Orthographic learning and self-teaching in a bilingual and biliterate context

Journal of Experimental Child Psychology 117 (2014) 45–58

Contents lists available at ScienceDirect

Journal of Experimental Child Psychology journal homepage: www.elsevier.com/locate/jecp

Orthographic learning and self-teaching in a bilingual and biliterate context Mila Schwartz a,b,⇑, Janina Kahn-Horwitz a,b, David L. Share c a

Department of Research and Evaluation Authority, Oranim Academic College of Education, Tivon 36006, Israel Edmond J. Safra Brain Research Center for the Study of Learning Disabilities, Department of Learning Disabilities, Faculty of Education, University of Haifa, Haifa 31905, Israel c Department of Learning Disabilities, Faculty of Education, University of Haifa, Israel b

a r t i c l e

i n f o

Article history: Received 28 November 2012 Revised 13 August 2013 Available online 15 October 2013 Keywords: Biliterate Bilingualism English foreign language Orthographic learning Self-teaching Script dependence

a b s t r a c t The aim of this study was to examine self-teaching in the context of English as a foreign language literacy acquisition. Three groups comprising 88 sixth-grade children participated. The first group consisted of Russian–Hebrew-speaking bilinguals who had acquired basic reading skills in Russian as their first language (L1) and literacy and who were literate in Hebrew as a second language. The second group consisted of Russian– Hebrew-speaking bilinguals who had not learned to read in their native Russian but had acquired Hebrew as their first literate language. The third group consisted of Hebrew-speaking monolingual children who were literate in Hebrew. This design facilitated examining the effect of biliteracy and bilingualism on basic English reading skills. We hypothesized that due to the proximity between the Russian and English orthographies as opposed to the Hebrew–English ‘‘distance,’’ the Russian– Hebrew-speaking biliterate group who acquired basic reading and spelling skills in L1 Russian would have superior self-teaching in English as opposed to the two other groups. The standard two-session self-teaching paradigm was employed with naming (speed and accuracy) and orthographic choice as posttest measures of orthographic learning. Results showed that after 4 years of English instruction, all three groups showed evidence of self-teaching on naming speed and orthographic recognition. The Russian–Hebrew-speaking biliterate group, moreover, showed a partial advantage over the comparison groups for initial decoding of target pseudowords and clear-cut superiority for measures of later orthographic learning, thereby showing self-teaching while supporting the script dependence hypothesis. Ó 2013 Elsevier Inc. All rights reserved.

⇑ Corresponding author at: Department of Research and Evaluation Authority, Oranim Academic College of Education, Tivon 36003, Israel. Fax: +972 4 9539617. E-mail address: [email protected] (M. Schwartz). 0022-0965/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jecp.2013.08.008

46

M. Schwartz et al. / Journal of Experimental Child Psychology 117 (2014) 45–58

Introduction There is growing concern among social scientists that conclusions from studies based on highly educated populations from wealthy industrialized Western cultures (especially native speakers of English and other Indo-European languages) may have limited generalizability regarding human behavior in general (Henrich, Heine, & Norenzayan, 2010). These reservations also extend to the field of language and literacy (Beveridge & Bak, 2012; Evans & Levinson, 2009; Share, 2008a). Share (2008a), for example, argued that the idiosyncracies of English orthography have confined reading science to an insular Anglocentric research agenda addressing theoretical and applied issues with only limited relevance for a universalistic science of reading (see also Frost, 2012). Despite the fact that nearly all models of reading are exclusively Anglocentric, most languages are not like English (Evans & Levinson, 2009). Furthermore, more individuals in the world learn to read in additional language literacy than in their first language literacy (Gunderson, Odo, & D’Silva, 2011). Even in the English-speaking world, the majority of those learning to read English are no longer native English speakers (Crystal, 1997). Therefore, it behooves researchers to test their models and theories beyond the confines of monolinguals learning to read in their native tongue. The aim of the current study was to examine a popular theory of reading acquisition, the self-teaching hypothesis (Share, 1995, 2008b), among children who are acquiring literacy in a non-native tongue. In view of the fact that many studies of literacy learning in a foreign or additional language often confound bilingualism and biliteracy, our study sought to disentangle these two factors. We also wished to avoid the potential confounds associated with studying native and non-native languages that belong to the same language subfamily or share a common orthography. Our study focused on English because this is the most commonly learned non-native language in the world (Crystal, 2008). Our sample comprised English language learners who are confronted with the task of learning to read and write in a structurally unrelated language and dissimilar orthography (as are the majority of English language learners worldwide). Our study took into account the influence of both biliteracy and bilingualism in three distinct languages (English, Russian, and Hebrew) and three distinct orthographies (Roman, Cyrillic, and Hebrew). To this end, three groups were recruited. All three groups were learning to read and write in English in the same instructional context but differed in their prior language and literacy background. The first group comprised Russian–Hebrew-speaking bilinguals who had acquired basic reading skills in alphabetic (Cyrillic) Russian as their first language (L1) and literacy and who were literate in Hebrew as a second language (L2) and were in the process of acquiring English as a foreign language (EFL) literacy where English was the third sequential language (L3). The second group comprised Russian–Hebrew-speaking bilinguals who had not learned to read in their native Russian but had acquired Hebrew as their first literate language and were also in the process of acquiring EFL literacy in English as their second sequential language. The third group comprised Hebrewspeaking monolingual children who had acquired literacy in their native Hebrew and were in the process of acquiring EFL literacy in English, which was their second sequential language. This design enabled us to distinguish the effect of biliteracy and bilingualism. If bilingualism per se is the critical factor in English language learning, then both Russian–Hebrew bilingual groups (monoliterates and biliterates alike) will be superior to the Hebrew monolinguals. If biliteracy per se is the crucial ingredient, then only the biliterates will enjoy an advantage in English literacy learning compared with the two monoliterate groups (Russian–Hebrew bilinguals and Hebrew monolinguals). Returning to the overarching goals of this study, two issues were examined: (a) the self-teaching hypothesis in the context of learning to read in a non-native context and (b) the challenges of acquiring EFL literacy among non-native English language learners from different L1 orthographic backgrounds within the framework of the script-dependent hypothesis.

The self-teaching hypothesis To become a skilled reader, recognition of individual words must be fluent and near effortless in order to free the reader’s attention to meaning. It is widely agreed that the skilled reader is able to

M. Schwartz et al. / Journal of Experimental Child Psychology 117 (2014) 45–58

47

recognize words almost instantly on the basis of the printed word form (i.e., orthographic knowledge) without laboriously translating graphemes to phonemes and synthesizing these to form known pronunciations. Although skilled readers resort to this decoding mechanism only when encountering a new word, Share (1995) emphasized that every word is a novel letter string at some point in reading development. The self-teaching hypothesis maintains that to attain fluent word identification, beginner readers need efficient phonological recoding (i.e., decoding) to independently develop both wordspecific and general orthographic knowledge (Share, 1995, 2008b). In this way, phonological recoding functions as a self-teaching mechanism or built-in teacher. After a relatively small number of successful decoding attempts with a novel word, a child is able to rely on rapid memory retrieval mechanisms for word identification based on a newly acquired word-specific orthographic representation. Thus, the precondition for acquiring new orthographic representations using the self-teaching mechanism is accurate and fluent decoding of these words (Bowey & Muller, 2005; Share, 1999). A number of studies have examined the self-teaching hypothesis among L1 children (see Share, 2008b, for a review). These studies have used a similar experimental design. This design involves exposing children to pseudowords representing names of novel content words such as places, animals, and fruit embedded within short texts. After a break of between 3 days and 1 week, orthographic learning is typically assessed in the following ways: homophonic orthographic choice between target and control (foil) items (e.g., yait vs. yate as the name of a place; Cunningham, Perry, Stanovich, & Share, 2002), reading aloud lists of targets versus homophonic foils, and (in many studies) also spelling of targets (Shahar-Yames & Share, 2008). Evidence for L1 self-teaching has been obtained in English (e.g., Bowey & Miller, 2007; Bowey & Muller, 2005; Cunningham et al., 2002; Nation, Angells, & Castles, 2007), in Dutch (de Jong & Share, 2007), and in pointed (fully voweled) Hebrew (e.g., Share, 1999, 2004). Self-teaching occurs in both oral and silent reading (e.g., Bowey & Miller, 2007; Bowey & Muller, 2005; de Jong & Share, 2007). Orthographic learning is rapid and robust, emerging within a few exposures to a novel letter string (Share, 2004) and being maintained for a considerable period of time (weeks or months; Hogaboam & Perfetti, 1978; Share, 2004). There are several lines of evidence that support the self-teaching notion that orthographic learning is the product of decoding (phonological recoding). First, levels of orthographic learning appear to be closely tied to levels of decoding success, whether induced experimentally (Kyte & Johnson, 2006; Share, 1999) or naturally occurring as in the case of disabled readers (Ehri & Saltmarsh, 1995; Manis, 1985; Reitsma, 1983, 1989; Share & Shalev, 2004). Second, at the individual level, the data have been quite consistent in showing a significant positive association between target decoding success and orthographic learning—a relation that, furthermore, does not appear to be simply the by-product of the general relation between preexisting decoding ability and orthographic learning (Cunningham, 2006; Cunningham et al., 2002; Kyte & Johnson, 2006). Third, the role of context in reducing attention to orthographic detail (see, e.g., Cunningham, 2006; Ehri & Roberts, 1979; Ehri & Wilce, 1980) and spelling in enhancing it (Ouellette, 2010; Shahar-Yames & Share, 2008) also confirms the basic self-teaching premise that phonological recoding contributes to orthographic learning by drawing attention to letter detail and word-specific spelling– sound relations. As reviewed above, the role of self-teaching in orthographic learning has now been supported in a number of studies of L1 literacy acquisition (Cunningham et al., 2002; de Jong & Share, 2007). The question addressed in the current study was whether self-teaching extends beyond a child’s mother tongue.

Challenges of acquiring English as a foreign language English has spread throughout the world, capturing the status of the primary world language (Crystal, 2008). Israel is no exception in using English as a universal lingua franca. English, which does not have the status of an official language in Israel (as do Hebrew and Arabic), is widely used in higher education and in many other fields of life and is a mandatory subject taught from early elementary school (Spolsky & Shohamy, 1999). Success in English literacy acquisition is a precondition for academic studies in Israel. For many Hebrew-speaking students, however, English literacy acquisition constitutes a serious obstacle (Ressissi & Kahn-Horwitz, 2007). There are several reasons for this.

48

M. Schwartz et al. / Journal of Experimental Child Psychology 117 (2014) 45–58

First, there is wide agreement that English is an unusually challenging orthography to acquire due to its orthographic complexity even for native-speaking students (Seymour, Aro, & Erskine, 2003; Share, 2008a). Although there may be more than one dimension of orthographic complexity (see, e.g., Borgwaldt, Hellwig, & de Groot, 2005), English is a deep orthography primarily owing to the complexity of grapheme–phoneme correspondences, with multiple phonemic possibilities for nearly all graphemes that depend on complex positional and morphemic constraints. Thus, grapheme–phoneme translation alone might not ensure accurate pronunciation of a word (Coltheart, 1978, 2005). Additional morphological, orthographic, and contextual (semantic and syntactic) resources are necessary to ensure accurate decoding (Nation & Snowling, 1998, 2004; Ouellette, 2006; Ricketts, Nation, & Bishop, 2007; Tunmer & Chapman, 1998). Given that English as a deep orthography is especially challenging even for native English speakers (Seymour et al., 2003), it is even more challenging for non-native learners. In considering the non-native context of English literacy learning, it is noteworthy that vocabulary knowledge, in addition to phonological and orthographic knowledge, has been found to contribute significantly to orthographic learning (Ricketts, Bishop, Pimperton, & Nation, 2011), especially in the case of irregular words (Nation & Snowling, 1998; Ouellette & Beers, 2010; Ricketts et al., 2007). For foreign language literacy acquisition, phonological, orthographic, and vocabulary knowledge may be severely limited, particularly during the early stages of literacy acquisition when elementary school children are not studying EFL on a daily basis. In addition, EFL literacy acquisition may be challenging because of the limited scope of instruction (e.g., instructional methods, time allocated for instruction). Whereas L1 literacy instruction starts with home language and literacy input from the time a child is born and is an integral and daily part of the preschool and elementary curriculum, EFL literacy instruction in many countries, including Israel, is not scheduled on a daily basis and must compete with EFL oral acquisition. Finally, another consideration is the orthographic proximity between the native and non-native languages. This dimension is well conceptualized by the script-dependent hypothesis (Geva & Siegel, 2000; Geva & Wade-Woolley, 1998), which maintains that L1 orthographic structure and features will affect L2 literacy acquisition due to cross-linguistic transfer (as elaborated below). Most previous studies supported this hypothesis by focusing on examples of negative transfer from L1 to L2 English orthography within a ‘‘contrastive analysis’’ framework. This was based on the assumption that differences in grapho-phoneme and phoneme–grapheme mappings between first and second languages interfere with decoding and spelling in English (see Dressler & Kamil, 2006). However, Schwartz, Leikin, and Share (2005) demonstrated a positive effect of L1 literacy beyond the well-known deep–shallow contrast (i.e., the orthographic depth hypothesis; Frost, 2005). Specifically, the differences between a fully fledged (plene) alphabet as in Russian’s Cyrillic script and a primarily consonantal alphabet (or abjad) such as Hebrew’s pointed script can facilitate the spelling of certain distinctive features in Hebrew. The current study was designed to examine self-teaching within the constraints of script dependence. This might enable a better understanding of the positive effect of previous literacy experience in L2 on L2 literacy acquisition within the framework of the script-dependent hypothesis. The current study This study investigated self-teaching among sixth-grade EFL children who had 4 years of exposure to English. Previous studies on L1 self-teaching have investigated children in first, second, and third grades (Bowey & Muller, 2005; Cunningham et al., 2002; Shahar-Yames & Share, 2008; Share, 1999). A period of 1 to 3 years is considered a time length whereby intensive exposure to L1 phonology and orthography lays the foundation for effective orthographic learning. Participants in the current study had been exposed to the English language and orthography for 3 h per week over 4 years, and our first overall aim was to determine whether this period is sufficient to enable functional decoding and self-teaching. More specifically, we were interested in investigating the interaction between experience with different languages and orthographies and EFL orthographic learning. We examined the extent to which previous orthographic knowledge and oral knowledge in Russian as opposed to mostly oral language skills might contribute to English orthographic learning. This is because the Russian orthography is alphabetic (representing both consonants and vowels), making

M. Schwartz et al. / Journal of Experimental Child Psychology 117 (2014) 45–58

49

it more similar to the English orthography (as opposed to Hebrew, which is a consonantal writing system, or abjad, primarily representing consonants). Russian uses a Cyrillic alphabet that was westernized by Peter the Great, czar of Russia during the 18th century. The Russian alphabet has 13 uppercase letter shapes that, although similar to the Latin alphabet, have only 6 phonemes in common with English. There are 12 lowercase Russian letters (a, d, e, r, v, y, o, p, c, n, y, and [) that are similar in shape to English letters and that share the phonemes /a/, / k/, /m/, /o/, /s/, and /t/. The Russian and English orthographies have many other common characteristics, including a fully fledged alphabetic writing system (consonants and vowels are accorded the same graphemic status) and the facts that writing direction is from left to right and uppercase versus lowercase letters are distinguished (see Appendix A). All of these orthographic similarities would be expected to facilitate orthographic learning in English. In contrast to the Russian and English orthographies, the Hebrew orthography is a Semitic orthography that is primarily consonantal. In the pointed script, vowels in the form of diacritical marks appear within, above, or below the consonants (e.g., /dK/ as in the word /dKg/ ‘fish’). Pointed Hebrew consists mainly of a succession of syllable ‘‘blocks,’’ each comprising a consonant with attached vowel diacritic. In this way, pointed Hebrew resembles the akshara unit of South and South-East Asian alphasyllabaries (which developed from Semitic scripts via Brahmi). The fully vowelized Hebrew orthography is shallow in that all phonemic information is provided. Fully vowelized Hebrew appears in beginning reader texts, poetry, and biblical texts. During the first 3 years of elementary school, children are gradually exposed to more and more unpointed texts and are required to use orthographic, morphological, semantic, and contextual information in order to decode this deeper form of the Hebrew orthography. As opposed to the Russian and English orthographies, the Hebrew orthography has different (more square-shaped) letter shapes, is written from right to left, and does not differentiate between uppercase and lowercase letters. In our study, the sixth-grade Russian–Hebrew-speaking biliterates, the Russian–Hebrew-speaking monoliterates, and the Hebrew-speaking monoliterates were expected to be experienced readers in the pointed Hebrew script and were expected to transfer their orthographic skills from Hebrew (L1 or L2) to EFL. At the same time, the script-dependent hypothesis predicts that, owing to the proximity between the Russian and English orthographies as opposed to the Hebrew–English ‘‘distance,’’ the Russian–Hebrew-speaking biliterate group who acquired basic reading and spelling skills in L1 Russian would have superior self-teaching in EFL. The reason for their superiority is based on the head start that these children received in acquiring literacy in their L1, alphabetic Russian. Taking into account that support has been found for self-teaching in L1 English and L1 Hebrew, we sought to examine the self-teaching hypothesis within the context of acquiring a non-native orthography. Thus, we posed the following question: After 4 years of EFL study (3 h of study per week), is self-teaching observed and, if so, how does this mechanism operate within the context of EFL reading? In addition, we questioned whether different orthographic backgrounds influence self-teaching in EFL. On the basis of the script-dependent hypothesis, we predicted that Russian–Hebrew-speaking biliterates would show superiority on EFL self-teaching tasks due to their previous experience with an orthography that shares many features with the English orthography as opposed to their Russian–Hebrew-speaking and Hebrew-speaking monoliterate peers. In this case, the Russian–Hebrewspeaking monoliterate group who had only oral Russian experience was used as a control or comparison group. This design permitted us to distinguish the contribution of bilingualism per se (by comparing established bilinguals with monolinguals) as opposed to biliteracy (comparing the biliterates with the monoliterates).

Method Participants A sample of 88 sixth-grade children were recruited from a group of 99 fifth-grade children who participated in a previous study (Kahn-Horwitz, Schwartz, & Share, 2011). Eleven children who had switched schools were dropped from the original sample of 99. The participants came from five

50

M. Schwartz et al. / Journal of Experimental Child Psychology 117 (2014) 45–58

socioeconomically middle-rank elementary schools in the north of Israel. A school’s socioeconomic status (SES) index is based on parental reports of income, occupation, and ranking of family residential area and is rated on a 10-point scale ranging from 1 to 10, with higher scores indicating lower SES. In the current study, the SES index for the schools ranged from 4.7 to 5.1 (M = 4.96). The sample consisted of three groups of children. The first group comprised 31 Russian–Hebrewspeaking biliterates. Their Russian literacy skills were reported previously in Kahn-Horwitz and colleagues (2011). The second group comprised 16 Russian–Hebrew-speaking bilinguals with L1 oral Russian but only very rudimentary knowledge of the Russian alphabet. In other words, these children had knowledge of approximately half of the letter–sound correspondences of the Russian alphabet, knew a few letter names, and could write their first names in Russian. The reason for the smaller size of this group was that a majority of Russian–Hebrew-speaking children learn to read Russian. The third group was a monolingual Hebrew-speaking group (n = 41). Both groups of Russian–Hebrew-speaking participants were children of Russian Jewish immigrants. In response to their consent forms, which included a home language use questionnaire, parents uniformly reported that the dominant language between parents and children was Russian at home. Measures General nonverbal ability Raven’s Colored Matrices. Participants were required to choose one of five possible options (Set A, B, C, D, or E) to identify a missing segment in a geometric array (Raven, Raven, & Court, 1976). English phoneme analysis. Participants analyzed words into constituent phonemes (Kahn-Horwitz, Schwartz, & Share, 2009). The experimenter pronounced a target word, and participants repeated the word and were then asked to segment the word. Two examples were given before the 20 target monosyllabic items were presented. All items in this test had a closed CVC (consonant–vowel–consonant) syllabic structure. There were 5 CVC items (e.g., dog), 5 items with an initial CCVC consonant cluster (e.g., trip), 5 CVCC items with a final consonant cluster (e.g., milk), and 5 items with both initial and final consonant clusters (CCVCC; e.g., stand). Two scores were calculated. First, an overall score was computed by allocating 1 point for each correctly analyzed word (maximum score of 20 for this task). Second, 1 point was allocated for correctly analyzing medial vowels (maximum score of 20 for this task). Internal consistency (coefficient alpha) was .87. Self-teaching task We employed the standard two-session self-teaching paradigm (Share, 1999) with naming (speed and accuracy) and orthographic choice as posttest measures of orthographic learning. During the first session, children were exposed to 12 short texts. These texts were adapted to be suitable for EFL learners at the elementary level. Each text contained approximately five sentences each with a simple syntactic structure and basic vocabulary. Two versions of each text were created (see Appendix B for both versions), each with two alternate spellings of a target pseudoword. Thus, 12 pairs of pseudowords were presented across the 24 texts: 6 homophonic pairs such as roo/rew and 6 non-homophonic pairs (e.g., snup/frup) sharing the same rime pairs (e.g., -up) but different onset consonant clusters (e.g., sn- fr-) The reason why only half of the items were homophones was that prior work had shown that elementary school EFL learners have not yet acquired many of the more complex orthographic conventions, especially vowel digraphs that are commonly used to construct homophonic pairs in L1 self-teaching investigations. For each pair, half of the sample was exposed to one spelling (e.g., roo, snup) and the other half was exposed to the alternative spelling (e.g., rew, frup). These target pairs were based on prior work (Kahn-Horwitz et al., 2011) and were structured to comply with English orthographic conventions. The current study focused on a selection of orthographic conventions appearing in four elementary school English textbooks. Twelve experienced elementary school teachers ranked each orthographic convention on a scale of 1 (no exposure or instruction) to 5 (high levels of exposure and instruction). Spelling patterns assigned a ranking of 4 or 5 by at least 80% of teachers were included as target items for the current study. The following patterns were included: the vowel digraphs (oo as in roo, ew as in rew, ee as in een or jeet, ei as in ein, oa as in zoam, ow as in smown, ea as

M. Schwartz et al. / Journal of Experimental Child Psychology 117 (2014) 45–58

51

in jeat, and ai as in raim), consonant digraphs (sh as in shesk and ch as in charn), split digraphs (a–e as in rame and o–e as in smone or zome), closed syllables with short vowels (vaps, faps, frup, snup, hesk, shesk, sneg, freg, vab, and fab), and r-controlled syllables (ar as in harn or charn). The target homophones (containing these spelling patterns) and the non-homophonic onset-switched pairs were presented as novel letter strings representing fictitious names for places, names, animals, colors, and the like. Target pseudowords each appeared four times per text. Measures of posttest orthographic learning Measures of posttest orthographic learning were administered during a second session 1 week later. Naming. Children were presented with two printed lists of 12 pseudowords. One list was made up of the target pseudowords presented in randomized order. The parallel list consisted of the 12 alternative target words presented in the same way. Naming for each list was recorded on a voice file for later error analysis. Both accuracy and speed of reading was recorded. Orthographic choice. Children were presented with each of the six homophonic word pairs, one at a time. One member of the pair was the original target word spelling read 7 days earlier, and the other was the alternate homophonic spelling. Children were asked to identify the target spelling to which they had been exposed 1 week earlier (see examples above). The location of these alternatives was rotated across trials. This orthographic choice task did not include the 6 non-homophonic onset rime alternatives because these pairs were not phonologically equivalent. Procedure As noted above, all children participated in two sessions, each lasting approximately 40 min. During the initial presentation session, children completed the Raven’s Colored Matrixes and the phoneme analysis task and were exposed to the 12 texts with embedded target words. During the posttest orthographic learning session, children completed the two self-teaching tasks with order counterbalanced. Testing was conducted individually in a quiet room by EFL teachers who were studying in a graduate program. Instructions were administered in Hebrew to avoid misunderstandings. Results Background biosocial measures and nonverbal intelligence Table 1 presents the background data for biosocial, nonverbal ability, and phonemic analysis measured in English for the three groups. The role of bilingualism and biliteracy can be evaluated only if the groups are of comparable ability. No significant differences were found between groups in age, nonverbal IQ, or measures of English phoneme analysis. However, there was a significant gender difference, with boys outnumbering girls in the Russian–Hebrew-speaking monoliterate group and girls outnumbering boys in the Russian–Hebrew-speaking biliterate group, v2(1) = 7.69, p = .05. We return to this difference below. Target decoding during initial presentation Target decoding success is displayed in Table 2. It can be seen that overall performance levels were relatively low—approximately 50%. Analysis of variance (ANOVA) indicated significant between-group differences, F(2, 83) = 2.72, p < .05, and post hoc follow-up comparisons showing that the bilinguals were significantly more accurate than the Hebrew-speaking monolinguals. Although the biliterate bilinguals scored higher than the monoliterate bilinguals, this difference did not reach significance, t(1, 45) = 1.37, ns. The Russian–Hebrew-speaking monoliterate bilinguals and the Hebrew-speaking monolinguals obtained similar results.

52

M. Schwartz et al. / Journal of Experimental Child Psychology 117 (2014) 45–58

Table 1 Means and standard deviations for background biosocial factors and nonverbal intelligence. Variable

Russian–Hebrewspeaking biliterates (n = 31)

Russian–Hebrewspeaking monoliterates (n = 16)

Hebrew-speaking monolinguals (n = 39)

F

v2

Age (years:months) Gender (boys:girls) Nonverbal IQ (% success) English phoneme analysis (% overall success)

10:7 (0.53) 12:19 66.5 (18.27) 62.7 (29.87)

10:7 (0.45) 13:3 71.7 (14.18) 42.8 (34.25)

10:6 (0.53) 20:19 68.0 (11.19) 56.5 (32.27)

0.53 – 0.65 2.07

– 7.69* – –

Note. Standard deviations are in parentheses. p < .05.

*

Table 2 Means and standard deviations for initial target decoding and posttest orthographic learning. Variable

Russian–Hebrew speaking biliterates (n = 31)

Russian–Hebrew-speaking monoliterates (n = 16)

Hebrew-speaking monolinguals (n = 39)

Target

Non-target

Target

Non-target

Target

Non-target

First session Initial decoding (% success) Comprehension (% success)

55.4 (19.34) 83.6 (21.35)

– –

47.8 (15.09) 83.3 (17.48)

– –

44.9 (19.75) 84.0 (23.37)

– –

Second posttest session Naming accuracy (% success) Naming speed (s) Orthographic choicea (% success)

70.4a (16.51) 16.0a (5.82) 76.9a (15.32)

69.4a (18.05) 18.1a (6.42) –

58.9b (15.05) 20.7b (9.14) 61.5b (4.13)

58.3b (13.94) 21.1b (7.25) –

53.0b (22.25) 20.4b (8.30) 64.1b (24.64)

54.9b (19.56) 22.6b (8.66) –

Note. Standard deviations are in parentheses. Significant group differences are indicated by the use of superscript letters, with different superscripts indicating statistically significant differences; groups sharing superscripts are not significantly different from one another.

There were no significant differences among the three groups in reading comprehension, F(2, 83) = 0.01, ns. In addition, the analyses were rerun with only the boys to examine the potential confound of gender. We obtained a similar pattern of results, with no significant difference among the three groups for initial decoding accuracy, F(2, 42) = 2.14, ns, or for reading comprehension, F(2, 42) = 0.23, ns. Posttest orthographic learning For each of our posttest measures (naming accuracy, naming speed, and orthographic choice), we performed three separate (2  3) ANOVAs with repeated measures on the target/non-target items to examine the within-participant effect, and between-participant effect with planned analyses (comparison among Russian–Hebrew-speaking biliterates, Russian–Hebrew-speaking monoliterates, and Hebrew-speaking monolinguals on the same self-teaching measures), and finally the effect of interaction between self-teaching and group. Table 2 summarizes the children’s performance on all measures. The analysis revealed a significant main effect of orthographic learning (targets vs. non-targets) for naming speed, F(2, 83) = 10.33, p < .01, but no differences in accuracy, F(2, 83) = 0.02, ns. Overall, target pseudowords were read more quickly than homophonic non-target pseudowords, and this effect was consistent across groups, with the group by target interaction being nonsignificant, F(2, 83) = 1.14, ns. A significant between-group effect was observed for naming accuracy, F(2, 83) = 7.78, p < .001, and naming speed, F(2, 83) = 3.53, p < .05, with the biliterates outperforming both of the other two groups.

M. Schwartz et al. / Journal of Experimental Child Psychology 117 (2014) 45–58

53

No interaction was found between orthographic learning and group for either naming accuracy or naming speed. For orthographic choice, an ANOVA with a planned post hoc analysis showed that there was again evidence for orthographic learning among all three groups, with results reliably above the 50% chance level of success. In addition, the Russian–Hebrew-speaking biliterates demonstrated a significant advantage over the Russian–Hebrew-speaking bilingual and Hebrew-speaking monolingual groups, F(2, 83) = 3.97, p < .05. A significant advantage was found for the bilingual biliterates as opposed to the bilingual monoliterates, t(1, 45) = 2.33, p = .03, and the monolinguals, t(1, 70) = 2.80, p = .007, whereas no significant differences were found between the two monoliterate groups, t(1, 55) = 0.33, ns. We also ran analyses for boys only. Here too, the same pattern of results emerged as in the combined analysis. There was a significant main effect of orthographic learning (targets superior to nontargets) for naming speed, F(2, 42) = 4.20, p < .05, but not for accuracy, F(2, 42) = 1.50, ns. Orthographic choice in all three groups of boys was above the 50% chance level (Russian–Hebrew-speaking biliterates: M = 80.5, SD = 11.96; Russian–Hebrew-speaking bilinguals: M = 61.5, SD = 25.81; Hebrew-speaking monolinguals: M = 61.7, SD = 23.57). Discussion This study adds to the existing self-teaching literature in two ways. First, this study showed that after 4 years of EFL instruction, all three groups showed evidence of self-teaching on all measures of orthographic learning; targets were read more quickly than homophonic non-target spellings, and target spellings were recognized more accurately. To our knowledge, this is the first study to examine self-teaching in a non-native EFL/ESL (English as a second language) literacy learning context. Second, the Russian–Hebrew-speaking biliterate group showed a numerical superiority over the Russian–Hebrew-speaking monoliterates and a significant advantage over the Hebrew-speaking monolingual group for initial decoding of target pseudowords. In addition, as predicted by the script-dependent hypothesis, the Russian–Hebrew-speaking biliterates showed a significant advantage over the Russian–Hebrew-speaking monoliterates and the Hebrew-speaking monoliterates for measures of subsequent orthographic learning. The following discussion elaborates on these findings. Self-teaching in EFL The current study highlights two salient characteristics that differentiate L1 as opposed to EFL orthographic learning. First, in EFL, overall decoding accuracy for the target pseudowords was relatively low (50%) for all three groups; Cunningham and colleagues (2002) reported a figure of 74% for second-grade English L1 children. Despite this relatively poor decoding accuracy for all three groups, results in the current study showed significant self-teaching. This supports findings obtained by Share and Shalev (2004), who found that despite lower initial decoding accuracy results (68% decoding accuracy), L1 Hebrew-speaking second graders performed well above chance (74%) on orthographic choice after two exposures to the targets. In the current study, all three groups were also faster on posttest naming speed of target pseudowords. Russian–Hebrew-speaking biliterates’ success on the orthographic choice task, and the speed of naming target pseudowords compared with non-target homophonic foils, may partly reflect the application of a visual–orthographic learning strategy for acquiring new words in English. In this study, our results seem to point to evidence that when self-teaching in English as a foreign language takes place with a background of a similar orthography, this facilitates the acceleration of visual–orthographic learning strategies that contributes to self-teaching. Support for this finding can be found in Cunningham and colleagues’ (2002) finding that over and above decoding, orthographic processing also contributes independently to self-teaching. In the case of our Israeli EFL children, the results may reflect teaching practices whereby after initial acquisition of grapheme–phoneme correspondence of the English alphabet, less emphasis is placed on independent decoding and more emphasis is placed on a larger unit identification strategy (Kahn-Horwitz, Sparks, & Goldstein, 2012).

54

M. Schwartz et al. / Journal of Experimental Child Psychology 117 (2014) 45–58

To summarize, EFL orthographic learning was observed in the current study despite relatively low levels of initial decoding. Low decoding accuracy can possibly be explained by the complexities of the English spelling system as well as possible instructional limitations in EFL settings as opposed to L1 literacy instruction. It could be that in the case of acquiring a more transparent L2 orthography such as Spanish or Italian, orthographic learning may be more rapid. In a recent self-teaching study by Van Daal, Rønneberg, and Wass (2013), readers of Danish, a relatively deep orthography, evinced poorer target decoding and lower subsequent levels of orthographic learning compared with readers of Norwegian and Swedish, which are much more consistent orthographies. Biliteracy versus bilingualism No significant differences were found between the groups on initial decoding accuracy and reading comprehension. This was strengthened when analyses were rerun with the boys only to examine the potential confound of gender. Here, a similar pattern of results was found, with no significant differences among the three groups for either initial decoding accuracy or reading comprehension. However, our findings pointed to numerical superiority of the Russian–Hebrew-speaking biliterates over the Russian–Hebrew-speaking bilingual monoliterates and significant superiority of the Russian–Hebrew-speaking biliterates over the Hebrew-speaking monolinguals on initial target pseudoword decoding. This superiority was also significant for all three self-teaching measures. Over and above the extreme difficulty of accurate English as a foreign language decoding after 3 years of English instruction, a possible explanation for the lack of a statistically significant advantage for initial decoding between the Russian–Hebrew-speaking biliterates and the Russian–Hebrewspeaking monoliterates could be a result of the relatively small numbers of participants in the Russian–Hebrew-speaking monoliterate group, which is a limitation that could be addressed in future studies. Adding participants to this Russian–Hebrew-speaking monoliterate group might have resulted in a significant difference in initial pseudoword decoding. At the same time, the orthographic learning measures showed clear-cut advantages of the Russian–Hebrew-speaking biliterates over the two monoliterate groups and, as a result, showed a self-teaching effect. This superiority suggests that experience in another alphabetic orthography provides a head start in orthographic learning in English. These findings are consistent with the script-dependent hypothesis (Geva & Siegel, 2000; Geva & Wade-Woolley, 1998), which explains the advantages in both target decoding and subsequent self-teaching as a result of the orthographic proximity between Russian and English, in contrast to between Hebrew and English (e.g., letter architecture, writing direction, full vowel representation). This explanation is strengthened by the fact that the Russian–Hebrew-speaking monoliterates were no different from the Hebrew-speaking monoliterates on any measures of selfteaching. In other words, it was not oral (bilingual) skills that made the difference; rather, it was the literacy experience with the Russian alphabetic script that resulted in a higher level of orthographic learning in English for the Russian–Hebrew-speaking biliterates. Additional support for the script-dependent and orthographic proximity hypotheses was reported in a recent study (Faris, 2012) whereby Arab–Hebrew-speaking biliterates showed a linguistic and orthographic advantage over Hebrew-speaking monoliterates in their decoding and spelling of English pseudowords that included both short and long vowel sounds. The Arabic orthography represents both long and short vowel sounds, whereas this distinction does not exist in Hebrew. Our data are also consistent with Cunningham and colleagues’ (2002) claim that orthographic processing is an independent factor contributing to self-teaching. In other words, children who have acquired the orthographic conventions of alphabetic Russian find it easier to learn alphabetic English and show superior orthographic learning in English. Further support for this assumption comes from a Dutch study by Bekebrede, van der Leij, and Share (2009), who investigated the role of native language orthographic abilities in EFL. The researchers compared a subgroup of dyslexic students with stronger orthographic abilities (in their native Dutch) with a subgroup of dyslexics with weaker orthographic scores. Both groups had comparable difficulties with phonological recoding. The group with stronger orthographic skills experienced greater success with English reading, and this was explained by these students’ ability to process larger units in the English orthography.

M. Schwartz et al. / Journal of Experimental Child Psychology 117 (2014) 45–58

55

To recap, this first study in English as a foreign language indicates that, as in the case of L1 (e.g., English, Hebrew, Dutch), evidence was found for orthographic learning. In addition, results from the current study point to a slower and more laborious process for EFL orthographic learning. Part of the problem, as alluded to in the Introduction, may also stem from the non-native speakers’ impoverished vocabulary and, specifically, the lack of well-specified phonological and semantic representations that may help to secure spellings, and irregular spellings in particular, in long-term memory (Hilte & Reitsma, 2011; McKay, Davis, Savage, & Castles, 2008; Nation & Snowling, 1998, 2004; Ouellette & Beers, 2010; Ricketts et al., 2007; Wang, Castles, Nickels, & Nation, 2011). A growing number of studies of orthographic learning (in L1 English) have now demonstrated that phonological knowledge aids orthographic learning, although evidence for the role of semantic knowledge over and above phonology is still mixed (Chalmers & Burt, 2008; Cunningham, 2006; Duff & Hulme, 2012; Hilte & Reitsma, 2011; McKague, Pratt, & Johnston, 2001; Nation et al., 2007; Ouellette, 2010; Ouellette & Fraser, 2009). It may be the case that phonological (and perhaps semantic) knowledge may be especially important for English reading development given the complexities of English letter–sound correspondence and the large number of exceptions that oblige readers to rely on additional sources of information to bridge the gap between a partially decoded or approximate form and a known pronunciation (Elbro, de Jong, Houter, & Nielsen, 2012; Share, 1995; Tunmer & Chapman, 2012). Here, knowledge of the form of the spoken words may be crucial, and of course the non-native reader is at a considerable disadvantage owing not only to a smaller phonological lexicon and but probably also to imprecise forms of even known words.

Limitations and future research directions Further studies integrating L2 self-teaching and orthographic proximity should include a posttest pseudoword spelling task as a measure of orthographic learning. An additional measure in future studies would be an L2 vocabulary measure that could assist in understanding the link between vocabulary and self-teaching in a second language. An additional direction for future research is to explore the interaction between L2 self-teaching and orthographic proximity within different language and literacy dyads and triads. It would also be of interest to examine foreign language self-teaching in the context of two shallow Romanbased orthographies such as Spanish and Italian or two Semitic orthographies such as Arabic and Hebrew. This would enable an understanding of self-teaching that extends beyond the limitations of the deep English orthography with its complex grapheme–phoneme correspondence (Share, 2008a). All of the above would broaden our understanding of the self-teaching mechanism in a second language. To further investigate evidence regarding the linguistic and orthographic proximity hypothesis, it would be of value to examine diverse language and literacy combinations that are linguistically and orthographically distant to differing extents such as Japanese versus Russian L1 bilingual biliterates versus Japanese and Russian L1 bilingual monoliterates. A research design of this nature would enable a stronger claim to be made for the role of linguistic and orthographic proximity versus the general advantage of being biliterate before literacy acquisition in the third language. In sum, this first study of L2 English self-teaching supports the script-dependent hypothesis owing to the fact that a different magnitude of orthographic learning was shown for children with different literacy backgrounds. Prior experience with an orthography that is more similar to English seems to facilitate orthographic learning as opposed to an orthography that is more remote from English.

Acknowledgments This research was jointly funded a grant from the MOFET Institute and the Department of Teacher Education at the Ministry of Education, Israel, awarded to the second author (2009–2010) and by a grant from the Oranim College M.Ed. Program awarded to the first two authors.

56

M. Schwartz et al. / Journal of Experimental Child Psychology 117 (2014) 45–58

Appendix A. Differences and similarities in orthographic characteristics in Russian, English, and Hebrew orthographies.

Orthographic convention or pattern

English

Russian

Hebrew

Letter shapes

13 shared letter shapes Left to right

13 shared letter shapes Left to right

X

Writing direction Cursive writing Uppercase and lowercase letters All vowels represented by letters

John

Mbka

Right to left X X Some vowels represented by letters

Appendix B. Texts for testing the self-teaching paradigm with target words italicized 1. This is the Snup/Frup family. The Snup/Frup family lives in a house in the country. There are three children in the Snup/Frup family, two girls and a boy. The Snup/Frup family has a pet cat. 2. Freg/Sneg likes shopping. It is five o’clock and Freg/Sneg is in the candy shop. Freg/Sneg is in the book shop at six o’clock. Freg/Sneg buys two new books and then she sees her friend, Tamar. Tamar says, ‘‘Hello Freg/Sneg’’. 3. A chesk/hesk is a fruit that you find in South America. A chesk/hesk is big, yellow, and sweet. Children in South America take a chesk/hesk with them to school to eat with lunch. Last week I saw a chesk/hesk in a supermarket. 4. A Jeet/Jeat is a small car. A Jeet/Jeat has four doors and it doesn’t use gas. A Jeet/Jeat does not cost a lot of money. Many people want to buy a Jeet/Jeat because it saves money and it keeps the air clean. 5. The een/ein is a big animal that lives in the jungle. The een/ein lives in groups and eats plants and leaves from the trees. The een/ein is a friendly animal. The een/ein lives for up to 30 years. 6. Zome/Zoam is a small city in Canada. It is very cold in Zome/Zoam, and people who live in Zome/ Zoam wear very warm clothes. In the short summer, people from Zome/Zoam swim outdoors every day. 7. My favorite TV program is Roo/Rew. Roo/Rew is about three children who travel around Africa. Every week there is a new adventure on Roo/Rew. Roo/Rew is on TV at seven o’clock on Monday evenings. 8. The color rame/raim is beautiful. The color rame/raim is orange–red and looks like fire. I have a rame/raim shirt. I see many rame/raim clothes in the shops now. 9. Every summer holidays we visit Smone/Smown. Smone/Smown is a small town near Eilat. It is very hot in Smone/Smown. We stay awake at night and sleep during the day. It gets dark very late in Smone/Smown. 10. Faps/Vaps are pretty flamingos, and they live near the water. Faps/Vaps are tall and pink. Faps/ Vaps run very fast. Faps/Vaps drink water and eat small fish. 11. Vab/Fab is in a band. She plays a guitar and sings. Vab/Fab sings many songs. Vab/Fab has a new CD. Lots of children see Vab/Fab singing on TV. 12. Harn/Charn likes to help at home. Harn/Charn feeds the dog. Harn/Charn waters the flowers. Harn/Charn makes his bed, and then he rides his bike with his friends.

M. Schwartz et al. / Journal of Experimental Child Psychology 117 (2014) 45–58

57

References Bekebrede, J. I., van der Leij, A., & Share, D. L. (2009). Dutch dyslexic adolescents: The phonological core variable orthographic difference model. Reading and Writing, 22, 133–165. Beveridge, M. E. L., & Bak, T. H. (2012). Beyond one-way streets: The interaction of phonology, morphology, and culture with orthography. Behavioral and Brain Sciences, 35, 18–19. Borgwaldt, S. R., Hellwig, F. M., & de Groot, A. M. B. (2005). Onset entropy matters: Letter-to-phoneme mappings in seven languages. Reading and Writing, 18, 211–229. Bowey, J. A., & Miller, R. (2007). Correlates of orthographic learning in third-grade children’s silent reading. Journal of Research in Reading, 30, 115–128. Bowey, J. A., & Muller, D. (2005). Phonological recoding and rapid orthographic learning in third-graders’ silent reading: A critical test of the self-teaching hypothesis. Journal of Experimental Child Psychology, 92, 203–219. Chalmers, K. A., & Burt, J. S. (2008). Phonological and semantic information in adults’ orthographic learning. Acta Psychologica, 128, 162–175. Coltheart, M. (1978). Lexical access in simple reading tasks. In G. Underwood (Ed.), Strategies of information processing (pp. 151–216). London: Academic Press. Coltheart, M. (2005). Modeling reading: The dual-route approach. In M. J. Snowling & C. Hulme (Eds.), The science of reading: A handbook (pp. 6–23). Oxford, UK: Blackwell. Crystal, D. (1997). English as a global language. Cambridge, UK: Cambridge University Press. Crystal, D. (2008). Two thousand million? English Today, 24(1), 3–6. Cunningham, A. E. (2006). Accounting for children’s orthographic learning while reading text: Do children self-teach? Journal of Experimental Child Psychology, 95, 56–77. Cunningham, A. E., Perry, K. E., Stanovich, K. E., & Share, D. L. (2002). Orthographic learning during reading: Examining the role of self-teaching. Journal of Experimental Child Psychology, 82, 185–199. de Jong, P. F., & Share, D. L. (2007). Orthographic learning during silent and oral reading. Scientific Studies of Reading, 11, 55–71. Dressler, C., & Kamil, M. (2006). First- and second-language literacy. In D. August & T. Shanahan (Eds.), Developing literacy in second-language learners: A report of the National Literacy Panel on Language-Minority Children and Youth (pp. 197–238). Mahwah, NJ: Lawrence Erlbaum. Duff, F. J., & Hulme, C. (2012). The role of children’s phonological and semantic knowledge in learning to read words. Scientific Studies of Reading, 16, 504–525. Ehri, L. C., & Roberts, K. (1979). Do beginners learn printed words better in context or isolation? Child Development, 50, 675–685. Ehri, L., & Saltmarsh, J. (1995). Beginning readers outperform older disabled readers in learning to read words by sight. Reading and Writing, 7, 295–326. Ehri, L. C., & Wilce, L. S. (1980). The influence of orthography on readers’ conceptualization of the phonemic structure of words. Applied Psycholinguistics, 1, 371–385. Elbro, C., de Jong, P. F., Houter, D., & Nielsen, A.-M. (2012). From spelling pronunciation to lexical access: A second step in word decoding? Scientific Studies of Reading, 16, 341–359. Evans, N., & Levinson, S. C. (2009). The myth of language universals: Language diversity and its importance for cognitive science. Behavioral and Brain Sciences, 32, 429–448. Faris, R. (2012). English spelling and decoding acquisition amongst students from Semitic orthographic backgrounds (Arabic and Hebrew) (M.Ed. thesis). Israel: Oranim Academic College of Education. Frost, R. (2005). Orthographic systems and skilled word recognition processes in reading. In M. J. Snowling & C. Hulme (Eds.), The science of reading: A handbook (pp. 272–295). Malden, MA: Blackwell. Frost, R. (2012). Toward a universal model of reading. Behavioral and Brain Sciences, 35, 1–67. Geva, E., & Siegel, L. S. (2000). Orthographic and cognitive factors in the concurrent development of basic reading skills in two languages. Reading and Writing, 12, 1–30. Geva, E., & Wade-Woolley, L. (1998). Component processes in becoming English–Hebrew biliterate. In A. Y. Durgunoglu & L. Verhoeven (Eds.), Literacy development in a multilingual context (pp. 85–110). Mahwah, NJ: Lawrence Erlbaum. Gunderson, L., Odo, D. M., & D’Silva, R. (2011). Second language literacy. In E. Hinkel (Ed.). Handbook of research in second language teaching and learning (Vol. 2, pp. 472–487). New York: Routledge. Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the world? Behavioral and Brain Sciences, 33, 61–135. Hilte, M., & Reitsma, P. (2011). Activating the meaning of a word facilitates the integration of orthography: Evidence from spelling. Journal of Research in Reading, 34, 333–345. Hogaboam, T. W., & Perfetti, C. A. (1978). Reading skill and the role of verbal experience in decoding. Journal of Educational Psychology, 70, 717–729. Kahn-Horwitz, J., Schwartz, M., & Share, D. L. (2009). Phoneme analysis. Israel: Unpublished test, Oranim Academic College of Education. Kahn-Horwitz, J., Schwartz, M., & Share, D. L. (2011). Acquiring the complex English orthography: A tri-literacy advantage? Journal of Research in Reading, 34, 136–156. Kahn-Horwitz, J., Sparks, R. L., & Goldstein, Z. (2012). English as a foreign language spelling development: A longitudinal study. Applied Psycholinguistics, 33, 343–363. Kyte, C. S., & Johnson, C. J. (2006). The role of phonological recoding in orthographic learning. Journal of Experimental Child Psychology, 93, 166–185. Manis, F. R. (1985). Acquisition of word identification skills in normal and disabled readers. Journal of Educational Psychology, 77, 78–90. McKague, M., Pratt, C., & Johnston, M. B. (2001). The effect of oral vocabulary on reading visually novel words: A comparison of dual-route-cascaded and triangle frameworks. Cognition, 80, 231–262. McKay, A., Davis, C., Savage, G., & Castles, A. (2008). Semantic involvement in reading aloud: Evidence from a nonword training study. Journal of Experimental Psychology: Learning, Memory, and Cognition, 34, 1495–1517.

58

M. Schwartz et al. / Journal of Experimental Child Psychology 117 (2014) 45–58

Nation, K., Angells, P., & Castles, A. (2007). Orthographic learning via self-teaching in children learning to read English: Effects of exposure, durability, and context. Journal of Experimental Child Psychology, 96, 71–78. Nation, K., & Snowling, M. (1998). Semantic processing and the development of word-recognition skills: Evidence from children with reading comprehension difficulties. Journal of Memory and Language, 39, 85–101. Nation, K., & Snowling, M. J. (2004). Beyond phonological skills: Broader language skills contribute to the development of reading. Journal of Research in Reading, 27, 342–356. Ouellette, G. P. (2006). What’s meaning got to do with it: The role of vocabulary in word reading and reading comprehension. Journal of Educational Psychology, 98, 554–566. Ouellette, G. P. (2010). Orthographic learning in learning to spell: The roles of semantics and type of practice. Journal of Experimental Child Psychology, 107, 50–58. Ouellette, G., & Beers, A. (2010). A not-so-simple view of reading: How oral vocabulary and visual-word recognition complicate the story. Reading and Writing, 23, 189–208. Ouellette, G., & Fraser, J. R. (2009). What exactly is a yait anyway?: The role of semantics in orthographic learning. Journal of Experimental Child Psychology, 104, 239–251. Raven, J., Raven, J. C., & Court, J. H. (1976). Raven’s Progressive Matrices. Oxford, UK: Oxford Psychologists Press. Reitsma, P. (1983). Printed word learning in beginning readers. Journal of Experimental Child Psychology, 36, 321–339. Reitsma, P. (1989). Orthographic memory and learning to read. In P. G. Aaron & R. M. Joshi (Eds.), Reading and writing disorders in deferent orthographic systems (pp. 51–73). Dordrecht, The Netherlands: Kluwer Academic. Ressissi, N., & Kahn-Horwitz, J. (2007). A multi-year intervention program and its impact on young English-as-an-additionallanguage students in three municipal outlying areas [in Hebrew]. Tivon, Israel: Oranim Research and Evaluation Authority. Ricketts, J., Bishop, D. V. M., Pimperton, H., & Nation, K. (2011). The role of self-teaching in learning orthographic and semantic aspects of new words. Scientific Studies of Reading, 15, 47–70. Ricketts, J., Nation, K., & Bishop, D. V. M. (2007). Vocabulary is important for some, but not all, reading skills. Scientific Studies in Reading, 11, 235–257. Schwartz, M., Leikin, M., & Share, D. L. (2005). Bi-literate bilingualism versus mono-literate bilingualism: A longitudinal study of reading acquisition in Hebrew (L2) among Russian-speaking (L1) children. Written Language and Literacy, 8, 179–207. Seymour, P. H. K., Aro, M., & Erskine, J. M. (2003). Foundation literacy acquisition in European orthographies. British Journal of Psychology, 94, 143–174. Shahar-Yames, D., & Share, D. L. (2008). Spelling as a self-teaching mechanism in orthographic learning. Journal of Research in Reading, 31, 22–39. Share, D. L. (1995). Phonological recoding and self-teaching: Sine qua non of reading acquisition. Cognition, 55, 151–218. Share, D. L. (1999). Phonological recoding and orthographic learning: A direct test of the self-teaching hypothesis. Journal of Experimental Child Psychology, 72, 95–129. Share, D. L. (2004). Orthographic learning at a glance: On the time course and developmental onset of self-teaching. Journal of Experimental Child Psychology, 87, 267–298. Share, D. L. (2008a). On the Anglocentricities of current reading research and practice: The perils of over-reliance on ‘‘outlier’’ orthography. Psychological Bulletin, 134, 584–616. Share, D. L. (2008b). Orthographic learning, phonology, and the self-teaching hypothesis. In R. Kail (Ed.). Advances in child development and behavior (Vol. 36, pp. 31–82). Amsterdam: Elsevier. Share, D. L., & Shalev, C. (2004). Self-teaching in normal and disabled readers. Reading and Writing, 17, 1–31. Spolsky, B., & Shohamy, E. (1999). The languages of Israel: Policy, ideology, and practice. Clevedon, UK: Multilingual Matters. Tunmer, W., & Chapman, J. W. (1998). Language prediction skills, phonological recoding ability, and beginning reading. In R. M. Joshi & C. Hulme (Eds.), Reading and spelling: Development and disorders (pp. 33–67). Mahwah, NJ: Lawrence Erlbaum. Tunmer, W. E., & Chapman, J. W. (2012). Does set for variability mediate the influence of vocabulary knowledge on the development of word recognition skills? Scientific Studies of Reading, 16, 122–140. Van Daal, V., Rønneberg, V., & Wass, M. (2013). Effects of orthographic depth on learning to read and spell in a first and in a second language: A ‘‘natural’’ Scandinavian experiment. Developmental Psychology . submitted for publication. Wang, H.-C., Castles, A., Nickels, L., & Nation, K. (2011). Context effects on orthographic learning of regular and irregular words. Journal of Experimental Child Psychology, 109, 39–57.