilinguals progress through the same stages of grammatical development as monolingual children, and further, that their processes of acquiring the grammar of each language follow the same underlying logic.

Genesee, F. (1989).   "Early bilingual development: One language or two?" Journal of Child Language, 16, 161-179.

There is a common conception that children who learn two languages from an early age are unable to differentiate between them. Genesee challenges this idea, arguing that bilingual children are capable of differentiating between their two languages from an early age. Bilingual children are able to attend to contextual differences between when one language is being used versus another. Further, Genesee notes that much of the language-mixing observed in young children might actually be a product of mixed utterances by parents.

Williams, L. (1980).   "Phonetic variation as a function of second-language learning." In G. Yeni-Komshian, J. Kavanagh, & C. Ferguson (eds) Child Phonology: Volume 2: Perception, New York: Academic Press (pp. 185-215).

Sequential bilingualism has different cognitive consequences than simultaneous bilingualism for a child acquiring language. Williams notes that children who learn a second language after they have acquired much of their first language, typically by the age of 3 or 4, process the second language in terms of their knowledge of the first. This can make learning a second language more difficult when there is extreme phonetic variation between the two languages.

Mechelli, A., Crinion, J., Noppeney, U., O'Doherty, J., Ashburner, J., Frackowiak, R., & Price, C. (2004).   "Neurolinguistics: Structural plasticity in the bilingual brain." Nature, 431, 757.

Mechelli et al. used voxel-based morphometry to evaluate the degree of neural plasticity in bilingual speakers of English and Italian. Their results indicate that, even when a second language is learned later in life (post childhood), there is still possibility for changes in the structural density of the language processing areas of the brain to accommodate that knowledge. Learning a second language was shown to increase grey matter density in the left inferior parietal cortex. Further, there was a strong positive correlation between proficiency in the second language and increase in grey matter density. The authors conclude that bilingualism affects not only functional, but structural properties of the brain, regardless of age of acquisition.

Chee, M., Tan, E., & Thiel, T. (1999).   "Mandarin and English single word processing studied with functional mangetic resonance imaging." Journal of Neuroscience, 19, 3050-3056.

Chee et al. studied the cortical organization of Mandarin-English bilinguals to determine the degree to which age of acquisition impacts the brain areas involved in processing a second language. They found no significant differences between the cortical areas involved in single-word reading in Mandarin and English, regardless of whether the second language was acquired before or after age 12. There were differences in the activation patterns for processing in the two languages, but these were consistent across subjects as well.

Bialystok, E., & Hakuta, K. (1999).   "Confounded age: Linguistics and cognitive factors in age differences for second language acquisition." In D. Birdsong (Ed) Second Language Acquisition and the Critical Period Hypothesis, Mahwah, NJ: Lawrence Erlbaum (pp. 161-181).

Bialystok & Hakuta evaluate research on second language learning in light of Lenneberg's critical period hypothesis. They cite evidence that supports difficulty in acquiring some aspects of language (e.g., phonology) when the second language is learned after puberty, but conclude that, overall, proficient acquisition of a second language is possible later in life. They suggest a few alternative explanations of the advantage young learners have in acquiring a second language.

Klein, D., Miner, B., Zatorre, R., Zhao, V., & Nikelski, J. (1999).   "Cerebral organization in bilinguals: A PET study of Chinese-English verb generation." Neuroreport, 10, 2841-2846.

The researchers used PET to study the cortical organization of language processing in Chinese speakers who learned English later in life. They found no significant differences in patterns of activation across the two languages. They concluded that even with languages as disparate as Mandarin and English, bilinguals have a single, dedicated neural substrate for processing language.

Wartenburger, I., Heekeren, H., Abutalebi, J., Cappa, S., Villringer, A., & Perani, D. (2003).   "Early setting of grammatical processing in the bilingual brain." Neuron, 37, 159-170.

The authors present the first fMRI study demonstrating that the age of L2 acquisition is crucial for grammatical processing. Grammatical and semantic judgments in three groups of Italian-German bilinguals were addressed with fMRI. Only in the case of L2 acquired very early in life did the neural substrates for L1 and L2 grammatical processing overlap. This suggests that grammatical processing, given its dependence on age of acquisition, is based on competence that is neurologically 'wired-in.' In addition, proficiency is the main determinant of the cerebral organization of both grammar and semantics in late bilinguals.

d. Spoken language disorders and the biological basis of language

Often, we can learn a great deal about how something works by studying the various cases in which it doesn't work or breaks down. This is particularly true in the case of language. The discovery of various forms of aphasia (speech disorder) has allowed scientists to uncover information about the biological basis of language. Further disorders of language acquisition also provide insight into the mechanisms by which language skills develop in normal populations.

Bishop, D. (1997).  Uncommon Understanding: Development and Disorders of Language Comprehension in Children. East Sussex, UK: Psychology Press.

In this book Bishop provides a comprehensive assessment of the various disorders that occur in spoken language acquisition, with a particular focus on SLI: specific language impairment. He integrates research from neuropsychology, psycholinguistics and developmental psychology.

Musso, M., Moro, A., Glauche, V., Rijntjes, M., Reichenbach, J., Buechel, C., & Weiller, C. (2003).   "Broca's area and the language instinct." Nature Neuroscience, 6, 774-781.

fMRI study providing insight into the neural underpinnings of universal grammar. German-speaking adults were taught first a small vocabulary in a foreign language, and subsequently UG-consistent rules and UG-inconsistent manipulations of that foreign language. During scanning, subjects had to judge whether sentences belonged to the consistent or inconsistent rules. Activation with the pars triangularis of Broca's area was specific only for UG-consistent rule learning leading the authors to hypothesize that Broca's area is the neural counterpart of UG. In addition, they found that hierarchy within grammatical structures can modulate activity in Broca's area across a wide-range of typological variations, such as those found between Japanese and Italian.

Heiss, W., Thiel, A., Kessler, J., & Herholz, K. (2003).   "Disturbance and recovery of language function: Correlates in PET activation studies." Neuroimage, 20, S42-S49.

Heiss et al. compared PET activation of subjects who had suffered from language loss due to some sort of neurological dysfunction both immediately and years after the onset of the damage. Their results indicate that there is some plasticity for language processing in the brain that allows people to recover language abilities post-aphasia. Years after damage to primary language areas, subjects show activation in secondary language areas that are not usually engaged by normals during basic language processing tasks. These results show that although there are areas of the brain dedication to language processing, it is possible for at least some language processing to occur elsewhere.

Geschwind, N. (1972).   "Language and the brain." Scientific American, 226, 76-83.

Geschwind offers a model of language processing – developed in response to evidence of various types of language disorders – that divides the language system into three distinct subsystems or modules. First, there is the processing of auditory information in Wernicke's area. Second, that information is carried along the arcuate fasciculus. Finally, the information reaches Broca's area, which assembles the articulatory programs for producing speech. Geshwind's paper is important because it was one of the first attempts to systematize the relationship between brain function and language processing, and it did so in a way that allowed for distinct behavioral predictions about types of brain lesions.

Goodglass, H. (1993).   Understanding Aphasia. San Diego: Academic Press.

Goodglass's book provide an overview of the current understanding of aphasic language impairments. The book details various types of aphasia, and the anatomical and processing impairments that typically accompany them. Both theoretical and clinical approaches are included.

Caramazza, A., & Zurif, E. (1976).  "Dissociation of algorithmic and heuristic processes in language comprehension: Evidence from aphasia." Brain and Language, 3, 572-582.

By studying aphasics, Caramazza & Zurif showed that there are two distinct processes of language comprehension. On tasks of sentence comprehension, Broca's and conduction aphasics could not process sentences whose meaning depended on syntactic differentiations, but performed at normal levels on interpreting sentences that required reliance on semantic information. Wernicke's aphasics showed the opposite pattern. The researchers concluded that there is a neuropsychological dissociation between these two ways of accessing sentence meaning.

Linebarger, M., Schwartz, M., & Saffran, E. (1983).   "Sensitivity to grammatical structure in so-called agrammatic aphasics." Cognition, 13, 361-392.

Broca's aphasia is typically associated with severe difficulties in processing in syntax. This study by Linebarger et al. suggests that Broca's aphasics may have greater knowledge of syntax than has previously been indicated. By asking subjects with Broca's aphasia to judge the grammaticality of various types of sentences, the researchers were able to show that Broca's aphasics do well on sentences with structural deformity, but do worse on those that require them to recognize inappropriate pronouns. They suggest that the earlier findings that Broca's patients are syntactically impaired may stem from their difficulties comprehending syntactic distinctions in real-time. When asked to judge the grammaticality of utterances, they may have more time to recognize and respond to judgments of at least some types syntactic errors.

Gazzaniga, M. & Sperry, R. (1967).   "Language after section of the cerebral commissures." Brain, 90, 131-148.

For patients who suffer from intractable seizures it is common medical practice to sever the forebrain commisssure and massa intermediary. Gazzaniga and Sperry examined the impact of this surgery on language abilities. The abilities for speech and writing appeared to be localized entirely in the left hemisphere, but language comprehension appeared to be represented in both the left and right hemisphere. They thus conclude that the amount of disconnection between the two hemispheres is not as stark as Geschwind reported.

2. Written language acquisition

a. Modelling reading

Dual Route

Marshall, J. & Newcombe, F. (1973).   "Patterns of paralexia: A psycholinguistic approach." Journal of Psycholinguistic Research, 2, 175-199.

Marshall and Newcombe's review of the literature on types of acquired dyslexia is what led to the dual-route model of word reading. Marshall and Newcombe identified two different types of dyslexia – surface and deep – that are associated with different patterns of psychological and neurological deficit.

Coltheart, M., Rastle, K., Perry, C., Langdon, R., & Ziegler, J. (2001).   "DRC: A dual route cascaded model of visual word recognition and reading aloud." Psychological Review, 108, 204-256.

Coltheart et al. present the Dual Route Cascaded model of visual word recognition. This model of word reading involves two different routes through which words can be processed. The first, nonlexical, route follows the rules of grapheme-phoneme relations, allowing a person to sound out and pronounce all regular words, as well as plausible pseudowords (e.g., MINT and BINT, respectively). Irregular words (like PINT) are named through use of the second, lexical route. Instead of breaking the word down into its phonemic components, this route functions as a memorized look-up table for words whose pronunciations are recalled as a whole. The researchers claim that this dual-route perspective provides the only model of word recognition that can account for human data across both lexical decision and reading aloud tasks.

Coltheart, M. (2005).   "Modeling reading: the dual route approach." In M. Snowling & C. Hulme (eds), The Science of Reading: A Handbook, Oxford: Blackwell (pp. 6-23).

In this review article of the dual route approach to modeling reading, Coltheart claims that – despite differences between connectionist and nonconnectionist approaches to modeling – there has emerged a general consensus as to the need for two distinct pathways/procedures/processes for reading print. He goes on two discuss the two pathways, how they are modeled in the dual route approach, and the differences between connectionist and nonconnectionist approaches.

Single Route

Seidenberg, M, & McClelland, J. (1989).   "A distributed, developmental model of word recognition." Psychological Review, 96, 523-568,

In this paper, Seidenberg and McClelland introduce a PDP (parallel-distributed-processing) model of word reading that challenges many of the assumptions of the classic, dual route approach to modeling reading. First, the model does not require two distinct routes for word reading. Instead, it uses one general learning procedure and acquires pronunciations through attending to statistical regularities in its inputs. Second, the model does not require the explicit representation of rules. Seidenberg and McClelland show that this model is able to simulate a great deal of the phenomena seen in human performance data. It also offers two distinct advantages. First, the model goes through a stage of learning which closely resembles the challenges young children face in learning to read. Second, damaging the model in various places leads to the types of graceful degradation in function that are often seen in lesion cases. From this the authors conclude that connectionist approaches represent a clear competitor to dual-route models.

Plaut, D. (1997).  "Structure and function in the lexical system: Insights from distributed models of word reading and lexical decision." Language and Cognitive Processes, 12, 765-805.

In this article, Plaut introduces connectionist modeling approaches to visual word recognition and differentiates them from dual route models. Dual route models require separate pathways for processing regular and exception words and requires representations of each of the words contained in the system. In contrast, connectionist models have a single processing route and allow for distributed representations. Plaut's models have the ability to account for much of the data on how people process words. Thus, connectionist models present a challenge to the traditional understanding of the functional structure required for visual word recognition.

Plaut, D. (2005).  "Connectionist approaches to reading." In M. Snowling & C. Hulme (eds), The Science of Reading: A Handbook, Oxford: Blackwell (pp. 24-38).

Plaut provides a survey of connectionist approaches to reading over the past 30-40 years. He begins by explaining the basic principles of connectionism, and highlights the similarity between connectionist principles and neural computation as the distinct advantage of this approach. Although current models still have limitations in the amount of performance data they can account for, Plaut offers reasons for thinking that connectionist approaches to language learning remain a strong competitor to dual route models.

Modeling semantic knowledge

Landauer, T., & Dumais, S. (1997).  "A solution to Plato's problem: The latent semantic analysis theory of acquisition, induction and representation of knowledge." Psychological Review, 104, 211-240.

Landauer and Dumais' paper begins from the puzzling question that people seem to know more information than what they've been exposed to directly. They attempt to account for this ability through a model of using Latent Semantic Analysis (LSA). LSA works on a basic mechanism of induction, representing similarity relations as a feature of co-occurrence. The researchers conclude that this is a plausible model of semantic development, as the model was capable of performing similarly to the TOEFL scores of foreign students.

Steyvers, M., & Tenenbaum, J. (2005).  "The large-scale structure of semantic networks: Statistical analyses and a model of semantic growth." Cognitive Science, 29, 41-78.

Steyvers and Tenenbaum take a different approach to modeling semantic development. They argue that semantic development as a small-world structure, where certain words – learned early – act as 'hubs' around which new words cluster. In this way, they conclude that the structure of one's semantic network will be a function of lexical development, the frequency of word use, and the age of acquisition.

b. Phonological Awareness and Reading

In order to learn how to read children have to understand that writing represents speech by symbolizing the sounds of spoken language. The ability to identify and distinguish the sounds used in one's own language – termed phonological awareness – is thus often considered to be important for successful written language development. Many researchers believe phonological awareness is the key skill that children must acquire to learn how to read. However, in recent years other researchers have begun to question the strength of the relation between phonological awareness and reading.

Ehri, L., Nunes, S., Willows, D., Schuster, B., Yaghoub-Zadeh, Z., & Shanahan, T. (2001).   "Phonemic awareness instruction helps children learn to read: Evidence from the National Reading Panel's meta-analysis." Reading Research Quarterly, 36, 250–287.

The National Reading Panel's meta-analysis of the relation between phonemic awareness and learning to read and spell was evaluated. The researchers state that the 52 studies involved reveal a significant influence of phonemic awareness on development of reading and spelling ability. Phonemic awareness was found to be strongly related to children's word recognition and reading comprehension skills.

McBride-Chang, C & Ho, C. (2000).   "Developmental issues in Chinese children's character acquisition." Journal of Educational Psychology,92, 50-55.

Researchers tested Chinese children on their phonological processing, speech perception, and recognition of Chinese characters. Phonological awareness was strongly related to children's ability to for character recognition. The researchers conclude that phonological awareness plays a central role in learning to read in Chinese, similar to the role in plays in reading development in Western societies.

Gottardo, A., Chiappe, P., Yan, B., Siegel, L., & Gu, Y. (2006).  "Relationships between first and second language phonological processing skills and reading in Chinese-English speakers living in English-speaking contexts." Educational Psychology, 26, 367-393.

The authors examined the relationship between a child's ability to categorize and process phonemes and their reading performance. Subjects were Chinese-English speaking children. The researchers found that, in both languages, the ability to process phonemes was related to reading performance. Further, they conclude that exposure to reading in Chinese influences language processing in English.

McBride-Chang, C., Cho, J., Liu, H., Wagner, R., Shu, H., Zhuo, A., Cheuk, C., & Muse, A. (2005).   "Changing models across cultures: Associations of phonological awareness and morphological structure awareness with vocabulary and word recognition in second graders from Beijing, Hong Kong, Korea, and the United States." Journal of Experimental Child Psychology, 92, 140-160.

Children from 4 different language communities were studied to determine relations amongst their abilities in phonological awareness, morphologiacl structure awareness, vocabulary and word reocnition. Particularly of interest here is their finding that phonological awareness is more strongly related to word recognition in English and Korean than in Chinese.

Laing, E., & Hulme, C. (1999).   "Phonological and semantic processes influence beginning readers' ability to learn to read words." Journal of Experimental Child Psychology, 73, 183–207.

When taught associations between short abbrevations and spoken words four- to seven-year-old children where shown to do better at recalling the associations when the relation between the abbreviation and spoken word was phonetic. The researchers conclude that, when relevant, children attend to phonological information in associating print and speech.

Castles, A., & Coltheart, M. (2004).  "Is there a causal link from phonological awareness to success in learning to read?" Cognition, 91, 77–111.

Castles and Coltheart examine previous research to determine whether or not claims of a causal relation between phonological awareness and reading ability are truly warranted. Previous work, they conclude, has not gone far enough to preclude the possibility that reading ability may have a causal influence on phonological ability, not the reverse. From this they conclude that, though related, the causal relation from phonological awareness to reading ability is not presently supported.

c. Developmental stages of reading and writing

In order to learn how to read, children must understand how written language works. That is, they must understand that writing is a representation of language and that there is a relation between the sounds of spoken language and the way print symbolizes language. Writing involves this understanding, as well as spelling ability and the refinement of motor skills.

Treiman, R., & Kessler, B. (2007).   "Learning to Read." In M. Gaskell (ed), Oxford Handbook of Psycholinguistics, Oxford: Oxford University Press (pp. 657-666).

In this chapter, Treiman and Kessler provide an overview of the cognitive achievements that children must make in order to learn to read. They discuss various theories of the stages involved in this developmental process. Additionally, they discuss the two predominant approaches to modeling reading development: dual-route and connectionist models.

Chi, M. (1988).  "Invented spelling/writing in Chinese-speaking children: The developmental patterns." National Reading Conference Yearbook, 37, 285-296.

Chi studied early spelling and writing development in Chinese-speaking children. Chi identified several stages of development in the children's writing. They were: scribbling, using the last name for every word, iconic pictographic writing, pictographic invented writing, invented writing with mild distinguishable quality, invented writing with distinguishable quality, writing with few absent/added strokes, writing by homonym character, writing by phonemic symbols, and, finally, correct writing.

Hanley, J. (2005).  "Learning to read in Chinese." In M. Snowling & C. Hulme (Eds) Science of Reading: A Handbook, Oxford: Blackwell (pp. 301-330).

Hanley's review article of the processes of learning to read in Chinese discusses the nature of the Chinese logographic writing system, how reading is taught in China, how children learn to read compound characters, the role of phonological awareness in learning to read Chinese, and the types of developmental dyslexia seen in Chinese readers.

Ehri, L. (1998).   "Grapheme-phoneme knowledge is essential for learning to read words in English." In J. Metsala & L. Ehri (eds) Word Recognition in Beginning Literacy, Mahwah, NJ: Erlbaum (pp. 3-40).

Ehri offers a distinction between two periods in the process of written language development: the pre-alphabetic and alphabetic periods. The pre-alphabetic period is the stage in which children do not understand the relation between spellings and sounds, and thus are unable to decode writing phoneme by phoneme. Their reading and writing reflects this inability, as they view each word as the most basic unit of print. For example, young children may treat their names as a basic unit, not recognizing that any one else's name that starts with the same sound as theirs may start with the same letter. During the alphabetic period, children learn to decode words at the level of individual phonemes.

Frith, U. (1985).   "Beneath the surface of developmental dyslexia." In K. Patterson, J. Marshall, & M. Coltheart (eds), Surface Dyslexia: Neuropsychological and Cognitive Studies of Phonological Reading , London: Erlbaum (pp. 316-335).

Frith argues for a two stage view of written language development that is very similar to Ehri's. Firth refers to the first stage as the logographic stage, because children in this period treat written language as if it were a logography and not an alphabet. The alphabetic period occurs when children understand that English is a language that depicts spoken language at the level of individual phonemes.

McBride-Chang, C., & Treiman, R. (2003).   "Hong Kong Chinese kindergartners learn to read English analytically." Psychological Science, 14, 138-143.

McBride-Chang and Treiman examined how Chinese-speaking children in Hong Kong learn to read English. They were taught English using a logographic "look and say" method. The researchers examined the children's ability to learn novel spelling based on one of three relations: letter name, letter sound, or visual cues. Studied over three years the children performed best in the letter name and letter sound conditions. McBride-Chang and Treiman thus concluded that letter-name and letter-sound knowledge are an important aspect of learning to read English, both as a first and second language.

Treiman, R., & Bourassa, D. (2000).   "The development of spelling skill." Topics in Language Disorders, 20, 1-18.

Treiman and Bourassa review the existing empirical literature on the development of spelling ability for children learning alphabetic languages. The first task of a young child learning to spell is the recognition that writing represents the sounds of spoken language. Once children know this, they can begin to learn the particular sound-spelling correspondences for their alphabet. Children require a great deal of exposure to written language before they learn all of the intricacies involved in transferring phonology into orthography and morphology. The authors conclude that research into children's spelling development must take all of the factors that children must learn into account when describing children's burgeoning spelling abilities.

Seymour, P. & Duncan, L. (2001).  "Learning to read in English." Psychology: The Journal of the Hellenic Psychological Society, 8, 281-299.

In this review article of learning to read in English, Seymour and Duncan defend a 'dual foundation model' for learning to read in alphabetic and syllabic languages. This model proposes a series of phases at each of which children become increasingly aware of the units of the language. In the initial phase of learning English, children learn the smallest units of the language. Next comes the orthographic phase in which children learn the spellings of syllables or rime units. At the final stage children begin to learn the spellings of entire words.

Ferreiro, E. & Teberosky, A. (1982).   Literacy Before Schooling. New York: Heinemann.

In this book Ferreiro and Teberosky claim that children who are learning alphabetic languages may go through a stage in which they treat their language as if it were a syllabary. That is, there writings reflect individual units corresponding to each syllable of the spoken word they are trying to write. This view has been considered somewhat controversial, as the researchers have not examined the extent to which the correspondence between children's spelling and syllables deviates from chance. Further, given that most of their research has been done in Spanish-speaking environments, they have not given ample consideration to the role of letter names (particular vowel names) in children's spellings. Since the number of vowels correlates well with the number of syllables, this could provide an alternative explanation of their findings.

d. Cross-linguistic Studies

Most research on reading and spelling development has been carried out in English-speaking settings. Given the differences across writing systems, it is helpful to examine this developmental process in non-English contexts. This cross-linguistic perspective helps us to see the universal difficulties that children face in acquiring written language skills, and also gives us an appreciation of the particular challenges children face in specific language communities. Further, many of the studies listed here explore the benefits or drawbacks children may have in learning to read and write in two language systems.

Treiman, R., & Kessler, B. (2005).   "Writing systems and spelling development." In M. Snowling & C. Hulme (Eds) Science of Reading: A Handbook, Oxford: Blackwell (pp. 120-134).

In this survey chapter Treiman and Kessler focus on the challenges children face in their development of spelling ability, and how these challenges are modified by the child's writing system. The authors distinguish various types of writing systems and the specific sound-to-spelling relations that children must learn in each. They conclude that although there are certain variations in what and how much children must learn about sound-spelling relations, all children face the central developmental challenge of learning that writing symbolizes spoken language.

Harris, M., & Hatano, G. (1999).   Learning to Read and Write: A Cross-linguistic Perspective. Cambridge: Cambridge University Press.

This anthology offers one of the first systematic discussion of reading and spelling development in young children that does not focus largely or entirely on English-speaking children. The book includes chapters investigating written language development in various languages, from Chinese to German, Hebrew to Greek. There are also articles on universal topics for written language acquisition, such as phonological awareness and spelling errors.

Bialystok, E., McBride-Chang, C., & Luk, G. (2005).   "Bilingualism, langauge proficiency, and learning to read in two writing systems." Journal of Educational Psychology, 97, 580-590.

In this study, the researchers compared five- and six-year-old children who were monolingual English, simultaneously bilingual Chinese-English, or sequentially bilingual Chinese-English. They were interested in differences in phonological awareness and ability to decode words across languages. They found that skill in phonological awareness was equally as beneficial for simultaneous and sequential bilinguals, but that there was no transfer of ability to decode words in one language to ability to decode words in the other, regardless of age of second language acquisition. They concluded that bilingualism has no general effect on learning to read.

Wang, M., Koda, K., & Perfetti, C. (2004).   "Language and writing systems are both important in learning to read: A reply to Yamada." Cognition, 93, 127-132.

The authors claim that there are no general conclusions to be drawn about the possibility of transfer of reading and spelling ability from one language to another. Any conclusions about the possibility of transfer will be relative to the two languages and writing systems involved.

Perfetti, C, & Liu, Y. (2005).   "Orthography to phonology and meaning: Comparisons across and within writing systems." Reading and Writing, 18, 193-210.

There are basic principles of written language that are taken to apply to all writing systems. First, there is the Universal Writing System Constraint, which states that all writing systems encode spoken language. Second, the Universal Phonological Principle says that skilled readers – across languages – activate word pronunciations while reading. Perfetti & Liu evaluated these principles in light of findings from research across many writing systems. They conclude that these general principles do hold true, but propose supplementing them with the System Accommodation Hypothesis, which states that the specific neural structures and cognitive processes that support understanding of written language vary across writing systems.

Liow, S., & Lau, L. (2006).  "The development of bilingual children's early spelling in English." Journal of Educational Psychology, 98, 868-878.

The researchers used an existing paradigm (Treiman et al. (1994) for testing children's spelling of flaps) to test the metalinguistic awareness of bilingual 6-year-olds learning English from three different language backgrounds. Liow & Lau found that difficulty with the flap task was influenced by the child's original language. They conclude that there is knowledge transfer of phonological information across languages in bilingual children, and suggest implications for education of bilingual children.

e. Biological Basis of Reading

Although reading is an evolutionarily newer ability, the importance of reading to members of literate societies has led to many brain areas being specialized for the processing of written language. It has long been known that language processing tends to occur in the left hemisphere of the brain. Here we summarize research that explores the specific parts of the left hemisphere involved in various aspects of language processing, and further research as to how and when these areas may become specialized. Also, information on the genetic basis of reading is included.

Turkeltaub, P., Gareau, L, Flowers, D., Zeffiro, T., Eden, G. (2003).   "Development of neural mechanisms for reading." Nature Neuroscience, 6, 767-773.

In this study, researchers used fMRI to investigate whether skilled reading involves activation of areas of cortex dedicated to this ability. Using data from subjects age 6-22, Turkletaub et al. found that learning to read was associated with decreased activation in right inferotemporal areas of the cortex, and an increase in activation in the left middle temporal area. Further, they found that a subject's ability to engage in phonological processing determined the amount of activation in the left posterior superior temporal region.

Turkeltaub, P., Eden, G., Jones, K., & Zeffiro, T. (2002).   "Meta-analysis of the functional neuroanatomy of single-word reading: Method and validation." Neuroimage, 16, 765-780.

The authors of the present study used a meta-analysis of PET data on visual word recognition to prescind individual differences in sensorimotor abilities and thereby isolate the areas of cortext involved in the processing of written language. Turkeltaub et al. used a new approach to meta-analysis – evaluating activation of subjects across studies through use of a localization probability distribution. The results were then compared to fMRI data on the same tasks. The results show consistent activation of in bilateral motor and superior temporal cortex, as well as pre-SMA, left fusiform gyrus, and the cerebellum during the process of single-word reading.

Cohen, L., & Dehaene, S. (2004).   "Specialization within the ventral stream: The case for the visual word form area." Neuroimage, 22, 466-476.

Cohen and Dehaene investigated the possibility that the visual ventral stream contains a specialized pathway for processing visual word information in literate human adults. The authors review experiments addressing this issue. They conclude that there is indeed specialization of the ventral stream for word processing. Reading activates a specific area of the left ventral occipitotemporal sulcus across proficient readers, and further, lesions to this area are consistently associated with forms of alexia. Further, the authors conclude that the process is specialized enough to attend to particular features of the reader's writing system (e.g., upper- and lower-case). They conclude that reading development in normal populations involves specialization of processing pathways and cortex to the task of recognizing visually-presented words.

Gaillard, W., Balsamo, L., Ibrahim, Z., Sachs, B., & Xu, B. (2003).   "fMRI identifies regional specialization of neural networks for reading in young children." Neurology, 60, 94-100.

Gaillard et al. used fMRI to identify the neural networks that are involved in the acquisition of reading skills in young children. The authors studied English-speaking children from 5 to 7 years old. They found significant activation in the left inferior temporal occipital junction, left fusiform gyrus, middle temporal gyrus, middle frontal gyrus, and supplementary motor areas. These areas of activation largely correspond to the areas of activation seen in literate adults, leading the authors to conclude that the neural networks for reading are strongly developed and left-lateralized by the age of 6 or 7.

McDermott, K., Petersen, S., Watson, J., & Ojemann, J. (2003).  "A procedure for identifying regions preferentially activated by attention to semantic and phonological relations using functional magnetic resonance imaging." Neuropsychologia, 41, 293-303.

McDermott et al. introduce an fMRI procedure for identifying the neural regions that are associated with attention to the semantic and phonological properties of written words. Subjects were asked to attend to the relations between words that were either semantically or phonologically related. The researchers found different patterns of activation across the two tasks. Further, they conclude the success of using this task to parse subtle processing differences intrasubjectively can be used to explore other areas of language processing, including language development in young children and processing in dyslexic and blind subjects.

Pennington, B., & Olson, R. (2005).   "The genetics of dyslexia." In M. Snowling and C. Hulme (eds), The Science of Reading: A Handbook (pp. 453-472).

The ability to read, at its base, depends on the brain to perceive and process the written information. This, in turn, is dependent upon genetic mechanisms. Reading itself is far too evolutionarily young to be an ability that is specifically coded for in the human genome. However, researchers are amassing a body of evidence suggesting various genetic differences that contribute to reading ability, including the development of dyslexia. In this article, Pennington and Olson review current research in this area and conclude that the genetics of reading are a fruitful new area of investigation for understanding the biological basis of reading.

Mechelli, A., Gorno-Tempini, M., & Price, C. (2003).   "Neuroimaging studies of word and pseudoword reading: Consistencies, inconsistencies, and limitations." Journal of Cognitive Neuroscience, 15, 260-271.

In this review article, Mechelli et al. survey the literature on fMRI studies of word and pseudoword reading. They conclude that, for the most part, the differences reported when comparing word to pseudoword reading are weak and inconsistent. The more robust comparisons are possible when pseudoword reading is compared to word reading.

Rayner, K., & Pollatsek, A. (2006).  "Eye-movement control in reading." In M. Traxler & M. Gernsbacher (eds), Handbook of Psycholinguistics, London: Elsevier (pp. 609-653).

Investigation of how a person's eyes move across the page while she is reading provide a useful source of information on how moment-to-moment reading processing occurs. For instance, by knowing how long a person's eyes fixate on a particular word we can draw inferences on the difficulty of processing that word. In this article Rayner and Pollatsek survey the vast literature available on eye-movement and reading.

f. Disorders of Reading and Spelling

There are many distinct types of social, neurological, and psychological impairments that can lead to problems in reading and spelling. Perhaps the most famous impairment of reading and spelling ability is dyslexia. Exploration of these various disabilities and the way in which they impact the cognitive systems dedicated to language can help to illuminate the processes by which normal written language development occurs.

Leybaert, J. (2005).   "Learning to read with a hearing impairment." In M. Snowling & C. Hulme (eds), The Science of Reading: A Handbook, Oxford: Blackwell (pp. 379-396).

In this survey article, Leybaert reviews the various types of hearing impairment that can cause difficulties for learning to read. Problems in hearing can lead to difficulty understanding the phonological patterns in written language. Leybaert discusses the types of treatment which improve reading ability for different hearing-impaired populations, the neurological basis of these impairments, and reading outcomes for patients diagnosed with various types of hearing disorders.

Sanders, R. & Caramazza, A. (1990).   "Operation of the phoneme-to-grapheme conversion mechanisms in a brain injured patient." Reading and Writing, 2, 61-82.

Sanders and Caramazza report the behavior of a patient who acquired a difficulty with spelling. His dysgraphia, as spelling disorders are called, were largely phonological in nature, suggesting that he was relying heavily on the sublexical route for processing words. This report is of interest because it shows that difficulties in spelling are not only developmental, but can be acquired as a result of brain damage or other similar trauma.

Castles, A. & Coltheart, M. (1993).   "Varieties of developmental dyslexia." Cognition, 47, 149-180.

In this article, Castles and Coltheart review various types of developmental dyslexia. They find that there are analogous varieties of dyslexia in both the developmental and acquired dyslexia populations. In particular, they examined the sublexical reading skills of boys with developmental dyslexia and compared them to nondyslexics. Their results show two distinct subtypes of dyslexia, which differ drastically from controls and one another. Thus the researchers conclude that dyslexica cannot be described in terms of a general language disorder.

Mody, M. (2004).  "Neurobiological correlates of language and reading impairments." In C. Stone, E. Silliman, B. Ehren, & K. Apel (eds). Handbook of Language and Literacy: Development and Disorders, New York: Guilford Press (pp. 49-72).

In this chapter Mody offers a review of the neurobiological factors that are related to language and reading impairments. Given the importance of phonological information to learning to read, many brain areas involved in auditory processing are highly correlated with reading difficulty. Mody discusses the various neurological methods used to assess reading ability (MEG, EEG, PET, fMRI, etc) and discusses the areas involved in different aspects of the reading process.

Shaywitz, S., Shaywitz, B., Fulbright, R., Skudlarski, P., Mencl, W., Constable, R., Pugh, K., Holahan, J., Marchione, K., Fletcher, J., Reid, L., & Gore, J. (2003).   "Neural systems for compensation and persistence: Young adult outcome of childhood reading disability." Biological Psychiatry, 54, 25-33.

In this study, Shaywitz et al. used fMRI to measure brain activation in reading tasks for adult developmental dyslexics. They found that the extent of their reading difficulty in later life was modulated by IQ. Higher IQ dyslexics show more activation of additional brain areas when reading, suggesting that they may bring additional resources (vocabulary, reasoning skills, etc) to bear on the task of deciphering new words.

Siok, W., Perfetti, C., Jin, Z., & Tan, L. (2004).   "Biological abnormality of impaired reading is constrained by culture." Nature, 431, 71-76.

Generally, developmental dyslexia is defined as presence of severe reading disabilities in persons who otherwise have normal education and intelligence. However, as Siok et al. show, the biological nature of the dyslexia varies widely across languages and cultures. In alphabetic languages (such as English) dyslexia is most often associated with dysfunction of the left temporoparietal regions of the brain. As their results show, different brain areas are involved for Chinese dyslexics. Using fMRI, the researchers investigated patterns of activation in Chinese-speakers, both dyslexics and controls. Impaired readers of Chinese show heightened activation of the left middle frontal gyrus. The authors speculate that this difference in activation (from that seen of English speakers) is due to the types of decoding required for a logographic writing system.

Ho, C., Chan, D., Chung, K., Lee, S., & Tsang, S. (2007).   "In search of subtypes of Chinese developmental dyslexia." Journal of Experimental Child Psychology, 97, 61-83.

The dual-route model of word recognition has often been used to classify subtypes of developmental dyslexia. In this paper, Ho et al. explored the possibility of applying this model to Chinese developmental dyslexics. Through testing of Chinese children with and without reading difficulties, the authors concluded that there were not two distinct types of dyslexia in Chinese readers. About half of the children with reading difficulties could be classified as surface dyslexics, but there were no phonological dyslexics. Ho et al. offer two possible explanations of this finding: either the two routes to word reading – the lexical and sublexical – are highly interdependent in Chinese, or there may only be one route from phonology to orthography, as connectionist models suggest.

Shaywitz, B., Shaywitz, S., Blachman, B., Pugh, K., Fulbright, R., Skudlarski, P., Mencl, W., Constable, R., Holahan, J., Marchione, K., Fletcher, J., Reid., L., & Gore, J. (2004).   "Development of left occipital systems for skiled reading in children after a phonologically-based intervention." Biological Psychiatry, 55, 926-933.

This is the largest fMRI study of reading intervention in children. It indicates that the nature of the remedial education intervention is crucial to successful outcomes in children with reading disabilities. It argues for a phonologically based intervention for the development of fast-paced neural systems that underlie skilled reading.

Eckert, M., Leonard, C., Richards, T., Aylward, E., Thomson, J., & Berninger, V. (2003).  "Anatomical correlates of dyslexia: Frontal and celebellar findings." Brain, 126, 482-494.

Eckert et al. studied the neuroanatomy of dyslexic and non-dyslexic children to discover if there are any anatomical distinctions that can be made between the two groups. They found that there were many such distinctions. Through the use of structural MRI, children with dyslexia were shown to have smaller right anterior lobes of the cerebellum, smaller pars triangularis in both hemispheres, and an overall smaller brain volume.