This paper explores whether the poor mastery of morphosyntax exhibited by second language (L2) learners can be tied to difficulties with non-syntactic processing. Specifically, we examine whether problems with English regular and irregular past tense are related to poor L2 phonological ability and lexical access, respectively. In Experiment 1, L2 learners showed poorer past tense mastery than native English speakers in grammaticality judgment and production tasks. L2 phonological ability was positively correlated with correct performance on regular verbs and negatively with unmarked production. L2 lexical access was positively correlated with correct performance on irregular verbs, and negatively with overregularization production. Experiment 2 simulated these difficulties in native English speakers by placing them under phonological processing (noise) or lexical access (deadline) stress. Noise selectively impacted regular verbs in grammaticality judgment but impacted all verb types in production. Deadline pressure impacted irregular verbs while sparing regular verbs across both tasks. Thus, non-syntactic processing difficulties can have specific impacts on morphosyntactic performance in both non-native and native English speakers.

This self-paced reading study first tested the prediction that the garden path effect previously observed during the processing of subject–object ambiguities in native English would not obtain in a null subject language like Spanish. The investigation then further explored whether the effect would be evident among near-native readers of Spanish whose native language was a non-null subject language like English. Twenty-three near-native and 33 native readers of Mexican Spanish read sentences like Cuando el escultor acabó/volvió la obra tenía tres metros de altura “When the sculptor finished/came back the piece was three meters in height”. The results suggest that (i) Spanish differs from English for this type of processing and (ii) native and near-native processing can be guided by largely similar principles, at least where lexical information like verb transitivity is concerned.

We investigate here the contribution of code-switching and structural priming to variable expression of the Spanish first person singular subject pronoun in the New Mexican bilingual community. Comparisons with both Spanish and English benchmarks indicate no convergence of Spanish toward English grammar, including in the presence of code-switching, where the linguistic conditioning of variant selection remains unaltered. We find a language-internal and cross-language priming effect, albeit of differing strength, such that speakers’ preceding coreferential (Spanish and English) subject pronouns favor subsequent pronouns, whereas unexpressed subjects tend to be followed by unexpressed subjects. Given the rarity of unexpressed subjects in English, in the presence of code-switching fewer tokens occur with unexpressed primes. Thus, code-switching has no intrinsic effect. Instead, it results in associated shifts in the distribution of contextual features relevant to priming, contrary to the convergence-via-code-switching hypothesis and in accordance with the contextual distribution-via-code-switching hypothesis, which we put forward here.

Spoken language (unimodal) interpreters often prefer to interpret from their non-dominant language (L2) into their native language (L1). Anecdotally, signed language (bimodal) interpreters express the opposite bias, preferring to interpret from L1 (spoken language) into L2 (signed language). We conducted a large survey study (N = 1,359) of both unimodal and bimodal interpreters that confirmed these preferences. The L1 to L2 direction preference was stronger for novice than expert bimodal interpreters, while novice and expert unimodal interpreters did not differ from each other. The results indicated that the different direction preferences for bimodal and unimodal interpreters cannot be explained by language production–comprehension asymmetries or by work or training experiences. We suggest that modality and language-specific features of signed languages drive the directionality preferences of bimodal interpreters. Specifically, we propose that fingerspelling, transcoding (literal word-for-word translation), self-monitoring, and consumers’ linguistic variation influence the preference of bimodal interpreters for working into their L2.

This article reports the results of an experiment on production of his/her in English as a second language (L2) by proficient native speakers of Italian, Spanish, and Dutch. In Dutch and English, 3rd person singular possessive pronouns agree in gender with their antecedents, in Italian and Spanish possessives in general agree with the noun they accompany (possessum). However, while in Italian the 3rd person singular possessives overtly agree in gender with the possessums, in Spanish they lack overt morphological gender marking. Dutch speakers were found to make very few possessive gender errors in any condition, Spanish and Italian speakers, on the other hand, behaved like Dutch speakers when the possessum was inanimate, but made more errors when it was animate (e.g., his mother). Thus, even proficient L2 speakers are susceptible to the influence of automatic processes that should apply in their first language alone. The pattern of results has implications for pronoun production and models of bilingual language production.

In two cross-linguistic priming experiments with native German speakers of L2 English, we investigated the role of constituent order and level of embedding in cross-linguistic structural priming. In both experiments, significant priming effects emerged only if prime and target were similar with regard to constituent order and also situated on the same level of embedding. We discuss our results on the basis of two current theoretical accounts of cross-linguistic priming, and conclude that neither an account based on combinatorial nodes nor an account assuming that constituent order is directly responsible for the priming effect can fully explain our data pattern. We suggest an account that explains cross-linguistic priming through a hierarchical tree representation. This representation is computed during processing of the prime, and can influence the formulation of a target sentence only when the structural features specified in it are grammatically correct in the target sentence.

Heritage speakers who re-learn their childhood language in adulthood are an important group for the study of L3 acquisition. Such re-learners have selective advantages over other L2/L3 learners in phonetics/phonology, but lack a global advantage at re-learning the prestige variety of their L1. These learners show asymmetrical transfer effects in morphosyntax: transfer occurs only from the dominant language. Two tentative explanations for this asymmetry are suggested. First, re-learners may deploy the skills acquired in a classroom setting, where they have used only their dominant language. Second, re-learners may implicitly strive to increase the typological distance between their childhood language and the language of classroom instruction. These findings have implications for models of L3/Ln learning: the Cumulative Enhancement Model, the Typological Proximity Model, and the L2 Status Factor Model. The data discussed in this paper are most consistent with the latter model, but they also highlight the significance of the typological distance between languages under acquisition.

The present study follows the role of the first (L1, English) and second (L2, Japanese or Spanish) languages in ab initio third language (L3, Latin) acquisition. Participants (N = 25) were L2 classroom learners without immersion experience. In order to complement previous generativist studies and to offer a fuller developmental account of how transfer operates at the morphosyntactic level, the Competition Model (CM) was adopted as theoretical framework. Positive changes in overall accuracy and sentence processing patterns in role assignment in L3 Latin show L3 development as largely modulated by the L1, suggesting that higher levels of L2 resonance are necessary for integrated patterns of L1 and L2 cues to emerge.

Translations often do not align directly across languages, and indirect mappings reduce the accuracy of language learning. To facilitate examination of this issue, we developed a new continuous measure for quantifying the semantic relatedness of words with more than one translation (hereafter translation-ambiguous words). Participants rated the similarity of each translation to every other translation, yielding a Translation Semantic Variability (TSV) score, ranging from 1.00 (unrelated) to 7.00 (related). Then, we determined how relatedness between translations affects translation-ambiguous word learning from German to English. German words with low TSV scores were recognized as translations more slowly and less accurately than German words with high TSV scores. TSV explains unique variance beyond the previously-used dichotomous classification of words as form vs. meaning ambiguous. We propose that the relatedness of the translation alternatives influences learning because it affects the ease with which a one-to-one mapping can be established between form and meaning.

Carroll (Carroll) takes issue with the use of parent report to obtain quantity of language exposure measures in research on bilingual development. When discussing parent questionnaires, Carroll writes “Temporal units are crude measures of exposure and they tell us nothing about input”. While I agree that temporal units do not tell us much about the fine-grained details of the input within the temporal units, importantly, parent-report-based measures of input quantity have predicted variation in bilingual development of phonology, vocabulary and morphosyntax. These are robust and reliable findings across numerous studies, and yet, Carroll skates over them as if they did not matter, or dismisses them as trivial. Furthermore, Carroll seems to lead readers to believe that only coarse-grained, language-use temporal units have been obtained through this method; on the contrary, researchers have obtained fine-grained input quality details via parent report that also predict bilingual children's development. Finally, in some circumstances, parent report data is the only feasible method for obtaining language exposure information.

Our keynote, “Understanding heritage languages” (Polinsky & Scontras, 2019), and the commentaries to it make the following main points.

This study compared working memory ability in multilingual young adults and their monolingual peers on four components of working memory (verbal and visuospatial storage, verbal and visuospatial processing). The sample comprised 39 monolingual English speakers, and 39 multilinguals, who spoke an African language as their first and third languages, and English as their second language, all with high levels of proficiency. The multilingual young adults came from lower socioeconomic status (SES) backgrounds and possessed smaller English vocabularies than the monolinguals, features which make this group an under-researched population. Both when SES and verbal ability were and were not statistically controlled, there was evidence of a multilingual advantage in all of the working memory components, which was most pronounced in visuospatial processing. These findings support evidence from bilinguals showing cognitive advantages beyond inhibitory control, and suggest that multilingualism may influence the executive control system generally.

It has been suggested that bilinguals learn additional languages ‘better’ than monolinguals. However, evidence is sparse, particularly for grammar. We examined behavioral and neural correlates of learning an additional (artificial) language in early Mandarin–English bilinguals, compared to English monolinguals. Following grammar instruction, participants practiced comprehension and production, and judged grammaticality at low and high proficiency while event-related potentials (ERPs) were acquired. Bilinguals and monolinguals did not differ on behavioral measures, but showed distinct ERP patterns. At low proficiency only bilinguals showed a P600, a common ERP correlate of syntactic processing in native speakers of languages. At high proficiency both groups showed P600s, though the monolinguals also evidenced an anterior positivity not typically found in native speakers of languages during syntactic processing. These findings suggest that, even without bilingual/monolingual behavioral differences, bilinguals show ERP patterns for an additional language that are more similar to those of native speakers of languages.

Previous studies have demonstrated that language switching in bilinguals can be affected by a number of variables, including the processing context. Here, we used a modified language-switching task combined with a Stroop paradigm, which manipulated the context of the task, to examine the impact of processing context on switch costs. The results of both Experiment 1 and Experiment 2 showed that the switch costs and the level of asymmetry in the switch costs are larger in the conflicting context than in the non-conflicting context, suggesting that the processing context affects the switch costs. In addition, the results of Experiment 2 revealed that individual variances in cognitive control capacity also play a role in the overall magnitude of the switch costs. Critically, processing context effects can be modulated by individual variance in cognitive control capacities. The results of this study are discussed within the framework of classic models of bilingual language control (e.g., the inhibitory control model).

In the current study, we investigated bilingual error detection by measuring the repair rate of language intrusions (i.e., involuntary production of nontarget language words) that arose while bilinguals produced sentences in a language switching context. This allowed us to compare two prominent accounts of error detection in a bilingual setting. According to the conflict monitoring account, error detection is initiated by interference. Since language switching increases bilingual language interference, error detection should be better in switch relative to repetition trials. According to the perceptual loop theory, error detection is based on language comprehension. Since language switching is known to impair language comprehension, it follows that error detection should be worse in switch relative to repetition trials. The results showed that the repair rate of language intrusions was higher in switch than repetition trials, thus providing evidence that bilingual language interference instigates error detection, in line with the conflict monitoring account.

Our study investigated how attention paid to a target syntactic construction causative had is related to the storage capacity and attention regulation function of working memory (WM) and how these WM abilities moderate the change of knowledge of the target construction in different input conditions. 80 Sri Lankan learners of English were exposed to examples of the target construction in explicit and implicit learning conditions and their eye movements were tracked as they read the input. Correlational and multiple regression analyses indicated a very strong relationship between WM abilities and gains in the knowledge of the target construction. WM scores were closely associated with gains in receptive knowledge in all input conditions, but they had a weaker link to the improvement of productive knowledge in the implicit learning conditions. The amount of attention paid to input was also strongly related to WM abilities.

The current project examined whether and to what degree continued L2 input, operationalized as length of residence (LOR), and age of acquisition (AOA), defined as the first intensive exposure to the target language, can be predictive of adult Japanese learners’ production of word-initial English /ɹ/. Data were collected from 65 participants, consisting of three groups of Japanese learners of English (n = 13 for Short-, Mid-, and Long-LOR groups, respectively) and two groups of baseline speakers (n = 13 for Japanese- and English-Baseline groups, respectively). Their production of /ɹ/ was elicited via three oral tasks (i.e., word reading, sentence reading, timed picture description). Acoustic analyses were carried out along four dimensions: third formant (F3), second formant (F2), first formant (F1) frequencies, and formant transition duration. The results demonstrated that (a) all learners reached native-like proficiency with respect to the use of existing cues (F2, transition duration) within approximately one year of LOR, (b) their performance was negatively related to AOA to some degree, and (c) longer LOR was predictive of the development of the new cue (F3). These results suggest that late L2 speech sound acquisition and proficiency may be characterized by different levels of phonetic processing.

In a multiple-session training study, native English speakers learned foreign Dutch vocabulary items that mapped to English either in a one-to-one way (translation-unambiguous) or in a one-to-many way (translation-ambiguous), such that two Dutch words corresponded to a single English translation. Critically, these two translation-ambiguous Dutch words were taught on consecutive trials in the same session, or were presented separately, such that each word was taught in a separate session. Translation-ambiguous words were produced and recognized substantially less accurately than translation-unambiguous words on tests administered one and three weeks after training. An ambiguity advantage emerged, however, in a free-recall test. Interestingly, teaching both translations together led to superior performance over teaching them in separate sessions, in which case the translation learned first enjoyed a considerable advantage over that learned second. These findings underscore the importance of order of acquisition in second-language vocabulary learning, and have practical implications for language instruction.

In an attempt to explain first language attrition in emigrant populations, this paper investigates the explanatory power of a framework that has – until now – received little attention: the regression hypothesis (Jakobson, 1941). This hypothesis predicts that the order of attrition is the reverse of the order of acquisition. The regression hypothesis was tested in relation to the loss of morphology and syntax in Dutch immigrants in Anglophone Canada. Evidence in favor of regression was found, but mainly in the morphological domain. Syntax, on the other hand, was mostly characterized by L2 influences from English. As it is problematic to treat regression as a theory in its own right, these findings are then explained in the light of both generative and usage-based approaches, as well as the more recent Dynamic Systems Theory.

Bilingual children have long been held to have ‘separate linguistic systems’ from the start (e.g., Meisel, 2001). This paper challenges that assumption with data from five bilingual children's first 100 words. Whereas the prosodic structures represented by a child's words may or may not be differentiated by language, emergent phonological templates are not, the same patterns being deployed as more complex adult word forms are targeted in each language. Reliance on common (idiosyncratic) phonological templates for the two languages is ascribed to children's experience with their own voice (in production) as well as with others’ speech. Both experimental studies and spontaneous cross-linguistic speech errors in adults and older children are cited to support the view that, for a bilingual, unconscious processing draws on both languages throughout the lifespan, which suggests that the emphasis on ‘separate systems’ (from the start or thereafter) may be misconceived.