This study investigated differences in the English referencing behaviors in the context of oral narrative retell between typically developing first- and second-grade Spanish–English emergent bilingual children in dual language immersion and English-only instructional contexts (N = 105). Children heard and retold Mercer Mayer wordless picture books, and analyses were conducted to examine how they used nominals and pronominals to maintain and switch reference to potential thematic protagonists in the story. Multivariate analysis of variance showed significant grade-level differences in the proportion of pronominals used to switch and maintain reference to BOY/S (boy and dog or frog), as well as to switch to BOY. In contrast, instructional context differences were significant only for reference to the DOG or FROG. The finding that second graders in both dual language immersion and English-only programs continued to demonstrate an overreliance on pronominal forms to switch reference in a second language suggests that differences in literacy and oral language development may extend beyond the grades that we investigated. It is therefore important to continue investigating the referencing behaviors of emergent bilingual children throughout the elementary years of schooling.

We examined how proficiency influences the processing of emotion words in Spanish–English bilinguals (22 balanced and 20 unbalanced). All unbalanced bilinguals were more proficient in English than Spanish. Participants rated the valence of negative, neutral, and positive words in both languages while EEG was being recorded. ERP latencies and amplitudes were analyzed for two components. The language effect was significant on the late positive component (LPC) amplitude, which was larger for emotion than for neutral words for both groups in English. The unbalanced group presented larger LPC amplitudes for positive than for neutral and for neutral than for negative words in Spanish, suggesting emotion processing differences in these participants’ less proficient language. Valence effects were consistent across languages for the balanced group, but not for the unbalanced group, perhaps reflecting differences in reactivity to emotion words in the less proficient language.