18

Development of second language
reading skills

cross-linguistic perspectives

Keiko Koda

Introduction

Reading is a complex construct, involving multiple operations and a unique set of skills each of those operations entail. When learning to read occurs in a second language (L2), the complexity increases exponentially because each operation involves two languages (Koda, 2005). The primary objective of this chapter is to explore how the involvement of two languages affects L2 reading development. In probing highly complex issues, entailing multiple skills in multiple languages, it is important to clarify the basic assumptions underlying the conceptual explorations.

First, under the input driven view of learning (Ellis, 2002), learning—be it of language or reading—is regarded as the process of detecting, abstracting, and internalizing regularly co-occurring elements in input as corresponding units. Learning is achieved primarily through cumulative experience of mapping between corresponding elements. The more frequently particular patterns of mappings are experienced, the stronger the linkages holding them together. Under this view, reading skills are seen as an outcome of cumulative experience of symbol-to-sound, as well as symbol-to-morpheme, mappings.

Next, from a developmental perspective, Perfetti and associates (Perfetti, 2003; Perfetti and Dunlap, 2008) regard reading as a dynamic pursuit embedded in two interrelated systems: a language and its writing system. Because no writing systems directly encode meaning independent of language, reading acquisition involves making links between the two systems. In learning to read, therefore, children must uncover how spoken language elements are mapped onto the graphic symbols that encode them. It is hardly accidental that reading instruction does not commence until children acquire sufficient spoken language competence. Reading development thus necessitates substantial linguistic knowledge.

Transfer has been a major theoretical concept in second language (L2) research. Studies have repeatedly shown that L2 learning is affected by previously acquired competencies. Based on this notion of transfer, L2 reading research generally assumes that L2 reading skills are shaped jointly by transferred first language (L1) skills, L2 linguistic knowledge and L2 print input. In this regard, L2 reading is cross-linguistic, entailing continual interaction and assimilation of L1 and L2 factors. Theories of L2 reading must explain how the involvement of two languages affects L2 reading development; how transferred L1 skills are assimilated in L2 reading; and how the resulting L2 skills vary across learners with diverse L1 backgrounds.

Historical discussion

L2 reading research has evolved both as an extension of L1 reading and as a branch of second language acquisition (SLA). Reflecting such subordination, impacts of major conceptual changes in the two fields are apparent in its theoretical orientations. In the sections below, those changes are briefly described and their relevance to the emergence of cross-linguistic perspectives in L2 reading research discussed.

Shifting views of reading

Over the past 40 years, L1 reading research has gone through dramatic changes in the way the very construct of reading is conceptualized. In the early days, under the premise that reading is a conceptually driven process, it was likened to a “psycholinguistic guessing game” (Goodman, 1973). In this view, readers were believed to engage in forming and confirming a series of hypotheses about up-coming text contents. Two assumptions underlie this top-down view of reading. First, as a communication (meaning-making) system, reading is an indivisible whole and should be treated as such in examining its process and development (Goodman, 1967). As a result, early reading studies treated reading as a single unitary construct, relegating all component skills to the periphery. Second, as a conceptually driven enterprise, the basic process of reading is largely unaffected by language elements and their properties (Goodman, 1969). Reading was thus considered universal across languages. However, both these assumptions were later challenged and largely discredited.

In the following decade, advances in technology made it possible to incorporate more sophisticated measurements and data collection techniques in reading research, and as a consequence, more complex models of reading began to emerge. Unlike the earlier view, these newer models uniformly regard reading as a complex, multi-faceted, construct. In the component skills approach (Carr and Levy, 1990), for example, reading is seen as the product of a complex information-processing system, involving a constellation of closely related mental operations. These operations are theoretically distinct and empirically separable. To fulfill the assigned function, each of these operations necessitates a set of processing skills, which interact and impact upon one another to facilitate perception, comprehension, and memory of visually presented language. The component approach further presumes variances in virtually all facets of reading ability. Studies conducted under this view aim to determine which particular variances are centrally related to reading competence by isolating component skills for inspection. The approach thus provides solid bases for determining the sources of reading problems attributable to a single deficiency or a combination of multiple deficiencies in specific skills.

The universality assumed in the top-down view of reading has also been contested. Experimental psychologists have repeatedly challenged language-processing theories based on data obtained exclusively from native English-speaking participants. Sentence processing research, for example, has shown that the cognitive strategies involved in sentence comprehension and production are heavily constrained by the syntactic properties of the language inspected (Bates and MacWhinney, 1989; Bates et al., 1982; Kail, 1989; Mazuka and Itoh, 1995). Word recognition studies have also reported that different procedures are used during print information processing by skilled readers of a wide variety of languages employing typologically diverse writing systems, and that those varied procedures are identifiable with the distinct properties of their respective writing systems (e.g., Katz and Frost, 1992; Saito et al., 1999; Taft and Zhu, 1995; Vaid, 1995).

Of late, the notion of reading universals has reemerged. Unlike the earlier view, the current conceptualization (Perfetti and Dunlap, 2008; Perfetti and Liu, 2005; Perfetti, 2003) incorporates the multi-dimensionality of reading and cross-linguistic variations in its operations. By specifying the universally mandated demands for learning to read, imposed on all learners in all languages, the theory sets the limits on possible variations in learning to read across languages and in so doing, stipulates precisely where cross-linguistic variations occur. Such stipulation is critical because languages vary in virtually all aspects of meaning-making conventions, as well as in the methods of graphically encoding those conventions. Clearly, the theory laid a solid foundation on which the language-specific demands for learning to read can be identified and compared systematically across typologically diverse languages.

Changing conceptions of reading transfer

Although transfer has long been a major concern in L2 reading research, there is no explicit theory explaining how reading skills developed in one language affect learning to read in another. In the absence of general guiding principles, transfer studies pursued how L1 and L2 reading abilities are related primarily under the developmental interdependence hypothesis (Cummins, 1979, Cummins, 1991). The hypothesis posits that academic language competence is supported by a set of non language-specific capacities, referred to as “common underlying proficiency.” Hence, the development of L2 academic language competence, including literacy skills, is determined, to a major extent, by the degree to which the “common underlying proficiency” has been established in the primary language. Although the hypothesis has prompted a significant amount of useful research, it offers little clarification as to what constitutes the core construct of “common underlying proficiency.” As a result, subsequent studies examining the hypothesized interdependence vary in nearly all aspects of their design, including participants’ characteristics (age, language background, ethnicity, academic status), “academic” skills measured and assessment tools used. These deviations make it difficult to consolidate empirical findings across studies to draw collective conclusions regarding the role of shared competencies in learning to read in an additional language.

In more recent studies, reading skills have been more clearly defined within the component approach and more accurately measured in two languages. Such refinements make it possible to clarify the fundamental notion of “common underlying proficiency” and to empirically examine cross-linguistic relationships in a variety of skills both in isolation and in tandem in the pursuit of “developmental interdependence.” These newer studies have shown systematic connections in a still small, but critical, set of skills (e.g., phonological awareness, decoding, orthographic processing) across diverse combinations of languages (e.g., Spanish/English, Hebrew/English, Turkish/ Dutch, English/Chinese) in school-age biliteracy learners.

Core issues

Over the years, the conceptual bases of L2 reading research have become increasingly more sophisticated. Under the more refined views of reading, a growing number of studies have begun to use programmatic approaches in addressing complex, cross-linguistic, issues in L2 reading development. In particular, three interrelated questions are currently being addressed: (1) how do reading skills in two languages relate to one another?; (2) which factors affect the way transferred L1 skills are assimilated in L2 reading?; and (3) how do the resulting L2 skills vary across learners with diverse L1 backgrounds?

Cross-linguistic relationships in reading skills

What is actually transferred is one of the most fundamental questions in reading transfer research. Within the component skills approach, many of the skills vital to successful reading have been identified. While some skills—for example, those involved in linguistic information processing—are language-specific, directly affected by linguistic properties, others are cognitive and do not vary across languages. Given these differences, it is critical to determine whether all transferred skills are assimilated in L2 reading in the same manner and to the same extent. If the answer is yes, all reading skills in one language should relate systematically and similarly to corresponding skills in the other. If, however, disparities exist, there should be parallel divergence in the way corresponding skills are related between two languages. Two hypotheses— “central processing” (Da Fontoura and Siegel, 1995) and “script-dependent” (Gholamain and Geva, 1999)—are useful in exploring variations in the way the two types of skills (language-specific and non language-specific) affect reading development in an additional language.

The “central processing” hypothesis posits that differences in underlying cognitive factors (such as working memory and coding speed) are primarily responsible for variances in reading achievements in school-age bilingual learners in both their languages. Its central claim is that children who are good readers in their L1 have a much better chance of achieving higher reading proficiency in a second language. Conversely, children with specific reading disability in one language are likely to experience similar difficulties in another language. Obviously, the notion of “central processing” originated in the concept of “common underlying proficiency” in Cummins’ developmental interdependence hypothesis (1979, 1991). Grounded in the current theories of reading, however, the “central processing” hypothesis provides clear delineations of the skills that constitute the “central processing” capacity, making it possible to test the specific predictions the hypothesis has generated.

In contrast, the “script-dependent” hypothesis contends that decoding development is facilitated by phonological transparency (i.e., degrees of regularity in symbol-sound correspondences) of the writing system. The hypothesis aims to explain why decoding skills are acquired more easily in some languages than in others. The formulation underscores the significance of phonological transparency and other language-specific properties as a critical factor explaining variations in reading acquisition.

Although the two formulations seem diametrically opposite in their core contentions, the two are complementary in that each explains the specific way in which two types of reading skills, language-specific (e.g., symbol-to-sound mappings) and non language-specific (e.g., word segmentation) affect L2 reading development. The distinction is important because the two types of skills presumably behave differently when transferred to L2 reading. To illustrate, non language-specific skills, once developed in one language, should be available and serviceable in learning to read in another language. In principle, these skills, if available, should provide all L2 learners, regardless of their L1 background, with direct and equal facilitation in L2 reading acquisition. However, such uniformity cannot be expected for language-specific skills because they are closely attuned to L1 properties, which may or may not be shared between two languages. For those skills to become functional in a second language, they need to undergo varying degrees of modification. The question then is what determines how much modification is necessary for such functional adjustment.

Factors affecting the assimilation of L1 skills in L2 reading

Once the involvement of L1 skills in L2 reading is established empirically, the next step is to uncover what happens to language-specific skills when they transfer across languages. Because no two languages are identical, transferred L1 skills must be adjusted to L2 properties. Given that some combinations of languages are more closely related than others, the amount of modification necessary for functional adjustment should vary. The second cross-linguistic issue deals with the factors explaining such variation.

Linguistic distance (degrees of similarity) between two languages is one such factor because it is responsible in large part for individual differences in the rate in which L2 reading skills develop. When the two languages share similar structural properties, what is required for linguistic information processing should also be similar—if not identical. Alphabetic writing systems, for example, share the basic unit of orthographic representation, requiring similar symbol-to-sound mapping procedures for decoding regardless of the graphic form of the symbols (e.g., the Roman script in English vs. the Cyrillic script in Russian). When transfer occurs between two alphabetic systems, therefore, L1 mapping skills should be functional in L2 decoding with minimum modification. Conversely, when two languages are distinct, transferred skills must be substantially modified. L1 and L2 distance thus determines the extent of adjustment necessary for the assimilation of transferred L1 skills in L2 reading.

Ultimately, however, it is the quality (linguistic/orthographic properties) and quantity (frequency) of L2 print input that determine the form and level of L2 skills emerging from a particular instance of L2 literacy learning (Koda, 2008). It is essential that the nature of L2 input be analyzed and documented. Despite its significance, limited attention has been give to input properties, such as structural transparency, frequency, and formation regularity. Investigations of L2 factors have focused mainly on the effects of general language proficiency. Of late, however, a small, but growing, number of studies have begun to incorporate detailed analysis of L2 input properties as the basis for generating hypotheses regarding the relative ease at which varying L2 skills are acquired. Properly sustained, these on-going studies will eventually yield far more specific information on how transferred skills are assimilated in L2 reading.

Variations in L2 reading skills

As mentioned earlier, dual-language involvement is the defining characteristic of L2 reading. It is vital to clarify how two languages interact during L2 print information processing. In the past decade, interest in such interactions has risen. The basic assumptions are two-fold: (1) L2 reading skills are shaped through continual cross-linguistic interactions between transferred L1 skills and L2 print input, and (2) such interactions induce sustained assimilation of print processing experiences in two languages. It follows then that the resulting L2 skills should reflect the major properties of the two languages involved, and therefore, vary systematically across learners with diverse L1 backgrounds. The third, and final, issue has to do with the empirical validation of such variations.

Data and common elicitation measures

Empirical explorations of the cross-linguistic issues outlined above necessitate at the minimum clear specification of focal skills and linguistic demands for their utilization in two languages. Such clarification requires a series of analyses, including construct analysis (isolating reading skills involved in a particular task in learning to read), linguistic analysis (identifying the properties of the linguistic facet directly related to the task under consideration), and cross-linguistic analysis (determining how the properties of the relevant linguistic facet vary between two languages). Sequentially, these analyses enable researchers to convert highly complex cross-linguistic issues into empirically testable hypotheses by allowing them to set the scope of their investigations. The sections below briefly describe these analyses, and then discuss the methodologies commonly used to test the hypotheses generated through the analyses.

Setting the scope: Generating hypotheses

Construct analysis. As a complex psycholinguistic construct, reading involves three major operations: (a) decoding (extracting linguistic information directly from print), (b) text-information building (integrating the extracted information into phrases, sentences, and paragraphs), and (c) reader-model construction (synthesizing the amalgamated text-information with prior knowledge). In examining the impacts of dual-language involvement on L2 reading development, it is essential to clarify the requisite skills for each operation. As an illustration, phonological and morphological information extraction constitutes the requisite tasks for decoding, and as such, the operation necessitates the skills to map phonological and morphological elements onto units of graphic symbols in the writing system. Because languages differ in the way they graphically represent sub-lexical information, the skills optimal for the required mappings vary across languages. Linguistic analysis is thus in order.

Linguistic analysis. In order to properly address cross-linguistic issues in L2 reading, we must establish reliable methods of comparing corresponding skills between two languages. One such method is to identify the language-specific demands for a particular task by analyzing the properties of the linguistic facet directly related to the task. Using decoding as an example again, efficiency in phonological information extraction in English is largely determined by the ability to manipulate intra-syllabic sounds, such as onset, rhyme, and phoneme, by assembling and dissembling letters and letter clusters (e.g., Ehri, 1998; Shankweiler and Liberman, 1972). It has been widely recognized that competent readers are uniformly adept at pronouncing both individual letters and nonsense letter-strings (e.g., Hogaboam and Perfetti, 1978; Siegel and Ryan, 1988; Wagner et al., 1994); and that such ability depends on orthographic knowledge, that is, understanding the rules about how graphic symbols are used to represent speech sounds (e.g., Adams, 1990; Ehri, 1994, 1998; Seidenberg and McClelland, 1989). In principle, therefore, the linguistic demands for phonological information extraction can be identified by uncovering which phonological unit (e.g., syllable, phoneme) is directly encoded in each graphic symbol and how the symbols are combined to represent spoken words. Accurate descriptions of such demands in two languages make the subsequent, cross-linguistic, analysis possible.

Cross-linguistic analysis. Documentation of the linguistic demands for a particular task allows systematic comparisons of the requisite skills for the task in two languages. Such comparisons permit reasonably accurate estimates of the relative ease with which transferred skills gain functionality in a second language. Once again, using decoding as an illustration, cross-linguistic variation in the requisite skill for phonological information extraction can be identified by comparing the basic unit of orthographic representation in the two languages involved.

Korean children, for example, must develop sensitivity to both syllables and phonemes, because phonological information is encoded at these two levels in their writing system, Hangul. In the Hangul script, the graphic symbols, each representing a distinct phoneme, cannot appear on their own; instead, they must be packaged into blocks to form syllables. In learning to read, phoneme and syllable manipulation skills are both strong predictors of word reading ability in Korean children (McBride-Chan et al., 2005). In contrast, in consonantal Hebrew, children develop stronger sensitivity to consonants than to vowels, and this sensitivity plays a pivotal role in Hebrew literacy development (Geva, 2008; Tolchinsky and Teberosky, 1998). In logographic Chinese, moreover, the grapheme-morpheme connections are more prominent than the grapheme-syllable linkages. Reflecting the stronger morpheme representation, morphological awareness has been found a more powerful predictor of initial reading success in Chinese than is phonological awareness (Ku and Anderson, 2003; Li et al., 2002). Evidently, the requisite skill for a particular task is closely aligned with the properties of the linguistic facet directly related to the task. The incorporation of cross-linguistic analysis, thus, allows empirical examinations of the hypothesized variation in the way previously acquired reading skills affect L2 reading development.

Empirical examinations: Testing hypotheses

Correlational studies. The current views of transfer uniformly treat previously acquired skills as potential resources promoting reading acquisition in another language (August and Shanahan, 2006; Riches and Genesee, 2006). The first step in exploring facilitation benefits stemming from prior literacy experience is to determine whether L1 skills are indeed involved in L2 reading. If a new set of reading skills builds on previously acquired skills, there should be systematic connections between corresponding skills in two languages. Correlational analysis has been widely used to examine cross-linguistic relationships in a variety of reading skills (e.g., phonological awareness, morphological awareness, decoding).

Once the involvement of L1 skills is established empirically, the next step is to determine the directionality in the observed relationships. The problem is that the mere presence of a correlation between two variables does not indicate which of the two may be affecting the other. There is also a possibility that a correlation can be attributed to a “third” variable to which the two variables in the correlation are both related. For these reasons, correlational studies are limited in their simplest forms. They can be made more powerful, however, by incorporating statistical techniques that allow the partial control of “third” variables (Stanovitch and Cunningham, 2004). These techniques, including multiple regression, path analysis, and structural equation modeling, allow the recalculation of the correlation between two variables after the effect of other key variables (possible “third” variables) are removed.

In L2 reading research, hierarchical multiple regression has been used extensively in determining the relative contributions of a variety of competencies, such as oral vocabulary knowledge, phonological awareness and working memory, to L2 reading performance both within and across languages. In these studies, cognitive and maturational factors (e.g., non-verbal intelligence and age) are used as control variables. The inclusion of these and other control variables in the regression analysis ensures that the observed associations between any of the predictor variables (e.g., oral vocabulary knowledge, working memory) and the criterion variable (i.e., reading performance) are not mediated by other related factors. Given that a large proportion of data in educational research is correlational, this and other like techniques are highly advantageous in exploring the role of previously acquired skills in L2 reading development.

Group comparison studies. Another way of exploring the impact of prior literacy experience on L2 reading development is to demonstrate L1-induced variations in L2 processing behaviors among learners with diverse L1 backgrounds. Typically, empirical evidence for such variations is obtained through cross-linguistic experiments, wherein performances on experimental tasks are compared between two comparable L2 learner groups, differing only in their L1 backgrounds. Experimental tasks are designed to elicit particular processing behaviors, which are clearly identifiable with specific L1 properties. By design, experimental manipulations should induce a greater impact on the group whose L1 processing behaviors are targeted for manipulation. If the two groups respond differentially to such manipulations, the observed variation can be taken as an indication that L1 skills are indeed used during L2 print information processing.

To illustrate, L2 learners of English (ESL) with alphabetic and logographic L1 backgrounds can be contrasted in their reliance on phonemic analysis during phonological information extraction. Because alphabetic literacy requires segmenting and assembling phonological information at the phonemic level, alphabetic readers rely heavily on phonemic analysis. In contrast, phonological decoding in logographic languages does not entail similar analysis because the spoken sound of a single morpheme corresponds holistically to a single graphic symbol. A word's phonology (usually a syllable) is mapped onto the visual display (a single character) as a whole. Hence, phonological information extraction in logographic writing systems requires no analysis or segmentation at the phonemic level. Instead, logographic readers are required to distinguish a large number of visually complex symbols. In light of these disparities, two experimental manipulations—one blocking a word's phonological information and the other distorting the visual configuration—should affect phonological decoding differently among ESL learners with alphabetic and logographic L1 backgrounds. When phonological information is made unavailable, decoding performance of alphabetic ESL learners will be more seriously disrupted, whereas visual distortion will impede phonological decoding among logographic learners to a greater extent.

These predictions have been tested using a variety of experimental manipulations, including case alterations (e.g., replacing “read” with “ReAd”; Akamatsu, 2003) and phonological information blocking (e.g., replacing words with unpronounceable symbols; Koda, 1990). Because these manipulations are designed to cause processing disruption by blocking the information presumed to be critical for phonological decoding in participants’ respective first languages, performance decline, if occurred, can be taken as evidence for the presence of L1 decoding skills in L2 print information processing. The section that follows discusses empirical results from studies examining the three core issues.

Empirical verification

Cross-linguistic relationships in reading skills. On the assumption that L2 reading success depends largely on common underlying competencies, early bilingual studies investigated how L1 and L2 reading abilities are related (e.g., Cummins et al., 1981; Legarretta, 1979; Skutnabb-Kangass and Toukomaa, 1976; Troike, 1978). Their results consistently showed that reading abilities among school-age children were strongly related in their two languages. Such a relationship was taken as indicating that L1 reading competence is the chief determinant of L2 reading achievement (Cummins, 1979, 1991). In these early studies, however, reading was uniformly treated as a single unitary construct, and the critical question—how corresponding skills are related between two languages–remained unaddressed.

Under the componential view of reading, more recent studies have begun to incorporate larger batteries of tasks designed to measure a variety of skills and cognitive capacities, including phonological awareness (Branum-Martin et al., 2006; Durgunoglu et al., 1993; Wade-Woolley and Geva, 2000; Wang et al., 2006), decoding (Abu-Rabia, 1997; Da Fontoura and Siegel, 1995; Durgunoglu et al., 1993; Geva and Siegle, 2000; Gholamain and Geva, 1999; Wade-Woolley and Geva, 2000), syntactic awareness (Abu-Rabia, 1995; Da Fontoura and Siegel, 1995), and working memory (Abu-Rabia, 1995; Da Fontoura and Siegel, 1995; Geva and Siegle, 2000; Gholamain and Geva, 1999).

Because these studies involved children learning to read two alphabetic languages, it is less certain whether the reported cross-linguistic relationships can be generalized to others whose literacy learning involves two typologically dissimilar languages. A small, but growing, body of evidence suggests that phonological awareness is systematically related in Chinese and English (Bialystok et al., 2005; McBride-Chan et al., 2006; Wang et al., 2005), but phonological awareness in one language only minimally contributes to decoding in the other (Luk and Bialystok, 2008; Wang et al., 2005). This makes a stark contrast with the strong functional relationship between L1 phonological awareness and L2 decoding skills found in Spanish-English bilingual children (e.g., Durgunoglu et al., 1993). Unlike phonological awareness, moreover, other competencies, such as orthographic knowledge and morphological awareness, are not systematically related in Chinese and English in school-aged biliteracy learners (Wang et al., 2005).

A sub-set of the biliteracy studies listed above further explored, within the “central processing” hypothesis, common deficiencies explaining reading failure in both languages in at-risk bilingual school-age readers. Their analyses revealed that poor readers are uniformly weak in phonological skills in both languages; and that their deficiencies usually are domain-specific and cannot be explained by non-phonological factors (e.g., Abu-Rabia, 1995; August et al., 2002; Carlisle and Beeman, 2000; Cormier and Kelson, 2000; da Fontoura and Siegel, 1995; Gholamain and Geva, 1999; Verhoeven, 2000; Wade-Woolley and Geva, 2000). These results provide strong empirical support for the critical role that “central processing” capacities, or “common underlying proficiency” (Cummins, 1979), play in literacy learning in an additional languages.

Taken as a whole, these findings make it plain that L1 and L2 reading skills are systematically related; that, as in L1 reading, L2 decoding development depends on phonological awareness; and that such dependency occurs both within and across languages. Caution is necessary, however, in generalizing the reported cross-linguistic relationships because current studies have concentrated primarily on school-age children at the initial grades, as well as on the functional connection between phonological awareness and decoding between two alphabetic languages. Obviously, more studies are needed to expand the current data base on cross-linguistic relationships in reading skills.

Factors affecting the assimilation of L1 skills. Within the input driven view of learning, it is assumed that previously acquired reading skills continue to evolve through L2 print input processing until they reach optimal efficiency in the new language. The second cross-linguistic issue pertains to the factors affecting the assimilation of transferred L1 skills. Two factors are particularly germane: (1) L1 and L2 linguistic distance because it determines how much modification is necessary for L1 skills to be functional in L2 reading, and (2) L2 print pocessing experience because L2 input is the dominant force in reshaping transferred L1 skills.

Linguistic distance varies widely across diverse L2 learner groups. As noted earlier, it directly affects L2 processing efficiency at a given point in time among learners with similar and dissimilar L1 backgrounds. Although L2 studies have repeatedly shown faster and more accurate processing performance for learners from typologically related L1 backgrounds than those from unrelated backgrounds (e.g., Green and Meara, 1987; Hamada and Koda, 2008; Koda, 2000), to date, little is known about precisely how shared structural properties facilitate L2 reading development. Muljani and colleagues (1998) shed some light on the issue by testing orthographic distance effects on L2 intraword structural sensitivity. Comparing lexical-decision performance among proficiency-matched adult ESL learners with alphabetic (Indonesian) and logographic (Chinese) L1 orthographic backgrounds, the researchers demonstrated that only Indonesian participants benefited from intraword structural congruity (i.e., spelling-patterns consistent between English and Indonesian). However, their superiority was far less pronounced with incongruent items whose spelling patterns were unique to English. Thus, while orthographic distance appears to have a general facilitative impact, accelerated efficiency occurs only in the tasks whose demands are identical to those imposed by L1 properties.

Similar findings have been reported in studies comparing morphological segmentation skills among ESL learners with alphabetic (Korean) and logographic (Chinese) L1 backgrounds (Koda, 2000; Koda et al., 1998). As does phonological decoding, morphological information extraction requires intraword segmentation and analysis in alphabetic languages, whereas morphological processing in logographic languages entails a more holistic operation, which does not involve segmentation or analysis. As predicted, Korean learners were more efficient in morphological segmentation than their proficiency-matched Chinese counterparts, but their efficiency gap was substantially reduced when the groups were confronted with the items whose structural properties are unique to English. Clearly, segmentation efficiency in the items structurally unique to the target language is far less affected by linguistic distance presumably because their analysis requires insights not yet available to either Korean or Chinese ESL learners. These findings corroborate the results from the Muljani et al., study, suggesting that the distance effect is far more specific and localized than has been assumed.

Another factor directly affecting the assimilation of transferred reading skills is L2 print input. As noted earlier, reading skills are shaped through cumulative print processing experience. The nature of print input thus is a critical factor determining the level and form of the skills emerging from a particular instance of literacy learning. Documenting L2 input properties is crucial in predicting how transferred L1 skills are modified through L2 print processing experience. In a recent study (Koda et al., 2007), the impact of print input has been examined with bilingual school-age children learning to read English as their primary language and Chinese as a heritage language (CHL). The researchers analyzed properties of the Chinese characters explicitly taught in grades 1 to 6 Language Arts textbooks specifically designed for CHL learners. Their analysis revealed that CHL students are taught roughly 35 percent of the characters and 20 percent of the radicals (i.e., graphic components in characters providing semantic information) that are introduced in grades 1 to 6 Language Arts textbooks for native Chinese-speaking children in China. Despite these quantitative differences, however, the types of characters and proportions of structurally and functionally regular characters were remarkably similar between the two textbook corpora. Based on the analysis, the researchers examined how the identified input (Chinese characters) properties related to morphological awareness among grade 3 to 5 CHL students. Their data demonstrated that CHL students in the three grade levels were similarly sensitized to the major properties of the morphologically complex characters, and that few children in any grade group showed sensitivity to refined facets of morphological awareness. Given that major growth in morphological awareness occurs between grades 2 and 5 among native Chinese-speaking children (Ku and Anderson, 2003; Shu and Anderson, 1999), these results are astonishing. Apparently, the input available to CHL students provided a sufficient basis for forming sensitivity to the major input properties, but it was too restricted to allow those children to refine their preliminary understanding, and as a result, their morphological awareness remained basic. Viewed as a whole, findings from this and other initial studies (e.g., Wang et al., 2005) are illuminating, suggesting that input characteristics in the target language are directly related to the level and form of reading skills acquired through literacy learning in that language.

Cross-linguistic variations. The third and final cross-linguistic issue addressed in L2 reading research has to do with L1-induced variations in L2 reading skills. Two basic assumptions underlie empirical explorations of such variations: (1) L2 skills are shaped through cross-linguistic interactions between transferred L1 skills and L2 print input; and (2) the resulting L2 skills reflect both L1 and L2 linguistic properties. Consequently, studies aim to determine whether systematic differences exist in L2 print processing behaviors among learners with contrasting L1 backgrounds. Their results generally confirm that L2 learners respond differently to a variety of experimental manipulations (e.g., Akamatsu, 2003; Brown and Haynes, 1985; Green and Meara, 1987; Koda, 1998, 1999); and more critically, that the observed differences are attributable to the structural variations in participants’ respective L1 writing systems (e.g., Koda, 1989, 1990, 1993; Ryan and Meara, 1991).

In more recent studies, an additional step has been taken to explore cross-linguistic interactions during L2 processing. These studies have examined the relative impacts of L1 and L2 factors on L2 lexical processing, using diverse experimental tasks, including semantic category judgment (Wang et al., 2003), associative word learning (Hamada and Koda, 2008), and word identification (Wang and Koda, 2005). To distinguish the impact stemming from L1 factors from that associated with L2 factors, in these studies, L2 stimulus words were manipulated in one way or another, and the effects of such manipulations were compared between two learner groups each representing a distinct first language. In such a design, the extent that a particular manipulation affects both groups is used as the basis for gauging the L2 impact, and the degree to which the effect of the manipulation varies between the participant groups serves as an index of the L1 impact.

Through semantic category judgments, for example, Wang et al., (2003) compared the relative impacts of phonological and graphic manipulations on judgment performance (both accuracy and speed) among ESL learners with alphabetic (Korean) and logographic (Chinese) L1 backgrounds. In the study, participants were first presented with a category description, such as “flower,” and then showed a target word. They were then asked to decide whether the word was a member of the category. The trick was that the target words were manipulated either phonologically (using homophones as targets; e.g., “rows” for “rose”) or graphically (using similarly spelled words as targets; e.g., “fees” for “feet”). The primary hypothesis was that the two ESL groups would respond differently to the two types of manipulation: Korean participants would be more likely to accept homophones as category members while Chinese would make more false positive responses to graphically similar targets. The data demonstrated that both phonological and graphic manipulations significantly interfered with category judgment performance among ESL learners regardless of their L1 backgrounds. However, the magnitude of interference stemming from each type of manipulation varied between the groups. As predicted, Korean learners made more errors with homophonic (phonologically manipulated) items, while similarly spelled (graphically manipulated) targets seriously affected judgment performance among Chinese learners. These results seem to suggest that (1) proficiency-matched ESL learners are similarly sensitized to L2 properties, (2) that the two groups rely on different information sources during L2 lexical processing, and (3) that these differences are consistent with the variations predicted from participants’ L1 orthographic properties.

To summarize, the studies investigating cross-linguistic variations generally suggest that while L1 literacy experience has a long-lasting impact on L2 reading development, proficiency-matched L2 learners are similarly affected by L2 properties. These findings seem to imply that L2 input is a more dominant force in shaping L2 reading skills. In fact, in all studies comparing L1 and L2 impacts, L2 variables had stronger effects, reducing the L1 background x stimulus manipulation interaction effect. Viewed collectively, the empirical findings discussed above provide a sufficient basis for drawing two conclusions: (1) prior literacy experience affects L2 reading development, and therefore (2) previously acquired skills play a pivotal role in explaining individual differences in L2 print information processing. Beyond these, however, the currently available data indicate that cross-linguistic variations in L2 print processing behaviors are far more complex than previously assumed because they can be explained by a number of factors, including reading skills, linguistic distance, L2 input, experimental manipulations, and the tasks used for measuring the focal reading skills.

Applications

Translating research into practice is not an easy task because not all insights brought to light in empirical studies can be taught directly or even have direct utility in increasing instructional efficiency. Although there is an impressive body of information on L1 reading development, the information alone–without considering the unique nature of L2 reading—is not likely to improve reading proficiency among linguistically diverse L2 learners. Throughout the chapter, dual-language involvement has been underscored as the major characteristic of L2 reading. A great number of studies have repeatedly shown that both L1 and L2 factors play a role in shaping L2 reading skills. It is important therefore to reflect on how dual-language involvement can be dealt with effectively in L2 reading instruction.

The clear implication of L1-induced variations is that no one method works optimally for all learners. Teachers should be aware that previously acquired reading skills have lasting impacts on learning to read in a new language. Although it is virtually impossible for them to know every possible variation brought about through transferred L1 skills, the awareness is vital in accommodating the disparate needs of their students. For example, teachers could address the disparities by incorporating diagnostic reading assessments. Their awareness of long-term L1-impacts would guide them in two critical phases: constructing assessment tasks and interpreting assessment outcomes. In this way, diagnostic information could allow them to compare the requisite skills across diverse learner groups, and in so doing, identify deficiencies in any of those skills, and variations therein. It would also help teachers fine-tune their instruction to meet the specific challenges their students face in learning to read in the new language.

Given that reading skills are shaped through input exposure and experience, L2 reading development can be promoted by improving the quality and quantity of the L2 print input that can be incorporated in instruction. Although simulating the quantity of input typically available in L1 learning to read is not feasible within the L2 instructional context, it is certainly possible to maximize the quality of input experience by manipulating input presentation, as well as by increasing input processing experience. Input presentation can be enhanced by reorganizing instructional materials in such a way that cumulative input exposure leads to heightened awareness of the structural regularities (e.g., regularly spelled words and canonically ordered sentences) directly related to the requisite information processing for text meaning construction. Additionally, input experience can be enhanced by providing operations processing exercises carefully designed to engage students in input analysis in constructing local text meaning (e.g., word meaning inference).

Dual-language involvement has different implications for reading instruction for school-age children learning to read in two languages concurrently. Because reading skills in young learners are still developing in their primary language, we cannot assume that they all possess the requisite competencies for the initial tasks in learning to read in either language. This indicates that school-age learners may encounter reading difficulties fundamentally different from those experienced by adult L2 learners because their problems could stem from various combinations of multiple deficiencies, including underdeveloped L1 metalinguistic awareness, insufficient L2 linguistic knowledge, and inadequate “central processing” capacity. Here again, it is essential that teachers be clear about these multiple sources of difficulties their students might face in learning to read in their two languages.

Future directions

This chapter has explored how L2 reading development is jointly affected by L1 and L2 factors. The picture emerging from the existing database is extremely complex, showing reading skills developed in one language differentially affecting the formation of an additional set of skills in another language. Although the picture captures the multiple layers of diversities inherent in L2 reading, it is far from complete because the currently available data do not encompass the full spectrum of skills, a broad range of learners, and a sufficient variety of learning contexts. To gain a deeper understanding of L2 reading development, the scope should be extended to additional related factors. Future research directions can be set to purposefully expand the current scope in the following three areas: (a) incorporating a broader range of skills, (b) including a wider variety of languages, and (c) integrating learner-external factors into psycholinguistic accounts of L2 reading development.

Incorporating a broader range of reading skills

Reading transfer research of late has begun to address one of the most fundamental questions–the question of which skills are transferred. Currently, the investigative goal is focused primarily on cross-linguistic relationships in decoding and related skills. Successful text comprehension, however, entails far more than decoding. It is crucial to extend the investigative scope to higher-order operations, such as text information building and reader-model construction. Because the execution of the tasks involved in these higher-order operations depends on well-established decoding skills, future studies should probe the functional and developmental interconnections among skills at different processing levels both within and across languages. Such explorations could yield significant new insights into the cognitive and linguistic requisites for the execution of various higher-order tasks directly contributing to text comprehension.

Including a wider variety of languages

Although a solid body of evidence suggests that prior literacy experience does affect L2 reading development, little is known about how previously acquired skills become functional in a typologically distant second language. The information is vital in understanding the precise nature of facilitation stemming from reading skills transfer. Obtaining such information necessitates accurate descriptions of the language-specific demands for a particular operation (e.g., decoding) in two languages. Without clarifying what is required for the execution of the operation in both languages, it is virtually impossible to identify what has been previously established and then to deduce what has yet to be acquired. At present, however, only limited information is available in languages other than English on how learning to read is linguistically constrained. Clearly, we need more information on how reading skills develop in typologically diverse languages.

Integrating learner-external factors

Finally, literacy learning does not occur in a vacuum. A number of learner-external factors directly and indirectly contribute to reading development. Despite the obvious relevance, learner-external factors are rarely incorporated into psycholinguistic accounts of L2 reading development. Traditionally, L2 reading research has pursued learner-internal (e.g., cognitive and linguistic) and learner-external (e.g., contextual) factors separately under clearly distinct conceptual orientations, using widely disparate methodologies. Although the psycholinguistic focus informs us which factors promote reading acquisition, it does not explain how and why those factors that come into play in reading skills development. Much can be gained by examining learner-internal and learner-external factors in tandem under a single unified framework.

One way of achieving such integration is to incorporate input properties as the pivot connecting learner-internal and learner-external factors under the input-driven view of learning. Because this view assigns a vital role to linguistic input (both oral and print) in predicting and explaining variances in psycholinguistic outcomes, it mandates accurate descriptions of the qualitative (what kinds) and quantitative (how much) input characteristics available to a focal group of learners. Hence, a framework grounded in this view can easily incorporate learner-external factors in explaining how the cognitive and linguistic resources for reading acquisition vary across learners. Obviously, integrating psycholinguistic and contextual factors in a coherent framework is challenging, but it is vital, if we are to gain a more comprehensive grasp of L2 reading development.

References

Abu Rabia, S. (1995). Learning to read in Arabic: Reading, syntactic, orthographic and working memory skills in normally achieving and poor Arabic readers. Reading Psychology, 16, 351–394.

Abu-Rabia, S. (1997). Verbal and working memory skills of bilingual Hebrew-English speaking children. International Journal of Psycholinguistics, 13, 25–40.

Adams, M. J. (1990). Beginning to read. Cambridge, MA: The MIT Press.

Akamatsu, N. (2003). The effects of first language orthographic features on second language reading in text. Language Learning, 53, 207–231.

August, D., Calderon, M., and Carlo, M. (2002). Transfer of skills from Spanish to English: A study of young learners (Final Report). Washington, D.C.: Center for Applied Linguistics.

August, D. and Shanahan, T. (Eds.). (2006). Executive summary: Developing literacy in second-language learners: report of the National Literacy Panel on Language-Minority Children and Youth. Mahwah, NJ: Lawrence Erlbaum.

Bates, E. and MacWhinney, B. (1989). Functionalism and the competition model. In B. MacWhinney and E. Bates (Eds.), The crosslinguistic study of sentence processing (pp. 3–73). Cambridge: Cambridge University Press.

Bates, E., McNew, S., MacWhinney, B., Devescovi, A., and Smith, S. (1982). Functional constraints on sentence processing: A cross-linguistic study. Cognition, 11, 245–299.

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

Branum-Martin, L., Fletcher, J. M., Carlson, C. D., Ortiz, A., Carlo, M., and Francis, D. J. (2006). Bilingual phonological awareness: Multilevel construct validation among Spanish-speaking kindergarteners in transitional bilingual education classrooms. Journal of Educational Psychology, 98, 170–81.

Brown, T. and Haynes, M. (1985). Literacy background and reading development in a second language. In T. H. Carr (Ed.), The development of reading skills (pp. 19–34). San Francisco, CA: Jossey-Bass.

Carlisle, J. F. and Beeman, M. M. (2000). The effects of language of instruction on the reading and writing achievement of first-grade Hispanic children. Scientific Studies of Reading, 4, 331–353.

Carr, T. H. and Levy, B. A. (Eds.). (1990). Reading and its development: Component skills approaches. San Diego: Academic Press.

Cormier, P. and Kelson, S. (2000). The roles of phonological and syntactic awareness in the use of plural morphemes among children in French immersion. Scientific Studies of Reading, 4, 267–294.

Cummins, J. (1979). Linguistic interdependence and educational development of bilingual children. Review of Educational Research, 49, 222–251.

Cummins, J. (1991). Interdependence of first- and second-language proficiency in bilingual children. In E. Bialystok (Ed.), Language processing in bilingual children (pp. 70–89). New York: Cambridge University Press.

Cummins, J., Swain, M., Nakajima, K., Handscombe, J., and Green, D. (1981). Linguistic interdependence in Japanese and Vietnamese students. Report prepared for the Inter-America Research Associates. June. Tronto: Ontario Institute for Studies in Education.

Da Fontoura, H. A. and Siegel, L. S. (1995). Reading syntactic and memory skills of Portuguese-English Canadian children. Reading and Writing: An International Journal, 7, 139–153.

Durgunoglu, A. Y., Nagy, W. E., and Hancin, B. J. (1993). Cross-language transfer of phonemic awareness. Journal of Educational Psychology, 85, 453–465, read (pp. 119–147). New York: Springer-Verlag.

Ehri, L. C. (1994). Development of the ability to read words: Update. In R. Ruddell, M. Ruddell, and H. Singer (Eds.), Theoretical models and processes of reading (Fourth Edition) (pp. 323–358). Hillsdale, NJ: Erlbaum.

Ehri, L. C. (1998). Grapheme-phoneme knowledge is essential to learning to read words in English. In J. L. Metsala and L. C. Ehri (Eds.), Word recognition in beginning literacy (pp. 3–40). Mahwah, NJ: Erlbaum.

Ellis, N. (2002). Frequency effects in language processing: A review with implications for theories of implicit and explicit language acquisition. Studies in Second Language Acquisition, 24, 143–188.

Geva, E. (2008). Facets of metalinguistic awareness related to reading development in Hebrew: Evidence from monolingual and bilingual and bilingual children. In K. Koda and A. M. Zehler (Eds.), Learning to read across languages: Cross-linguistic relationships in first and second language literacy development. New York: Routledge.

Geva, E. and Siegle, L. S. (2000). Orthographic and cognitive factors in the concurrent development of basic reading skills in two languages. Reading and Writing, 12, 1–30.

Gholamain, M. and Geva, E. (1999). Orthographic and cognitive factors in the concurrent development of basic reading skills in English and Persian. Language Learning, 49, 183–217.

Goodman, K. S. (1967). Reading: A psycholinguistic guessing game. Journal of the Reading Specialist, 6, 126–135.

Goodman, K. S. (1969). Analysis of oral language miscues: Applied psycholinguistics. Reading Research Quarterly, 5, 9–30.

Goodman, K. S. (1973). Psycholinguistic universals of the reading process. In F. Smith (Ed.), Psycholinguistics and reading (pp. 21–29). New York: Holt: Rinehart and Winston.

Green, D. W. and Meara, P. (1987). The effects of script on visual search. Second Language Research, 3, 102–117.

Hamada, M. and Koda, K. (2008). Influence of first language orthographic experience on second language decoding and word learning. Language Learning, 58, 1–31.

Hogaboam, T. W. and Perfetti, C. A. (1978). Reading skill and the role of verbal experience in decoding. Journal of Educational Psychology, 70, 717–729.

Kail, M. (1989). Cue validity, cue cost, and processing types in sentence comprehension in French and Spanish. In B. MacWhinney and E. Bates (Eds.), The cross-linguistic study of sentence processing (pp. 77–117). New York: Cambridge University Press.

Katz, L. and Frost, R. (1992). Reading in different orthographies: The orthographic depth hypothesis. In R. Frost and L. Katz (Eds.), Orthography, phonology, morphology, and meaning (pp. 67–84). Amsterdam: Elsevier.

Koda, K. (1989). The effects of transferred vocabulary knowledge on the development of L2 reading proficiency. Foreign Language Annals, 22, 529–542.

Koda, K. (1990). The use of L1 reading strategies in L2 reading. Studies in Second Language Acquisition, 12, 393–410.

Koda, K. (1993). Transferred L1 strategies and L2 syntactic structure during L2 sentence comprehension. The Modern Language Journal, 77, 490–500.

Koda, K. (1998). The role of phonemic awareness in L2 reading. Second Language Research, 14, 194–215.

Koda, K. (1999). Development of L2 intraword structural sensitivity and decoding skills. The Modern Language Journal, 83, 51–64.

Koda, K. (2000). Cross-linguistic variations in L2 morphological awareness. Applied Psycholinguistics, 21, 297–320.

Koda, K. (2005). Insights into second language reading. NY: Cambridge University Press.

Koda, K., Lu, C., and Zhang, Y. (2007). Properties of Characters in heritage Chinese textbooks and their implications for character knowledge development among Chinese heritage Language Learners. In A. W. He and Y. Xiao (Eds.), Chinese as a heritage language. Hawaii: University of Hawaii Press.

Koda, K., Takahashi, E., and Fender, M. (1998). Effects of L1 processing experience on L2 morphological awareness. Ilha do Desterro, 35, 59–87.

Ku, Y.-M. and Anderson, R. C. (2003). Development of morphological awareness in Chinese and English. Reading and Writing: An Interdisciplinary Journal, 16, 399–422.

Legarretta, D. (1979). The effects of program models on language acquisition of Spanish speaking children. TESOL Quarterly, 13, 521–534.

Li, W., Anderson, R. C., Nagy, W., and Zhang, H. (2002). Facets of metalinguistic awareness that contribute to Chinese literacy. In W. Li, J. S. Gaffiney, and J. L. Packard (Eds.), Chinese children's reading acquisition: Theoretical and pedagogical issues (pp. 87–106). Boston: Kluwer Academic.

Luk, G. and Bialystok, E. (2008). Common and distinct cognitive bases for reading in English–Cantonese bilinguals. Applied Psycholinguistics, 29, 269–289.

Mazuka, R. and Itoh, K. (1995). Can Japanese speakers be led down the garden path? In R. Mazuka and N. Nagai (Eds.), Japanese sentence processing (pp. 295–330). Hillsdale, NJ: Erlbaum.

McBride-Chan, C., Cheung, B. W. -Y., Chow, C. S. -L., and Choi, L. (2006). Metalinguistic skills and vocabulary knowledge in Chinese (L1) and English (L2). Reading and Writing, 19, 695–716.

McBride-Chan, C., Wagner, R. K., Muse, A., Chow, B. W. -Y., and Shu, H. (2005). The role of morphological awareness in children's vocabulary acquisition in English. Applied Psycholinguistics, 26, 415–435.

Muljani, M., Koda, K., and Moates, D. (1998). Development of L2 word recognition: A Connectionist approach. Applied Psycholinguistics, 19, 99–114.

Perfetti, C. A. (2003). The universal grammar of reading. Scientific Studies of Reading, 7, 3–24.

Perfetti, C. A. and Dunlap, S. (2008). Learning to read: General printiples and writing system variations. In K. Koda and A. M. Zehler (Eds.), Learning to read across languages: Cross-linguistic relationships in first- and second-language literacy development. Mahwah, NJ: Lawrence Erlbaum.

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

Riches, C. and Genesee, F. (2006). Crosslanguage and crossmodal influences. In F. Genesee, K. Lindholm-Leary, W. Saunders, and D. Christian (Eds.), Educating English language learners: A synthesis of research evidence. New York: Cambridge University Press.

Ryan, A. and Meara, P. (1991). The case of invisible vowels: Arabic speakers reading English words. Reading in a Foreign Language, 7, 531–540.

Saito, H., Masuda, H., and Kawakami, M. (1999). Subword activation in reading Japanese single Kanji character words. Brain and Language, 68, 75–81.

Seidenberg, M. S. and McClelland, J. L. (1989). A distributed, developmental model of word recognition and naming. Psychological Review, 96, 523–568.

Shankweiler, D. and Liberman, I. Y. (1972). Misreading: A search for causes. In J. F. Kavanaugh and I. G. Mattingly (Eds.), Language by eye and by ear (pp. 293–317). Cambridge, MA: MIT Press.

Shu, H. and Anderson, R. C. (1999). Learning to read Chinese: The development of metalinguistic awareness. In A. Inhuff, J. Wang, and H. C. Chen (Eds.), Reading Chinese scripts: A cognitive analysis (pp. 1–18). Mahwah, NJ: Lawrence Erlbaum.

Siegel, L. S. and Ryan, E. B. (1988). Development of grammatical sensitivity, phonological, and short-term memory in normally achieving and learning disabled children. Developmental Psychology, 24, 28–37.

Skutnabb-Kangass, T. and Toukomaa, P. (1976). Teaching migrant children's mother tongue and learning the language of the host country in the context of soci cultural situation of the migrant family. Helsinki: The Finnish National Commission for UNESCO.

Stanovitch, K. E. and Cunningham, A. E. (2004). Inferences from correlational data: Exploring associations with reading experience. In N. K. Duke and M. H. Mallette (Eds.), Literacy research methodologies. New York: Guilford.

Taft, M. and Zhu, X. P. (1995). The representation of bound morphemes in the lexicon: A Chinese study. In L. B. Feldman (Ed.), Morphological aspects of language processing (pp. 109–129). Hillsdale, NJ: Erlbaum.

Tolchinsky, L. and Teberosky, A. (1998). The development of word segmentation and writing in two scripts. Cognitive Development, 13, 1–25.

Troike, R. C. (1978). Research evidence for the effectiveness of bilingual education. NABE Journal, 3, 13–24.

Vaid, J. (1995). Effect of reading and writing directions on nonlinguistic perception and performance: Hindi and Urdu data. In I. Taylor and D. R. Olson (Eds.), Scripts and literacy: Reading and learning to read the world's scripts (pp. 295–310). Dordrecht: Kluwer Academic.

Verhoeven, L. (2000). Components in early second language reading and spelling. Scientific Studies of Reading, 4, 313–330.

Wade-Woolley, L. and Geva, E. (2000). Processing novel phonemic contrasts in the acquisition of L2 word reading. Scientific Studies of Reading, 4, 295–311.

Wagner, R. K., Torgesen, J. K., and Rashotte, C. A. (1994). The development of reading-related phono-logical processing abilities: New evidence of bi-directional causality from a latent variable longitudinal study. Developmental Psychology, 30, 73–87.

Wang, M. and Koda, K. (2005). Commonalities and differences in word identification skills among learners of English as a second language, Language Learning, 55, 71–98.

Wang, M., Koda, K., and Perfetti, C. A. (2003). Alphabetic and non-alphabetic L1 effects in English semantic processing: A comparison of Korean and Chinese English L2 learners. Cognition, 87, 129–149.

Wang, M., Park, Y., and Lee, K. R. (2006). Korean-English biliteracy acquition: Cross language and orthography transfer. Journal of Educational Psycholgy, 98(1), 148–158.

Wang, M., Perfetti, C. A., and Liu, Y. (2005). Chinese-English biliteracy acquisition: Cross-language and writing system transfer. Cognition, 97, 67–88.