Bilingual children's language and literacy is stronger in some domains than others. Reanalysis of data from a broad-scale study of monolingual English and bilingual Spanish–English learners in Miami provided a clear demonstration of “profile effects,” where bilingual children perform at varying levels compared to monolinguals across different test types. The profile effects were strong and consistent across conditions of socioeconomic status, language in the home, and school setting (two way or English immersion). The profile effects indicated comparable performance of bilingual and monolingual children in basic reading tasks, but lower vocabulary scores for the bilinguals in both languages. Other test types showed intermediate scores in bilinguals, again with substantial consistency across groups. These profiles are interpreted as primarily due to the “distributed characteristic” of bilingual lexical knowledge, the tendency for bilingual individuals to know some words in one language but not the other and vice versa.

This study examined how three different sources of stimulus variability—overall amplitude, fundamental frequency, and speaking rate—affect second language (L2) vocabulary learning. Native English speakers learned Spanish words in presentation formats with no variability, moderate variability, and high variability. Dependent measures were accuracy and latency of picture-to-Spanish and Spanish-to-English recall. The findings indicated that variability sources that do not affect first language (L1) word identification (amplitude, fundamental frequency) do not affect L2 vocabulary learning. Conversely, variability in speaking rate, which negatively affects L1 word identification, positively affected L2 vocabulary learning. These findings are consistent with an integrated account in which sources of variability that affect phonetically relevant properties of speech impede L1 speech processing but improve L2 vocabulary learning.

This study examined second language (L2) experience effects on children's acquisition of fluency-(speech rate, frequency, and duration of pausing) and prosody-based (stress timing, peak alignment) suprasegmentals. Twenty Korean children (age of arrival in the United States = 7–11 years, length of US residence = 1 vs. 11 years) and 20 age-matched English monolinguals produced six English sentences in a sentence repetition task. Acoustic analyses and listener judgments were used to determine how accurately the suprasegmentals were produced and to what extent they contributed to foreign accent. Results indicated that the children with 11 years of US residence, unlike those with 1 year of US residence, produced all but one (speech rate) suprasegmentals natively. Overall, findings revealed similarities between L2 segmental and suprasegmental learning.

Two experiments tested the activation of gender stereotypes for Italian role nouns (e.g., teacher). The experimental paradigm was modeled on the one proposed by a study by Banaji and Hardin: participants were shown a prime word followed by a target pronoun (he or she) on which they performed a gender decision task. The prime words were Italian role nouns that had an associated female stereotype, an associated male stereotype, or no associated stereotype (control condition). The results of Experiment 1 showed no influence of stereotypes on the gender decision response to personal pronouns. However, when, in Experiment 2, the prime-target interval and the prime presentation times were prolonged, a stereotypical gender priming effect did emerge. A different inhibition pattern was found when the pronoun was preceded by a noun associated with a male stereotype instead of a female one.

The purposes of this study are to test whether the processing of subject relative (SR) clauses is easier than that of object relative (OR) clauses in Turkish and to investigate whether the comprehension of SRs can be better explained by the linear distance hypothesis or structural distance hypothesis (SDH). The question is examined in two groups of second language (L2) learners of different proficiency levels and a few agrammatics expected to show a similar pattern. Each participant is asked to comprehend 15 sentences containing SRs and ORs via a picture selection task. The results indicate that comprehension of SRs is easier than that of ORs for intermediate level L2 learners, whereas there is no significant difference between the types of relative clauses for early learners. Another result is that early learners produce errors similar to those of agrammatics, which are explained through trace deletion and referential strategy. These findings on Turkish provide significant support for the SDH.

Spanish-speaking (SS) children with language impairment (LI) present with deficits in morphology and verb argument structure. These language areas may be useful for clinical identification of affected children. This study aimed to evaluate the discrimination accuracy of spontaneous language measures with SS preschoolers to tease out what combination of grammatical measure(s) were responsible for the LI deficits, and to determine the role of verb argument structure and syntactic complexity in identifying SS children with LI. Two sets of experiments were conducted on the spontaneous language samples of SS preschoolers with and without LI using discriminant function analyses. The study found that (a) mean length of utterance in words (MLUW) and ungrammaticality index in combination are fair to good discriminators of preschoolers with LI; (b) a morphology model combining correct use of articles, verbs, and clitics fairly discriminates LI children but may miss children whose language has limited syntactic complexity; and (c) semantic–syntactic complexity measures, such as MLUW, theme argument omissions, and ditransitive verb use, should be considered in the assessment of Spanish LI. The children who were bilingual and Spanish dominant in the study were classified as accurately as the Spanish-only children.

The influence of development and literacy upon syllabification in French was evaluated by comparing the segmental behavior of 4- to 5-year-old preliterate children and adults using a pause insertion task. Participants were required to repeat bisyllabic words such as “fourmi” (ant) by inserting a pause between its two syllabic components (/fur/-/mi/). In the first experiment we tested segmentation over a range of 49 double intervocalic consonant clusters. A similar general segmentation behavior was observed in both age groups, with a pattern that fit the predictions from a legality principle-based model of syllabification. Experiment 2 revealed that opacity between phonological and orthographic representations lead to increased ambisyllabic responses and a reduction in segmentation consistency in adults. In total, these findings indicate that syllabic forms are consistently represented from an early age, but that segmentation in metalinguistic tasks is susceptible to contamination from spelling and etymological knowledge.

Cross-linguistic borrowing (overt use of words from the other language) and transfer (use of semantic or syntactic structures from the other language without active switching to that language) were examined during language production in Russian–English bilinguals. Grammatical category (noun/verb) and level of concreteness were found to influence language interaction. More cross-linguistic borrowings were found for nouns than for verbs and more cross-linguistic transfers were found for verbs than for nouns, suggesting that grammatical categories are differentially vulnerable to covert and overt language interaction. Moreover, concrete nouns and verbs were transferred more than abstract nouns and verbs, suggesting that level of concreteness influences lexical access in bilinguals. Overall, bilinguals transferred more when speaking their second and less proficient language and borrowed more when speaking their first and less recent language (especially if the described event took place in the other language). We suggest that language architecture (e.g., semantic representation, lexical access) and language environment influence the nature of cross-linguistic interaction.