Young Spanish-English Language Learners' Cognate ... · gual Hmong–English preschoolers (ages 3;4–5;2 [years;months]). They administered a 50-item recep-tive picture identification

Leacox et al.: Young Spanish–English Language Learners’ Cognate Facilitation 115

ABSTRACT: Purpose: Cognates are words that share meaning and form, where the translation-equivalent pairs between languages are phonologically similar (e.g., baby–bebé and telephone–teléfono). Noncog-nates are word pairs that share meaning but not form (e.g., bear–oso). The literature strongly establishes a cognate advantage with bilingual adults, although evidence for younger bilinguals is still emerging. This study investigated young Spanish-speaking English language learners’ (ELLs’) picture naming of cognates and noncognates. Method: Thirty-one Spanish-speaking ELL children completed a picture-naming task in English and in Spanish, in a counterbalanced order, to compare per-formance on cognates and noncognates. Data analysis using a repeated measures analysis of variance com-pared cognate and noncognate accuracy in English and Spanish. The number of translation-equivalent pairs

I

Contemporary Issues In CommunICatIon sCIenCe and dIsorders • Volume 43 • 115–128 • Spring 2016 © NSSLHA 1092-5171/16/4301-0115

Young Spanish–English Language Learners’ Cognate Facilitation on Picture Naming

Lindsey LeacoxUniversity of Northern Iowa, Cedar Falls

Carla WoodGretchen SundermanChristopher SchatschneiderFlorida State University, Tallahassee

n speech-language pathology, the term cog-nate has more than one meaning; for this bi-lingual investigation, the definition of interest

between cognates and noncognates (i.e., named in both languages) was also compared. Results: Young ELL children demonstrated higher naming accuracy on phonologically similar cognates than on phonologically dissimilar noncognates, which is similar to the cognate advantage found with bilingual adults and with young bilingual chil-dren on receptive tasks. Additionally, more transla-tion-equivalent pairs were found with cognates than with noncognates. Conclusion: Findings suggest that the cognate status of words should be considered for assessment prac-tices and clinical decision making. Future research should examine additional cognate measures and individual factors that influence cognate facilitation.

KEY WORDS: bilingualism, assessment, phonology, language

classifies cognates as words with shared meaning and linguistic similarity across languages (Harley, 2008). An example in English and Spanish is the cognate

word pair baby–bebé, which shares the concept of an “infant” as well as linguistic similarity in phonology (i.e., sounds produced), orthography (i.e., spelling pat-terns), and morphology (i.e., word structure).

In the average educated adult’s vocabulary, English and Spanish share approximately 10,000 to

116 Contemporary Issues In CommunICatIon sCIenCe and dIsorders • Volume 43 • 115–128 • Spring 2016

15,000 cognates, estimated as one-third to one-half of one’s vocabulary (Thomas, Nash, Thomas, & Rich-mond, 2005). In contrast, noncognates share meaning, but the translation pairs share less linguistic similar-ity (e.g., bear–oso), and false cognates share lexical similarity but have different semantic meanings (e.g., embarrassed–embarazada, in Spanish means pregnant).

Only 4.5% of the 161,163 American Speech- Language-Hearing Association (ASHA) members reported meeting the definition of being bilingual (ASHA, 2014), but at least 60% of school speech-language pathologists (SLPs) reported having children who are English language learners (ELLs) on their caseload (ASHA, 2010). Thus, SLPs may be interested in cognates due to the cross-language interaction and linguistic similarity that cognate pairs provide.

Cognates have received increased attention in the bilingual literature for their cross-linguistic overlap and their potential for cognate facilitation. Cognate facilitation is when the linguistic features of a word in one language influence and/or activate the features of a word in another language (Hoshino & Kroll, 2008; Sánchez-Casas & García-Albea, 2005). For this reason, bilingual or second language–learning adults recognize cognates quicker and with fewer errors than they do noncognates (Sunderman & Schwartz, 2008). Mono-lingual adults do not demonstrate cognate facilitation because only one language is present (Colomé, 2001).

Most evidence of cognate facilitation is based on studies of bilingual adults or older school-age chil-dren. Adults demonstrate cognate facilitation as both languages simultaneously activate regardless of the language presented during tasks (Blumenfeld & Mar-ian, 2007; Hermans, Bongaerts, de Bot, & Schreuder, 1998; Kroll, Gerfen, & Dussias, 2008). For older children, providing specific instruction to recognize cognates may increase word learning (Nagy, García, Durgunoğlu, & Hancin-Bhatt, 1993; Proctor & Mo, 2009). Bilingual adults and older elementary students may use the combined phonological and orthographic similarities of cognates to facilitate reading compre-hension (Cunningham & Graham, 2000; Malabonga, Kenyon, Carlo, August, & Louguit, 2008), whereas younger children may rely more on just the phono-logical properties of cognates (Kohnert, Windsor, & Miller, 2004; Pérez, Peña, & Bedore, 2010). Over-all, cognates’ linguistic overlap may increase word retrieval for bilingual individuals or second-language learners (Costa, Santesteban, & Caño, 2005; Hoshino & Kroll, 2008).

Despite the substantial research on cognates with adults and older children, additional research is warranted to investigate young bilingual children’s performance on cognates. The purpose of this study was to investigate young ELL children’s ability to

express phonologically similar cognates as compared to phonologically dissimilar noncognates. Exploring phonological overlap differences with young bilingual children has implications to guide word selection for assessment and potential utility in treatment.

Theoretical Models of CognatesA theoretical model that is sensitive to cognates would need to include a conceptual–semantic level and a lexical level (two labels: language A, language B) as well as a phonological level for similarity between words. As a basic framework, a monolin-gual learner knows one vocabulary concept and one name for that concept. For the child who is acquiring another language, Cummins (1981) explained how the word’s concept is the common underlying proficiency (CUP) or prior conceptual knowledge that facilitates the acquisition of another label in a different lan-guage for that same concept. If a Spanish-speaking child knows the concept of an orange and its Spanish label of naranja, the CUP is the foundation for learn-ing a second label in English; that is, orange.

Cummins’s (1981) model provides a rationale for cognate facilitation in relation to the overlapping semantic–lexical representation; however, the phono-logical overlap is not specifically addressed. The word production model (Garrett, 1975, 1976; Levelt, 1989) represents the activation of a word’s concept, the se-mantic label, and corresponding phonology across mul-tiple stages for bilingual adults. For example, in nam-ing pictures, the first stage involves concept activation; successive stages involve the retrieval of a spoken word with its syntactic and phonological properties.

It is hypothesized that the interaction among these word production stages is not separate like stair steps, but rather is overlapping and interactive among stages, similar to cascades or a waterfall (Humphreys, Rid-doch, & Quinlen, 1988). For adult bilingual speakers, cognates have cascaded activation among the stages, which influences word production when sounds are similar or overlap between languages. Even in lan-guages where the orthographic script is different (e.g., Spanish–English and Japanese–English), bilingual individuals naming pictures of cognates demonstrated activation across languages as compared to noncog-nates (Hoshino & Kroll, 2008). Costa, Caramazza, and Sebastian-Galles (2000) found that Catalán-Spanish adult bilinguals named pictures of cognates (e.g., gato–gat [cat]) more rapidly than they named pictures of noncognates (e.g., perro–gos [dog]), and this cognate effect disappeared when testing a monolingual control group. Costa et al. hypothesized that the overlapping phonology of the words in two languages permits quicker access. Perhaps not all


bilingual children will demonstrate a cognate effect, but an interacting bilingual system could lead to a child’s vocabulary containing more cognates because the overlapping phonology facilitates learning and ac-quisition in both languages due to the similar sounds (Kelley & Kohnert, 2012).

Receptive VocabularyMeasuring receptive vocabulary is a common pro-cedure used with young children. The vocabulary inventory of bilingual infants and toddlers contains either “doublets” (i.e., translation equivalents in both languages) or “singlets” (i.e., label only in one lan-guage). As an example of a doublet, a child produces the words perro and dog in each language; however, a child may only label oso as a Spanish singlet and not yet know the English equivalent of bear. When calculating vocabulary, a composite or conceptual score (Bedore, Peña, García, & Cortez, 2005; Pear-son, Fernández, & Oller, 1993) is recommended as a more accurate bilingual representation of the child’s total number of concepts across languages as com-pared to a single monolingual score.

In a study by Marchman and Martínez-Sussman (2002), Spanish–English toddlers demonstrated that a substantial amount of their vocabulary may be spe-cific to one language. Pearson, Fernández, and Oller (1995) found that approximately 30% of bilingual toddlers’ vocabulary words were doublets, with cog-nates comprising 20% of the doublets, or 7% of their total vocabulary. Peña, Bedore, and Zlatic-Giunta (2002) found that in a structured word generation task, older bilingual children (M = 6.5 years) had more doublets than did younger bilingual children (M = 5.1 years).

Kan and Kohnert (2005) investigated the recep-tive and expressive vocabulary skills of 19 bilin-gual Hmong–English preschoolers (ages 3;4–5;2 [years;months]). They administered a 50-item recep-tive picture identification task and a separate 50-item expressive picture-naming task of developmental words (from Fenson et al., 1993), using the same items presented across both languages in a coun-terbalanced order. The authors analyzed separate language scores in Hmong and English as well as a composite score and found that the older children outperformed the younger children in English but not in Hmong, and all of the children’s composite scores were greater than their single language scores.

Kan and Kohnert’s (2005) picture-naming results also showed that the doublet-to-singlet ratio increased with age, where younger children (M = 3;11, SD = 0;4) had 90% singlets and 10% doublets/translation equivalents, and older children (M = 5;0, SD = 0;2)

had 77% singlets and 23% doublets. The authors stated that the early acquired words were selected from developmental inventories, but they did not re-port whether any Hmong–English translation equiva-lents were cognates or not (Kan & Kohnert, 2005); cognates may not exist between Hmong and English (Vang, 2005).

Several studies have used receptive vocabulary tests to investigate performance differences based on words’ cognate status. In a preliminary investiga-tion, Umbel, Pearson, Fernández, and Oller (1992) compared bilingual first graders’ performance on Spanish and English items from two standardized receptive vocabulary tests: the Test de Vocabulario en Imágenes de Peabody (TVIP–H Spanish; Dunn, Padilla, Lugo, & Dunn, 1986) and the Peabody Picture Vocabulary Test—Revised Edition (PPVT–R, English: Dunn & Dunn, 1981). The bilingual stu-dents performed similarly on the English–Spanish cognate pairs compared to all of the words, 68% and 67% correct, respectively. Umbel et al. also analyzed appoximately 15% of the sample for priming effects on cognates, as both the TVIP–H and PPVT–R were administered in the same sitting. They hypothesized that if cognates did influence the children’s test per-formance, then more cognate doublets would exist; however, they found that cognates did not determine an “automatic doublet” (p. 1,016). Thus, Umbel et al. suggested that there was no preliminary evidence for cognate facilitation, and findings suggested that children performed similarly on cognate and noncog-nate vocabulary test items. Umbel and Oller (1994) found similar results with first, third, and sixth grad-ers, where they reported that no significant differ-ences existed in their performance on cognates (60% correct) compared to their performance on all words (65% correct). Although early studies of receptive vocabulary did not find evidence of a cognate effect with young children (Umbel et al., 1992; Umbel & Oller, 1994), initial research suggested that cognates influenced older bilingual children’s vocabulary (Nagy et al., 1993).

Researcher-Designed Investigations With Older ChildrenMost standardized vocabulary tests are not designed to examine cognates specifically; thus, many researchers have developed tasks to investigate phonological simi-larity. In a two-part study, Kohnert et al. (2004) were the first to develop a cognate scale to rate the phono-logical similarity between English and Spanish words, as children may rely more on the perceived phonologi-cal attributes of a word rather than on the orthographic properties. The Crosslinguistic Overlap Scale for


Phonology (COSP; Kohnert et al., 2004; see publica-tion for scale) quantifies cognate status as the amount of phonological overlap between a word in Spanish and its English translation equivalent. The COSP assigns values of 0 to 10 points based on four word characteristics: initial sound, number of syllables, percentage of overlapping consonants, and percentage of overlapping vowels. Higher scores indicate more phonological overlap, where cognates have values of six points or higher.

As a secondary study to the COSP scale develop-ment, Kohnert et al. (2004) developed a picture–word recognition task in Spanish for 8- to 13-year-old monolingual English-speaking children. The English-speaking children were shown two pictures (i.e., one Spanish–English phonologically similar word and one foil), simultaneously heard a Spanish word, and had to decide which picture matched the spoken Span-ish word by pressing a left or right button. Even monolingual English speakers without any exposure to Spanish could identify words with higher Span-ish–English overlap (e.g., heard “elefante,” saw “elephant” and “hammer”) with higher accuracy and quicker response times, with a medium effect size, as compared to words with lower COSP scores.

Malabonga et al. (2008) developed the Cognate Awareness Test (CAT) and administered it in English to Spanish–English bilingual students in fourth and fifth grade. The results indicated that fourth-grade students were aware of cognates, and a 1-year follow-up showed higher performance on cognates than noncognates. Spanish picture vocabulary was related to cognate performance, and English picture vocabulary was strongly related to noncognate per-formance. During test design, the researchers recom-mended using word frequencies specific to children (e.g., Zeno, Ivens, Millard, & Duvvuri, 1995) and quantifying the amount of overlap between words (e.g., Kohnert et al., 2004).

Schelletter (2002) designed a timed picture- naming task to assess the effects of cognates based on “form similarity” (i.e., translation equivalents sharing 50% of similar phonemes) with two groups of 8- to 9-year-olds who were either German lan-guage dominant (i.e., less fluent in English) or Ger-man–English bilinguals. The bilingual children named pictures in both languages (English and German) on cognate and noncognate items. Although both groups had only eight participants, results yielded a small yet significant cognate effect. Schelletter found that children in both groups named cognates quicker than noncognates in both languages. A cognate effect ex-isted regardless of whether the children were less flu-ent in their second language or had a mixed language exposure across their first and second languages.

Picture-naming accuracy performance may vary based on age. Kohnert, Bates, and Hernandez (1999) investigated picture naming with Spanish–English bilinguals ranging from 5 to 20 years of age in a cross-sectional design; response accuracy and speed (i.e., reaction times) showed a developmental lan-guage shift, with maturation from Spanish dominance at younger ages to a balanced cognitive processing in middle childhood, to English dominance in ado-lescence and young adulthood. Kohnert et al. admin-istered the picture-naming task in different blocks: Spanish only, English only, and mixed languages (i.e., cued to respond in a specific language). Despite using early-acquired vocabulary stimuli, the children in the 5- to 7-year age group had decreased accu-racy rates, as evidenced by the percentage of pictures they named correctly (Spanish blocked: 60%, English blocked 38%, Spanish mixed 46%, English mixed 30%). Although this study was foundational for examining young bilinguals’ picture-naming abilities, the potential influence of cognates was not reported.

Recent Cognate Investigations With Bilingual ChildrenMore than 15 years after the initial studies on testing cognate status using receptive vocabulary tests (e.g., Umbel et al., 1992), Pérez et al. (2010) compared cognates and noncognates on a standardized receptive vocabulary subtest (Test of Language Development—Primary: Third Edition; TOLD–P:3; Newcomer & Hammill, 1997) with kindergarten and first-grade bi-lingual students. Pérez et al. administered all 30 test items (18 cognates, 12 noncognates) and tested past the typical ceiling so as to get more opportunities to compare cognates to noncognates. Target words were administered in order of lesser to greater difficulty level and reported word frequency to describe word characteristics. Although previous receptive studies (e.g., Pearson et al., 1993) did not find a cognate difference, Pérez et al.’s results demonstrated higher performance accuracy on cognates over noncognates, with a 1% to 8% cognate advantage, depending on grade and language exposure. The advantage equated to a modest but significant finding of 1.08 to 3.08 words higher on cognates than noncognates. The chil-dren with high Spanish language exposure had higher accuracy on cognates than on noncognates.

With older children of a greater age range, Kel-ley and Kohnert (2012) more recently examined 8- to 13-year-olds’ cognate advantage through standardized administration of a receptive English vocabulary test, the Peabody Picture Vocabulary Test—Third Edition, Form A (PPVT–IIIA; Dunn & Dunn, 1997), and the English Expressive One-Word Picture Vocabulary Test


Third Edition (EOWPVT–3; Brownell, 2000). Span-ish translations of these English tests were made by a trained bilingual speaker in order to identify the degree of phonological overlap (i.e., COSP), but only English items were administered, not Spanish. Due to the increasing difficulty level of vocabulary items on both standardized tests, items were categorized across three difficulty levels as well. Kelley and Koh-nert found evidence of an overall cognate advantage. The children’s scores on cognates were significantly higher than their scores on noncognates for words with comparable difficulty level; however, not all of the children demonstrated a cognate advantage. An analysis of individual student performance on the PPVT–IIIA showed that age predicted 26% of the variance for the cognate advantage, where older children significantly outperformed younger children. On the EOWPVT–3, a nonverbal intelligence score approached significance (p = 0.56), which would account for 9% of the variance. Although conclusive results indicated evidence of cognate facilitation, the use of standardized vocabulary tests limits the ability to match cognates and noncognates on word-level factors such as length and frequency. Kelley and Kohnert also pointed out that the original purpose of the standardized testing tools was to evaluate vocabu-lary, not cognate performance.

Based on this review of the literature, recent studies’ results suggest that vocabulary words show preliminary evidence of a cognate advantage for bilingual children. Investigators have examined cog-nate status using standardized receptive vocabulary tests (Pearson et al., 1993; Pérez et al., 2010; Um-bel et al., 1992), standardized receptive and expres-sive vocabulary tests with older children (Kelley & Kohnert, 2012), and researcher-developed tasks to specifically examine cognates with older children (e.g., Kohnert et al., 2004; Malabonga et al., 2008; Schelletter, 2002). Expressive picture-naming tasks have been conducted with younger bilingual children, but cognate status has not been investigated (Kan & Kohnert, 2005). To date, no one has examined the expressive picture naming of cognates with younger bilingual children.

It was unknown how young children would perform on an expressive picture-naming task that was designed specifically to test cognates. We hypothesized that because bilingual children have doublets in their vocabulary (Pearson et al., 1995), it is plausible that children with more doublets could have higher picture-naming accuracy overall. Perhaps more doublets (i.e., translation equivalents) would be present in word pairs that were cognates due to the overlapping phonology between English and Spanish. If a cognate advantage was found, it was unknown

whether the cognate advantage would be present in only the second language (Malabonga et al., 2008), or, given the young age and early-acquired vocabu-lary, whether a cognate advantage would be present in both languages (Schelletter, 2002), evident of the interacting bilingual linguistic representation. Addi-tionally, if a cognate effect was evident, child factors may relate to cognate status, such as age and/or lan-guage exposure, as seen in receptive tasks (Kelley & Kohnert, 2012; Pérez et al., 2010), or nonverbal intel-ligence as significance was approached with older children (Kelley & Kohnert, 2012), or possibly even English phonological skills due to the phonological overlap of cognates.

The specific aim of this investigation was to ex-amine young ELL children’s performance on expres-sive picture-naming tasks comparing phonologically similar English–Spanish cognates to phonologically dissimilar English–Spanish noncognates. We asked the following research questions:

• Is there a difference in ELL children’s picture naming on cognates versus noncognates?

• Is there an association between cognate perfor-mance and child variables (e.g., age, nonverbal IQ, phonological awareness subtest scores)?

MEthod

Participants. Thirty-one ELL children (15 girls, 16 boys) with a mean age of 68.9 months (SD = 8.8) were included in the final sample. Most parents completed a questionnaire on the child’s language (e.g., age of exposure to each language), develop-ment, and home experiences (90% returned), with follow-up phone inquiries for missing questionnaires. Of completed questionnaires, the majority of the ELL children’s parents reported that they were born in Mexico (85%), and most of the children were born in the United States (93%). All of the children had significant exposure to English and Spanish either through English instruction at school with English and/or Spanish spoken at home or English instruction at school with only Spanish spoken at home. Most of the ELL children’s parents reported Spanish as the home language (78%); however, the ELL children’s parents reported that 42% of the children spoke only Spanish at home, and 58% of the children spoke both English and Spanish at home.

Language exposure and dominance is dynamic, not static, which presents bilingual children as a more heterogeneous population (Kan & Kohnert, 2005). For the current study, we defined exposure as language input and/or output per parent report. For


example, children who were exposed to English after 36 months and whose parents spoke only Spanish in the home would be considered to have more Spanish exposure. In cases where a child had at least one par-ent who spoke both English and Spanish in the home, and parents reported child output in both languages, then the child would have a more balanced exposure to both languages.

Preschool attendance was 61%. All of the chil-dren received free or reduced lunch rates. For most of the participants’ parents, education levels ranged from 4th to 12th grade. All of the children passed a hearing screening if the parents had reported hearing concerns. This study was approved by Florida State University’s Institutional Review Board.

Table 1 provides the ELL children’s scores from the measures that were administered. The Peabody Picture Vocabulary Test—Fourth Edition (PPVT–4; Dunn & Dunn, 2007) assesses children’s receptive English vocabulary, requiring the child to point to a picture from a field of four pictures, with high alpha reliability coefficients on internal consistency levels: .95–.96. The TVIP–H, which is the Spanish adapted version of the PPVT–4, was used to exam-ine the children’s Spanish receptive vocabulary. The TVIP–H has normative data on Spanish monolingual children, and its alpha reliability coefficients are .91–.94. The Primary Test of Nonverbal Intelligence (PTONI; Ehrler & McGhee, 2008) examines young children’s reasoning abilities, with instructions pro-vided in Spanish; alpha reliability coefficients were .90–.92. The Phonological Awareness subtest from the Test of Preschool Early Literacy (TOPEL; Lonigan, Wagner, Torgesen, & Rashotte, 2002) assesses young children’s early literacy skills across 27 items. The reliability coefficients for internal consistency are

.87 and for test–retest reliability are .83. Normative data are only available for children ages 3 to 5 years; however, raw scores were examined. It is notable that standardized monolingual tests may not be appropri-ate measures of bilinguals’ conceptual vocabulary abilities (Kester & Peña, 2002), which may lead to lower than average scores (Umbel et al., 1992); yet, these assessments are used as indices of relative standing for bilingual children as compared to their monolingual peers.

Oversampling occurred; children were excluded if (a) they demonstrated limited English or Span-ish exposure (i.e., unable to complete criterion for picture-naming practice items), (b) parents reported a third language in the home, or (c) there were con-cerns about language difficulties. To screen for pos-sible language impairment, the parent questionnaire included questions concerning the child’s speech and language skills as parent report provides contribut-ing information about the identification of language impairment (Restrepo, 1998). Two included children presented with one late-developing speech-sound error (i.e., interdental /s/); however, the error was not a phonological disorder, and performance on a vocabulary test and nonverbal intelligence test were average or above average in their native language. None of the included children was receiving speech, language, or hearing services. The ELL children had English-only speaking teachers who provided English-only instruction; however, to facilitate comprehensible input, the classrooms had Spanish–English bilingual assistants who provided basic directives and instruc-tion in Spanish.

Cognate–Noncognate TasksWe developed picture-naming tasks to examine the children’s production of English–Spanish cognates and noncognates. We defined cognates as phonologi-cally similar words containing overlapping pho-nemes in English and Spanish that had equivalent meanings (e.g., baby–bebé). Noncognates also had a shared meaning but with limited or no phonologi-cal similarities (e.g., bear–oso). The picture stimuli were dichotomized by COSP scores into cognates (≥6 points) and noncognates (≤5 points), without consideration to orthographic similarity (Kohnert et al., 2004). Second, we obtained words from the MacArthur Communicative Development Invento-ries (CDIs; Fenson et al., 1993) in English and in Spanish (Jackson-Maldonado et al., 2003), which are a noted predictor of picture-naming abilities (Mor-rison, Ellis, & Quinlan, 1992; Windsor & Kohnert, 2004) and assess normal vocabulary acquisition from language delays of young children. Pilot testing was

table 1. Mean scores and standard deviations for the descriptive measures.

English language learners n = 31

M SD

English receptive vocabularya 79.9 14.1Spanish receptive vocabularyb 87.1 17.5Nonverbal intelligencec 97.2 18.7Phonological awareness (raw)d 17.8 6.2

Note. All means and standard deviations were standard scores ex-cept for the Phonological Awareness subtests were raw scores.aPeabody Picture Vocabulary Test—Fourth Edition; bTest de Vocabulario en Imágenes de Peabody; cPrimary Test of Nonver-bal Intelligence; dPhonological Awareness subtest of the Test of Preschool Early Literacy.

Descriptive measure


completed with adults to assess feasibility, refine the word list, and test the equipment.

Stimuli list. Twenty-four images (12 cognates, 12 noncognates) of concrete nouns were selected from a standardized picture database of black line drawings on white backgrounds (Abbate & La Chappelle, 1984a, 1984b; Snodgrass & Vanderwart, 1980). An early childhood teacher, a certified speech-language patholo-gist, and one doctoral candidate reviewed an original list of 18 items to identify 12 cognates that were ap-propriate for young children. Cognates were matched pair-by-pair to noncognates first on word frequency in English (Zeno et al., 1995) and then on number of phonemes (length; Snodgrass & Yuditsky, 1996), for a total of 12 pairs (sample in the Appendix).

Procedure. Each child was tested individually in a quiet room. The child sat at a comfortable distance from the computer screen. An initial practice required the child to become familiar with quickly naming four pictures (e.g., perro–dog, barco–boat, pie–foot, casa–house) on the computer, with examiner feed-back for correct answers (e.g., head nodding, “that’s right”) and incorrect answers (e.g., “do you know the word in [specified language]?”) after the child named each picture. Then, the child had to name eight pic-tures without any feedback, except the initial instruc-tions (e.g., “Now, name these pictures as best as you can”); practice criterion was to name six out of eight correct. If criterion was not met, the training was re-administered for up to three opportunities. Practice items were not included as experimental stimuli in the picture-naming task.

Next, the ELL children completed two blocks of picture naming in English (EPT = English pic-ture-naming task) and in Spanish (SPT = Spanish picture-naming task), in a counterbalanced order. The language of the practice items matched the language of the picture-naming block. Each picture-naming block consisted of 24 trials in randomized order on a Macintosh laptop computer with SuperLab 4 software (Cedrus Corporation, 2011). On each trial, a bell ring was presented to alert the child to a simultaneously presented picture for the child to name; the child had a 5500-ms display, or until a time response was voice-activated. Within each randomized list, no more than three words from the same semantic category appeared in a row. Because matching stimuli on word frequency and length can be challenging, the same 24 items were used in both blocks of picture naming.

An age- and gender-matched group of 31 English monolingual children completed the picture-naming task in English twice in order to address concerns for a potential priming or repetition effect. The monolin-gual children demonstrated 91.5% reliability between the two administrations, indicating approximately

22 agreements out of a possible 24 total items. A paired-samples t test found no significant differences on the overall accuracy, t = .278, p > .05, suggesting minimal priming or repetition effect for the English monolingual speakers. All of the picture-naming tasks were videotaped. The computer software did not judge naming accuracy. Thus, each picture-naming task was scored by the first author using the video recording.

Scoring and Data AnalysesMean accuracy data for the picture-naming tasks was collected for each child. Picture-naming response was scored using a scoring protocol based on pic-ture-naming practices (Cycowicz, Friedman, Roth-stein, & Snodgrass, 1997; Snodgrass & Vanderwart, 1980). Correct responses were accurate naming of the pictures or alternate acceptable naming responses (Cycowicz et al., 1997). Acceptable responses in-cluded changed morphemes (e.g., named grapes as grape) and expanded label (e.g., named plato as plato de comida). The three primary naming errors were (a) incorrect or unacceptable label, (b) response in the opposite language, or (c) no response or “I don’t know”–“no sé.” Additional errors were lack of specificity (e.g., named penguin as bird) and seman-tic errors (e.g., named spoon as toothbrush). Cognate and noncognate pairs were counted. Analyses were conducted using IBM SPSS V.22 statistical software. Effect sizes were calculated using Cohen’s d (1988). Effect sizes were considered small if d is .02 or .03, medium if d is approximately .5, and large if d is greater than .8.

Reliability. A trained research assistant rescored 20% of all of the picture-naming tasks, with an inter-rater scoring reliability of 98%. To evaluate reliability of the picture-naming task scores, 20% of the partici-pants were re-administered the picture-naming tasks, with reliability within acceptable limits (English = 83.2%; Spanish = 85.8%).

RESuLtS

Picture NamingThe first research question asked whether there was a difference in ELL children’s picture naming on cog-nates versus noncognates. Table 2 presents the ELL children’s picture-naming percentage accuracy and standard deviations. A repeated measures analysis of variance (ANOVA) was conducted with two within-subject variables: language (English, Spanish) and cognate status (cognate, noncognate). Language did not have a main effect, F(1, 30) = 0.917, p = 0.346,


but cognate status had a significant main effect, F(1, 30) = 22.558, p < .001. The language × cognate status interaction effect was not significant, F(1, 30) = .001, p > .05. Main effects for cognate status showed that accuracy on cognates was significantly higher than accuracy on noncognates in EPT, F(1, 30) = 14.923, p < .01, d = .748, and in SPT, F(1, 30) = 20,186, p < .01, d = .850. Thus, ELL children’s performance accuracy was significantly higher on cognates than on noncognates both in Spanish picture naming and in English picture naming. Accuracy for cognates and noncognates respectively was 80.9% and 70.4% on the EPT, and 76.3% and 66.0% on the SPT (represented in Figure 1).

Cognate–noncognate word pair item analysis. In order to investigate whether the target word was named accurately in both languages across translation equivalents (e.g., plate–plato), we also calculated the number of cognate and noncognate word pairs (e.g., Umbel et al., 1992). A total of 12 pair opportunities for cognates and 12 pairs for noncognates was pos-sible. The ELL children named word pairs correctly in both Spanish and English: 56.7% of the time for cognates (211/372 = 56.7; 372 = 12 stimulus transla-tion paired-items × 31 participants), and 44.4% of the time for noncognates (165/372). A paired-samples t test found a significant difference between the number of cognate word pairs and noncognate word pairs, t = 5.019, p < .001. Divided across the 31 participants, English–Spanish cognate pairs averaged 6.8 (out of 12), and noncognate pairs averaged 5.3, which equals approximately 1 to 1.5 cognate pairs more than noncognate pairs.

Correlations The second research question asked whether there was any correlation between the bilingual children’s performance on the picture-naming tasks and the child

variables of age, nonverbal IQ, and phonological awareness subtest scores (see Table 3). A significant positive relationship was found between age and Spanish picture-naming accuracy of noncognates (r = .42, p < .05) and between PTONI standard scores and English noncognate accuracy (r = .41, p < .05). On the English Phonological Awareness subtest, signifi-cant positive correlations existed between the TOPEL raw scores and both the English cognates (r = .47, p < .01) and the English noncognates (r = .54, p < .01).

On the EPT, as expected, the ELL children dem-onstrated high positive correlations on the PPVT–4 English vocabulary test with cognate (r = .62, p < .01) and noncognate (r = .75, p < .001) accuracy. On the SPT, similarly, the ELL children had positive cor-relations on the TVIP–H Spanish vocabulary test with cognate (r = .56, p < .001) and noncognate (r = .63, p < .001) accuracy.

dISCuSSIoN

The purpose of this study was to examine young ELL children’s performance on picture-naming tasks in order to compare their performance on English– Spanish words, either cognates sharing or noncog-nates not sharing phonological overlap. We discuss the higher cognate accuracy finding for ELL children with respect to previous vocabulary studies and theo-retical and clinical considerations, as well as limita-tions and future study. Although a modest advantage, higher cognate accuracy on young bilinguals’ picture naming was a substantial finding that should be criti-cally discussed.

table 2. English language learners’ mean accuracy: Cognate versus noncognate on two picture-naming tasks.

Cognates Noncognates

Task M SD M SD

EPT 80.9% 17.0% 70.4% 25.0% 11.0%*

SPT 76.3% 18.0% 66.0% 20.2% 10.0%*

Note. EPT = English picture-naming task, SPT = Spanish picture-naming task.

*p < .05.

Cognate difference

Figure 1. Mean percent accuracy in English and Span-ish picture naming by cognate status (cognate vs. non-cognate), with significant effects for cognate status (p < .01) but not for language (p > .05).

Figure 1. Mean percent accuracy on English and Spanish picture naming by cognate status

(cognate vs. noncognate) with significant effects for cognate status (p < .01) but not for language

(p > .05).

80.9 76.370.4 66

0

10

20

30

40

50

60

70

80

90

100

English Spanish

Accu

racy

(%)

Cognates

Noncognates


Previous Vocabulary Studies

The present study’s findings contradict those of the original (1990s) studies with young bilingual children of no cognate effect (e.g., Umbel et al., 1992), where cognates and noncognates had similar accuracy per-centages. This discrepancy in results may have been due to differences in the assessment measure mo-dality (i.e., expressive vs. receptive) or the level of word difficulty. The present study used words from a developmental word list for young children (i.e., CDIs), and most of the previous studies administered standardized vocabulary tests that were not designed to measure cognate status but increased word dif-ficulty with successive items. However, the current study’s outcomes are similar to more recent stud-ies that suggest a cognate effect, where Pérez et al. (2010) found that young kindergarten and first-grade bilingual children demonstrated higher accuracy on cognates over noncognates, even when using a stan-dardized receptive vocabulary test (i.e., TOLD–P:3).

The current study’s results are also similar to Schelletter (2002), who found a cognate effect on an expressive picture-naming task with slightly older children (i.e., 8- to 9-year-olds). A few differences ex-ist between our study and Schelletter’s. This investiga-tion expanded Schelletter’s findings by including more participants (n = 31, n = 16) and increasing the age range to include younger bilingual children. Another difference was Schelletter’s definition of cognate’s phonological overlap (i.e., 50% of similar phonemes); more recent studies used a sensitive 10-point scale of Spanish–English overlap, but similar results were still found. Although previous research examined cognates with older and younger children, the present study was the first to investigate cognates in the expressive domain with younger bilingual children.

We also found evidence that a cognate effect ex-isted in both languages (Spanish and English), which corresponded to Schelletter’s (2002) findings with a cognate effect in English and German. Interestingly, few studies have looked for a cognate effect in both languages with bilingual children. The cognate effect may be present in both languages as children transi-tion from language dominance in their first language to language dominance in their second language. It is common for bilingual children in the United States to demonstrate a developmental linguistic shift between language dominance in their first language to their second language from early childhood development to middle childhood up and through adulthood (Koh-nert et al., 1999). A cognate effect in both languages requires further investigation but adds an important construct to theoretical bilingual models when includ-ing young children and the transition of language dominance.

Overall, limited significant associations were determined with nonverbal IQ, age, and vocabulary. Age was correlated with noncognate picture nam-ing in Spanish, but not in English; nonverbal IQ was correlated with English noncognate picture naming and PPVT–4 performance. These findings are dis-similar to Kelley and Kohnert (2012), who examined individual effects of a cognate advantage with older students (ages 8–13), where some older bilingual children demonstrated a cognate advantage, but not all students. Kelley and Kohnert found that age sig-nificantly predicted 26% of the variance on a cognate advantage for an English vocabulary test, and non-verbal IQ approached significance and would have contributed 9% of the variance.

Cognates may have a closer relationship with metacognitive abilities for older students, which may explain age and an approaching significance for

table 3. Pearson product–moment correlations among naming accuracy and descriptive measures.

Variable 1 2 3 4 5 6 7 8 9

1. Cognate accuracy – Spanish ___ .68** .01 –.02 .29 –.11 .56** .05 .222. Noncognate accuracy – Spanish ___ .00 –.05 .42* –.11 .63** .06 .283. Cognate accuracy – English ___ .63** .14 .62** –.08 .26 .47**4. Noncognate accuracy – English ___ .29 .75** –.17 .41* .54**5. Age (in months) ___ .08 .08 .23 .59**6. PPVT–4 ___ –.13 .47** .54**7. TVIP–H ___ .06 .238. PTONI ___ .40*9. TOPEL subtest ___

Note. All means were standard scores except the Phonological Awareness subtest (raw scores) and English and Spanish nam-ing accuracy (% correct).

*p < .05, **p < .01.


nonverbal IQ in Kelley and Kohnert’s (2012) study. For younger children, it is possible that other indi-vidual factors influence cognate status. Correlational analyses showed that ELL children’s English vocabu-lary had a significant positive association to picture naming of both cognates and noncognates, which may have been due to the less difficult word selection. Pérez et al. (2010) found that kindergarten and first-grade children’s exposure to Spanish was associated with their recognition of English cognates of Spanish words.

Additionally, we noted that the overall picture-naming percentages of cognates and noncognates combined were higher on English naming (80.9%) than on Spanish naming (76.3%), even though the TVIP–H standard scores were overall higher than the PPVT–4 standard scores. This observation could be due to the high frequency and early acquisition of the selected picture-naming items and may have impacted correlations not found in previous studies. Because prior investigations primarily used standardized vo-cabulary tests, a higher range of item difficulty may also have altered associations.

Theoretical and Clinical ConsiderationsTheoretical model. The children’s picture-naming results in our study demonstrated both a higher ac-curacy on cognates over noncognates as well as a higher percentage of accurate cognate word pairs (i.e., doublet) where a child names the word correctly in both languages (e.g., plate–plato). These findings are congruent with the word production model (Gar-rett, 1975, 1976) that is used with bilingual adults, suggesting a cascaded or simultaneous activation of the phonological characteristics of the word. This representational overlap may facilitate acquisition.

Kan and Kohnert (2005) found that an older group of Hmong–English speakers (M = 5;0) had more doublets than a younger group did; however, the cognates were not examined because cognates may be difficult to find between Hmong and English (Vang, 2005). It is possible the similar phonologi-cal overlap facilitates the acquisition of learning a word, which contributes to a bilingual child’s doublet vocabulary. Umbel et al. (1992) only examined ap-proximately 15% of the sample of correct doublets to investigate a cognate effect; it is possible that a complete analysis would lead to alternate results. When a child knows a word in Spanish, perhaps fewer exposures are required to learn a cognate word in English because the phonologically similar word representation is already present. Corresponding with previous research, cognate facilitation exists for bi-linguals on words with higher phonological overlap,

which is suggested to lead to higher accuracy (Koh-nert et al., 2004; Sunderman & Schwartz, 2008). Al-though previous theoretical models were reserved for bilingual adults, the word production model provides initial support that even young bilingual children have cross-linguistic transfer or influence between languages.

Clinical considerations. This study’s findings contribute to clinical implications in speech-language pathology. First, clinicians should note children’s accuracy on cognates. Cognate status may contribute to the difficulty level of items on vocabulary tests, which would relate to basal and ceiling rules. Thus, bilingual assessments could involve administering test items above the recommended ceiling to examine ELL children’s total performance (Pérez et al., 2010). Cognate status should be taken into consideration as well when making baseline probes for treatment. Higher cognate accuracy on baseline probes has been found in a case study with a bilingual poststroke adult (Kohnert, 2004) and a bilingual child with speech-sound disorders (Leacox, 2013).

A second important clinical consideration may be to provide additional time for word retrieval and/or to consider phonemic cueing as a form of dynamic assessment. Although each picture-naming task was blocked to a single language, the primary investigator noted anecdotally that several children demonstrated activation of the opposite language. For example, one child produced the initial onset of a word in one language, stopped, and then produced the word in the other language (e.g., target Spanish word was silla = chair: “ch-silla”). Children also named com-plete words in the opposing target language before realizing their error; it appeared that some children were cognizant of the wrong language choice error and then self-corrected (e.g., “chair…silla”); others were not. These errors exemplify how young chil-dren appear to have both languages activated at least to some degree. These errors are not uncommon in the literature and are evident even with trilingual speakers who demonstrate multilanguage activation (Francis & Gallard, 2005). Bilinguals are more likely to demonstrate a “tip of tongue” experience, where difficulty is noted in verbal retrieval of a word even though the speaker usually can produce the intended word (Harley, 2008).

Limitations and Future DirectionsOne limitation of this study was the sample’s vari-ability. The Spanish–English bilingual population is heterogeneous, with various cultural and linguistic differences and varying levels of exposure to each language (Goldstein, 2000). Many times, researchers


dichotomize bilinguals into subgroups of sequential or simultaneous, additive or subtractive, or high/low language proficiency levels. Pérez et al. (2010) cre-ated three subgroups of “high English exposure,” “balanced exposure,” and “high Spanish exposure.” The present study included both simultaneous learn-ers (e.g., one parent spoke only Spanish, one parent spoke Spanish and English) where English exposure began at relatively the same time as Spanish expo-sure. Some children did not begin speaking English until formal English instruction began at around 3 to 5 years of age. The sample defined “young” children as being 3 to 6 years old, which may have also added to group variability. Second, the present study did not directly address any confounds of low socioeconomic status. The sample was collected in one of the more underresourced counties in northern Florida. Although socioeconomic status is a multi-factored construct, a low socioeconomic status has been associated with differences in vocabulary (Hart & Risley, 1995). The present study’s group of young Spanish-speaking ELL children represented a het-erogeneous sample of bilingual children from lower socioeconomic backgrounds.

Lastly, although cognates demonstrated a posi-tive cognate effect, the findings were modest. When comparing the number of Spanish–English correct cognate pairs to correct noncognate pairs, cognate doublets had approximately 1 to 1.5 translation pairs more than noncognate pairs. Although a seemingly small advantage, findings were similar to Pérez et al. (2010), which showed a 1 to 3.8 word advantage, and Schelletter (2002), who reported only a slight advan-tage for form similarity (i.e., cognates) on bilingual children’s picture naming. It may be that cognates only contribute a small portion to bilingual vocabulary de-velopment during the early years of acquisition. Given the emphasis on closing the gap in vocabulary learn-ing for ELL children (e.g., Carlo et al., 2004), even small contributions to vocabulary and bridges between languages warrant further investigation.

Future studies can further investigate the influ-ence of vocabulary words’ cognate status. For exam-ple, a cognate sensitivity test for young ELL children could potentially be another indicator of language ability. Similar to the cognitive flexibility of code-switching between African American English and Standard American English dialects (Terry, Connor, Thomas-Tate, & Love, 2010), sensitivity to cognate status may be an indicator of a similar cognitive– linguistic flexibility or vocabulary skills. The Cognate Awareness Test (August et al., 2001) was developed to assess older children’s awareness to cognates during reading comprehension. Given that the pres-ent study’s task was relatively brief to administer (5

to 10 min), further revisions to create an informal assessment tool for younger children would be a feasible and interesting option. Additionally, future studies may investigate how to maximize this magni-tude for cognates by using expanded assessment tools (Malabonga et al., 2008) or even providing brief or extended instruction on cognates, as has been done with older children (Carlo et al., 2004; Nagy et al., 1993; Proctor & Mo, 2009).

In conclusion, young ELL children demonstrate awareness and sensitivity to words with higher degrees of phonological overlap rather than lesser de-grees of overlap. Although previous studies have ex-amined vocabulary tests, few are specifically designed to examine cognates in an expressive picture-naming task with young ELL children. The linguistic overlap of cognates necessitates further investigation. From the 1990s research suggesting that young children do not perform differently on cognates, the present study aligns with recent findings that even young children in preschool and early elementary school are sensitive to phonologically similar cognates.

ACkNoWLEdGMENtSWe sincerely appreciate the teachers and staff at Pan-

handle Area Educational Consortium for Migrant Education and at Leon and Gadsden County who facilitated our access for assessing participants. Special thanks to Maria Pouncey, Director of the Panhandle Area Educational Consortium; research assistant Kelly Worthington; and Marcy Vensel. This research was based on a portion of the first author’s dissertation and was partially supported by the U.S. Depart-ment of Education personnel preparation project at Florida State University, H325D070021.

REFERENCESAbbate, M. S., & La Chappelle, N. B. (1984a). Picture,

please! An articulation supplement. Tucson, AZ: Commu-nication Skill Builders.

Abbate, M. S., & La Chappelle, N. B. (1984b). Pictures, please! A language supplement. Tucson, AZ: Communica-tion Skill Builders.

American Speech-Language-Hearing Association. (2010). 2010 Schools survey report: SLP caseload characteris-tics. Retrieved from www.asha.org/uploadedFiles/Schools-10Caseload.pdf

American Speech-Language-Hearing Association. (2014). Demographic profile of ASHA members providing bi-lingual services. Retrieved from www.asha.org/upload-edFiles/Demographic-Profile-Bilingual-Spanish-Service-Members.pdf

August, D., Kenyon, D., Malabonga, V., Louguit, M., Caglarcan, S., & Carlo, M. (2001). Cognate Awareness Test. Washington, DC: Center for Applied Linguistics.


Bedore, L. M., Peña, E. D., García, M., & Cortez, C. (2005). Conceptual versus monolingual scoring: When does it make a difference? Language, Speech, and Hear-ing Services in Schools, 36, 188–200. doi:10.1044/0161-1461(2005/020)

Blumenfeld, H. K., & Marian, V. (2007). Constraints on parallel activation in bilingual spoken language pro-cessing: Examining proficiency and lexical status using eye-tracking. Language and Cognitive Processes, 22(5), 633–660. doi:10.1080/01690960601000746

Brownell, R. (2000). Expressive One-Word Picture Vocabu-lary Test, Third Edition. Novato, CA: Academic Therapy.

Carlo, M. S., August, D., McLaughlin, B., Snow, C. E., Dressler, C., Lippman, D. N.,…White, C. E. (2004). Closing the gap: Addressing the vocabulary needs of English-language learners in bilingual and mainstream classrooms. Reading Research Quarterly, 39(2), 188–215. Retrieved from www.jstor.org/stable/41151671

Cedrus Corporation. (2011). Superlab 4 [Computer soft-ware]. San Pedro, CA: Author.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.) New York, NY: Erlbaum.

Colomé, Á. (2001). Lexical activation in bilinguals’ speech production: Language-specific or language-indepen-dent? Journal of Memory and Language, 45, 721–736. doi:10.1006/jmla.2001.2793

Costa, A., Caramazza, A., & Sebastian-Galles, N. (2000). The cognate facilitation effect: Implications for models of lexical access. Journal of Experimental Psychol-ogy: Learning, Memory, and Cognition, 26(5), 1283–1296. Retrieved from http://dx.doi.org/10.1037/0278-7393.26.5.1283

Costa, A., Santesteban, M., & Caño, A. (2005). On the facilitatory effects of cognate words in bilingual speech production. Brain and Language, 94, 94–103. doi:10.1016/j.bandl.2004.12.002

Cummins, J. (1981). Empirical and theoretical underpin-nings of bilingual education. Toronto, Canada: Ontario Institute for Studies in Education.

Cunningham, T. H., & Graham, C. R. (2000). Increasing native English vocabulary recognition through Span-ish immersion: Cognate transfer from foreign to first language. Journal of Educational Psychology, 92(1), 37–49. Retrieved from http://dx.doi.org/10.1037/0022-0663.92.1.37

Cycowicz, Y., Friedman, D., Rothstein, M., & Snodgrass, J. G. (1997). Picture naming by young children: Norms for name agreement, familiarity, and visual complexity. Journal of Experimental Child Psychology, 65, 171–237. doi:10.1006/jeep.1996.2356

Davis, C. J., & Perea, M. (2005). BuscaPalabras: A pro-gram for deriving orthographic and phonological neigh-borhood statistics and other psycholinguistic indices in Spanish. Behavior Research Methods, 37(4), 665–671. doi:10.3758/BF03192738

Dunn, L. & Dunn, L. (1981). Peabody Picture Vocabulary Test—Revised. Circle Pines, MN: AGS.

Dunn, L. M & Dunn, L. M. (1997). Peabody Picture Vo-cabulary Test, Third Edition. Circle Pines, MN: AGS.

Dunn, L., & Dunn, L. (2007). Peabody Picture Vocabulary Test, Fourth Edition. Circle Pines, MN: AGS.

Dunn, L., Padilla, E., Lugo, D., & Dunn, L. (1986). Test de Vocabulario en Imágenes Peabody Adaptación Hispanoamericana [Peabody Picture Vocabulary Test: Hispanic-American Adaptation]. Circle Pines, MN: AGS.

Ehrler, D. J., & McGhee, R. L. (2008). Primary Test of Nonverbal Intelligence. Austin, TX: Pro-Ed.

Fenson, L., Dale, P. S., Reznick, J. S., Thal, D., Bates, E., Hartung, J.,…Reilly, J. S. (1993). The MacArthur Communicative Development Inventories: User’s guide and technical manual. San Diego, CA: Singular.

Francis, W. S., & Gallard, S. L. K. (2005). Concept mediation in trilingual translation: Evidence from response time and repetition priming patterns. Psycho-nomic Bulletin & Review, 12(6), 1082–1088. doi:10.3758/BF03206447

Garrett, M. F. (1975). The analysis of sentence production. In G. H. Bower (Ed.), The psychology of learning and motivation: Advances in research and theory (vol. 9, pp. 133–177). New York, NY: Academic Press.

Garrett, M. F. (1976). Syntactic processing in sentence production. In E. Walker & R. Wales (Eds.). New ap-proaches to language mechanisms (pp. 231–256). Amster-dam, Netherlands: North-Holland.

Goldstein, B. (2000). Cultural and linguistic diversity resource guide for speech-language pathologists. San Diego, CA: Singular.

Harley, T. (2008). The psychology of language: From data to theory. Sussex, UK: Psychology Press.

Hart, B., & Risley, T. R. (1995). Meaningful differences in the everyday experience of young American children. Baltimore, MD: Brookes.

Hermans, D., Bongaerts, T., de Bot, K., & Schreuder, R. (1998). Producing words in a foreign language: Can speakers prevent interference from their first lan-guage? Bilingualism: Language and Cognition, 1(3), 213–229. Retrieved from http://dx.doi.org/10.1017/S1366728998000364

Hoshino, N., & Kroll, J. F. (2008). Cognate effects in picture naming: Does cross-language activation sur-vive a change of script? Cognition, 106(1), 501–511. doi:10.1016/j.cognition.2007.02.001

Humphreys, G. W., Riddoch, M. J., & Quinlan, P. T. (1988). Cascade processes in picture identification. Cog-nitive Neuropsychology, 5(1), 67–104. doi:10.1080/02643298808252927

Jackson-Maldonado, D., Thal, D., Fenson, L., March-man, V., Newton, T., & Conboy, B. (2003). MacArthur Inventarios del Desarrollo de Habilidades Comunicativas [MacArthur Communicative Developmental Inventories]: User’s guide and technical manual. Baltimore, MD: Brookes.


Kan, P. F., & Kohnert, K. (2005). Preschoolers learning Hmong and English: Lexical–semantic skills in L1 and L2. Journal of Speech, Language, and Hearing Research, 48, 372–383. doi:10.1044/1092-4388(2005/026)

Kelley, A., & Kohnert, K. (2012). Is there a cognate advantage for typically developing Spanish-speaking English-language learners? Language, Speech, and Hear-ing Services in Schools, 43, 191–204. doi:10.1044/0161-1461(2011/10-0022)

Kester, E. S., & Peña, E. D. (2002). Language ability assessment of Spanish–English bilinguals: Future direc-tions. Practical Assessment, Research & Evaluation, 8(4). Retrieved from http://PAREonline.net/getvn.asp?v=8&n=4

Kohnert, K. (2004). Cognitive and cognate-based treat-ments for bilingual aphasia: A case study. Brain and Lan-guage, 91, 294–302. doi:10.1016/j.bandl.2004.04.001

Kohnert, K., Bates, E., & Hernandez, A. E. (1999). Balancing bilinguals: Lexical–semantic production and cognitive processing in children learning Spanish and English. Journal of Speech, Language, and Hearing Re-search, 42, 1400–1413. doi:10.1044/jslhr.4206.1400

Kohnert, K., Windsor, J., & Miller, R. (2004). Crossing borders: Recognition of Spanish words by English-speak-ing children with and without language impairment. Applied Psycholinguistics, 25, 543–564. Retrieved from http://dx.doi.org/10.1017/S0142716404001262

Kroll, J. F., Gerfen, C., & Dussias, P. E. (2008). Labora-tory designs and paradigms: Words, sounds, and sen-tences. In L. Wei & M. G. Moyer (Eds.), The Blackwell guide to research methods in bilingualism and multilin-gualism (pp. 108–131). Malden, MA: Blackwell.

Leacox, L. R. (2013, November). Spanish–English phono-logical overlap: Children’s cognate assessment & facilita-tion in speech therapy. Poster presented at the annual meeting of the American Speech-Language-Hearing As-sociation, Atlanta, GA.

Levelt, W. J. M. (1989). Speaking: From intention to ar-ticulation. Cambridge, MA: MIT Press.

Lonigan, C. J., Wagner, R. K., Torgesen, J. K., & Rashotte, C. A. (2002). Test of Preschool Early Literacy. Austin, TX: Pro-Ed.

Malabonga, V., Kenyon, D. M., Carlo, M., August, D., & Louguit, M. (2008). Development of a cognate aware-ness for Spanish-speaking English language learners. Language Testing, 25(4), 495–519. doi:10.1177/0265532208094274

Marchman, V., & Martínez-Sussman, C. (2002). Con-current validity of caregiver/parent report measures of language for children who are learning both English and Spanish. Journal of Speech, Language, and Hearing Re-search, 45, 983–997. doi:10.1044/1092-4388(2002/080)

Morrison, C. M., Ellis, A. W., & Quinlan, P. T. (1992). Age of acquisition, not word frequency, affect object naming, not object recognition. Memory and Cognition, 20(6), 705–714. doi:10.3758/BF03202720

Nagy, W., García, G., Durgunoğlu, A., & Hancin-Bhatt, B. (1993). Spanish–English bilingual students’ use of

cognates in English reading. Journal of Reading Behav-ior, 25, 241–259. doi:10.1080/10862969009547816

Newcomer, P., & Hammill, D. (1997). Test of Language Development—Primary: Third Edition. Austin, TX: Pro-Ed.

Pearson, B. Z., Fernández, S. C., & Oller, K. D. (1993). Lexical development in bilingual infants and toddlers: Comparison to monolingual norms. Language and Learning, 43(1), 93–120. doi:10.1111/j.1467-1770.1993.tb00174.x

Pearson, B. Z., Fernández, S., & Oller, K. D. (1995). Cross-language synonyms in the lexicons of bilingual in-fants: One language or two? Journal of Child Language, 22(2), 345–368. Retrieved from http://dx.doi.org/10.1017/S030500090000982X

Peña, E. D., Bedore, L. M., & Zlatic-Giunta, R. (2002). Category-generation performance of bilingual children: The influence of condition, category, and language. Journal of Speech, Language, and Hearing Research, 45, 938–947. doi:10.1044/1092-4388(2002/076)

Pérez, A. M., Peña, E. D., & Bedore, L. M. (2010). Cognates facilitate word recognition in young Span-ish–English bilinguals’ test performance. Early Childhood Services, 4, 55–67.

Proctor, C. P., & Mo, E. (2009). The relationship between cognate awareness and English comprehension among Spanish–English bilingual fourth grade students. TESOL Quarterly, 43(1), 126–136. doi:10.1002/j.1545-7249.2009.tb00232.x

Restrepo, M. A. (1998). Identifiers of predominantly Spanish-speaking children with language impairment. Journal of Speech, Language, and Hearing Research, 41, 1398–1411. doi:10.1044/jslhr.4106.1398

Sánchez-Casas, R., & García-Albea, J. E. (2005). The representation of cognate and noncognate words in bilin-gual memory: Can cognate status be characterized as a special kind of morphological relation? In J. Kroll & A. de Groot (Eds.), Handbook of bilingualism: Psycholin-guistic approaches (pp. 226–250). New York, NY: Oxford University Press.

Schelletter, C. (2002). The effect of form similarity on bilingual children’s lexical development. Bilingualism: Language and Cognition, 5(2), 93–107. Retrieved from http://dx.doi.org/10.1017/S1366728902000214

Snodgrass, J. G., & Vanderwart, M. (1980). A standard-ized set of 260 pictures: Norms for name agreement, image agreement, familiarity, and visual complexity. Journal of Experimental Psychology: Human Learning and Memory, 6, 174–215. Retrieved from http://dx.doi.org/10.1037/0278-7393.6.2.174

Snodgrass, J. G., & Yuditsky, T. (1996). Naming times for the Snodgrass and Vanderwart pictures. Behavior Research Methods, Instruments, & Computers, 28(4), 516–536. doi:10.3758/BF03200540

Sunderman, G., & Schwartz, A. I. (2008). Using cog-nates to investigate cross-language competition in second language processing. TESOL Quarterly, 42, 527–536. doi:10.1002/j.1545-7249.2008.tb00145.x


Terry, N. P., Connor, M. C., Thomas-Tate, S., & Love, M. (2010). Examining relationships among dialect varia-tion, literacy skills, and school context in first grade. Journal of Speech, Language, and Hearing Research, 53, 126–145. doi:10.1044/1092-4388(2009/08-0058)

Thomas, S., Nash, R., Thomas, G., & Richmond, D. (2005). The big red book of Spanish vocabulary. New York, NY: McGraw-Hill.

Umbel, V. M., & Oller, D. K. (1994). Developmental changes in receptive vocabulary in Hispanic bilingual school children. Language Learning, 44(2), 221–242. doi:10.1111/j.1467-1770.1994.tb01101.x

Umbel, V. M., Pearson, B. Z., Fernández, M. C., & Oller, D. K. (1992). Measuring bilingual children’s re-ceptive vocabularies. Child Development, 63, 1012–1020. Retrieved from www.jstor.org/stable/1131250

APPENdIx. SAMPLE WoRd StIMuLI FoR PICtuRE NAMING

Cognates Noncognates

Target Age of Target Age of words COSP #Phon Freq acquisition words COSP #Phon Freq acquisition

Eng Span (1–10) Eng Span Eng Span Eng Eng Span (1–10) Eng Span Eng Span Eng

flower flor 7 5 4 40 33.04 1 apple manzana 2 4 7 39 11.07 1plate plato 7 4 5 49 30.89 1 finger dedo 2 5 4 49 50.89 1baby bebé 8 4 4 113 16.96 1 eye ojo 1 1 3 125 71.25 1

Note. The complete picture-naming task stimuli included 24 words (12 cognates, 12 noncognates). COSP = Crosslinguistic Overlap Scale for Phonology (Kohnert et al., 2004). Sources were word frequency (English: Zeno et al., 1995; Spanish: Da-vis & Perea, 2005) and age of acquisition (CDI; Fenson et al., 1993).

Vang, C. T. (2005). Hmong American K–12 students and the academic skills needed for a college education: A re-view of the existing literature and suggestions for future research. Hmong Studies Journal, 5, 1–31.

Windsor, J., & Kohnert, K. (2004). The search for com-mon ground: Part 1. Lexical performance by linguisti-cally diverse learners. Journal of Speech, Language, and Hearing Research, 47, 877–890. doi:10.1044/1092-4388(2004/065)

Zeno, S. M., Ivens, S. H., Millard, R. T., & Duvvuri, R. (1995). The educator’s frequency guide. New York, NY: Touchstone Applied Science Associates.

Contact author: Lindsey Leacox, 230 Communica-tion Arts Center, Cedar Falls, IA 50614. E-mail: [email protected]

Documents

Young Spanish-English Language Learners' Cognate ... · gual Hmong–English preschoolers (ages 3;4–5;2 [years;months]). They administered a 50-item recep-tive picture identification