This paper sketches a comprehensive framework for modeling and interpreting language contact phenomena, with speakers’ bilingual strategies in specific scenarios of language contact as its point of departure. Bilingual strategies are conditioned by social factors, processing constraints of speakers’ bilingual competence, and perceived language distance. In a number of domains of language contact studies important progress has been made, including Creole studies, code-switching, language development, linguistic borrowing, and areal convergence. Less attention has been paid to the links between these fields, so that results in one domain can be compared with those in another. These links are approached here from the perspective of speaker optimization strategies. Four strategies are proposed: maximize structural coherence of the first language (L1); maximize structural coherence of the second language (L2); match between L1 and L2 patterns where possible; and rely on universal principles of language processing. These strategies can be invoked to explain outcomes of language contact. Different outcomes correspond to different interactions of these strategies in bilingual speakers and their communities.

Heritage language acquisition has been characterized by various asymmetries, including the differential acquisition rates of various linguistic areas and the unbalanced acquisition of different categories within a single area. This paper examines Arabic heritage speakers’ knowledge of subject–verb agreement versus noun–adjective agreement with the aim of contrasting their distributions and exploring areas of resilience and vulnerability within Arabic heritage speech and their theoretical implications. Two oral-production experiments were carried out, one involving two picture-description tasks, and another requiring an elicited narrative. The results of the study show that subject–verb agreement morphology is more maintained than noun–adjective morphology. Moreover, the unmarked singular masculine default is more robust than the other categories in both domains and is often over-generalized to other marked categories. The results thus confirm the existence of these asymmetries. We propose that these asymmetries may not be explained by a single factor, but by a complex set of morphological, syntactic, semantic, and frequency-related factors.

This study investigates L1–L2 convergence among bilinguals at an intermediate (CEFR-B2) level of L2 proficiency, focusing on the clausal packaging of Manner and Path of motion. Previous research has shown cross-linguistic differences between English and Japanese in this domain (Allen et al., 2003; Kita & Özyürek, 2003, though note Brown & Gullberg, 2012). We compared descriptions of motion from monolingual English and Japanese speakers to L1 and L2 descriptions from Japanese users of English as a second (ESL) and foreign (EFL) language. Results showed no significant difference between the monolinguals, who predominately used single-clause constructions packaging Manner and Path. However, bilinguals, both ESL and EFL speakers, used significantly more multi-clause constructions in both their L1 and L2. Following Pavlenko (2011a), findings are interpreted as evidence for L1–L2 convergence. We discuss potential bi-directional cross-linguistic influences underpinning the L1–L2 convergence and implications for the restructuring of bilingual systems.

Models are no new beasts to scholars of bilingualism. During the last several decades we have seen many interesting and important models that postulate how the bilingual mind works. But specific, computationally implemented, models are far less common than general, verbal, models of bilingualism. This is because the former require efforts on the part of the researcher to conduct algorithmic and representational implementations, whereas the latter do not. The central question is: What good does implementation do in telling us about the bilingual mind beyond what the verbal models do? This Special Issue is an attempt to address this question with seven computational models of bilingualism from different research labs.

Emotional words are better remembered than neutral words in the first language. Ferré, García, Fraga, Sánchez-Casas and Molero (2010) found this emotional effect also for second language words by using an encoding task focused on emotionality. The aim of the present study was to test whether the same effect can also be observed with encoding tasks not related to emotionality, as has been reported in monolinguals. We tested highly proficient bilinguals of Catalan and Spanish that were dominant in one of these two languages. At the encoding phase, we directed their attention to words’ features other than emotionality (participants had to either rate words’ concreteness or count the number of vowels they had). In both cases, we obtained an advantage for emotional words independently of the language in which they appeared. These results suggest that the emotional effect on memory has the same characteristics in the two languages of a bilingual.

Metasyntactic Ability (MSA) refers to the conscious reflection about syntactic aspects of language and the deliberate control of these aspects (Gombert, 1992). It appears from previous studies that heritage-language learners tend to demonstrate lower MSA than their monolingual counterparts (Lesaux & Siegel, 2003). In the present study, we verified whether the same results would be obtained among Portuguese heritage children living in a French-speaking environment when their MSA is measured using two different tasks. The participants were 22 Portuguese heritage children and 22 French monolingual elementary school children (mean age = 10.9 years). Five measurement instruments were used: a reading comprehension task; a language proficiency task; two metasyntactic tasks: a replication task in which the children had to identify and reproduce an error, and a repetition task, in which they had to repeat sentences containing syntactic errors; and a sociodemographic questionnaire. The results showed that when reading comprehension and language proficiency were controlled for, no effect of language background could be observed. However, reading comprehension and language proficiency differently influenced performances on MSA tasks.

To investigate potential causes of L2 performance deficits that correlate with age of onset, we use a computational model to explore the individual contributions of L1 entrenchment and aspects of memory development. Since development and L1 entrenchment almost invariably coincide, studying them independently is seldom possible in humans. To avoid this confound, we study neural network models that learn to solve gender assignment and agreement tasks in Spanish and French. We model the learner as a collection of recurrent cell assemblies that subserve working memory and are facilitated by trainable long-term connections. Varying the time-course over which assemblies and connections are added allows us to compare small, growing, child-like networks to fixed-size adult-like ones. Networks undergo variable-length exposure to L1 before L2 onset to control the amount of L1 entrenchment. This model, by allowing us independent control of both variables, lends us a novel glimpse of all sides of their interaction and affords a rare test of the less-is-more hypothesis. Network comparisons suggest that final L2 proficiency declines as L2 onset delays increase relative to L1, implicating an L1 entrenchment effect. However, aspects of memory development during learning play a key role in mitigating these impairments, lending support to less-is-more as a contributor to sensitive periods.

Muysken's article is a timely call for us to seek deeper regularities in the bewildering diversity of language contact outcomes. His model provocatively suggests that most such outcomes can be subsumed under four speaker optimization strategies. I consider two aspects of the proposal here: the formalization in Optimality Theory (OT) and the reduction of contact outcomes to four basic strategies.

Is parental language mixing related to vocabulary acquisition in bilingual infants and children? Bilingual parents (who spoke English and another language; n = 181) completed the Language Mixing Scale questionnaire, a new self-report measure that assesses how frequently parents use words from two different languages in the same sentence, such as borrowing words from another language or code switching between two languages in the same sentence. Concurrently, English vocabulary size was measured in the bilingual children of these parents. Most parents reported regular language mixing in interactions with their child. Increased rates of parental language mixing were associated with significantly smaller comprehension vocabularies in 1.5-year-old bilingual infants, and marginally smaller production vocabularies in 2-year-old bilingual children. Exposure to language mixing might obscure cues that facilitate young bilingual children's separation of their languages and could hinder the functioning of learning mechanisms that support the early growth of their vocabularies.

Much previous research has pointed to the need for a unified framework for language contact phenomena – one that would include social factors and motivations, structural factors and linguistic constraints, and psycholinguistic factors involved in processes of language processing and production. While Contact Linguistics has devoted a great deal of attention to the structural properties of contact phenomena and their sources in the input languages, the field has made much less progress in attending to Weinreich's observation that language contact can best be understood only “in a broad psychological and socio-cultural setting” (Weinreich, 1953, p. 4). There have been some attempts to establish links between the disciplines that investigate language contact, for example, the psycholinguistic and sociolinguistic (Walters, 2005), and the linguistic and psycholinguistic (Myers-Scotton 2002, Winford 2009, among others). Yet, so far, no one has come close to achieving the kind of integrative, multi-disciplinary framework that Weinreich envisaged. Muysken's paper is therefore a welcome reminder of the need for such a framework, and the complexity of the task involved in constructing it, if indeed it can be accomplished. The introduction to the paper outlines a very ambitious objective – “to explore the possibility of unifying these fields, all different approaches to language contact, creating a single framework within which it is possible to link results from different subfields” (Section 1.1).

We present an original model designed to study how a second language (L2) is acquired in bilinguals at different proficiencies starting from an existing L1. The model assumes that the conceptual and lexical aspects of languages are stored separately: conceptual aspects in distinct topologically organized Feature Areas, and lexical aspects in a single Lexical Network. Lexical and semantic aspects are then linked together during Hebbian learning phases by presenting L2 lexical items and their L1 translation equivalents. The model hypothesizes the existence of a competitive mechanism to solve conflicts and simulate language switching tasks. Results demonstrate that, at the beginning of training, an L2 lexicon must parasitize its L1 equivalent to access its conceptual meaning. At intermediate proficiency, L2 items may evoke their semantics independently of L1, but with a high risk of interference. At higher proficiency, the L2 representation becomes progressively similar to the L1 representation, according to Green's (2003) convergence hypothesis.

Until now, research on bilingual auditory word recognition has been scarce, and although most studies agree that lexical access is language-nonselective, there is less consensus with respect to the influence of potentially constraining factors. The present study investigated the influence of three possible constraints. We tested whether language nonselectivity is restricted by (a) a sentence context in a second language (L2), (b) the semantic constraint of the sentence, and (c) the native language of the speaker. Dutch–English bilinguals completed an English auditory lexical decision task on the last word of low- and high-constraining sentences. Sentences were pronounced by a native Dutch speaker with English as the L2, or by a native English speaker with Dutch as the L2. Interlingual homophones (e.g., lief “sweet” – leaf /liːf/) were always recognized more slowly than control words. The semantic constraint of the sentence and the native accent of the speaker modulated, but did not eliminate interlingual homophone effects. These results are discussed within language-nonselective models of lexical access in bilingual auditory word recognition.

While morphological awareness has received much attention to date, little is understood about how morphological awareness develops within bilingual children learning typologically different languages. Therefore, we investigated children's knowledge of inflections and derivations in Japanese and English, and also asked whether morphological awareness in one language predicted morphological awareness in the other. To that end, 24 Japanese learners of L2 English (ESL) and 21 English learners of Japanese as a heritage language (JHL) were recruited and participated in a range of tasks assessing both vocabulary and morphological knowledge. Cross-linguistic contributions of morphological awareness were identified in both directions (Japanese ↔ English), after controlling for age, IQ, and vocabulary knowledge. This bidirectional transfer was, however, identified only in the ESL group. The group-specific and reciprocal transfer observed is discussed in terms of morphological complexities and relative competence in each language. The potential role of different types of L2 instruction in morphological development is also discussed.

A self-paced reading and translation task was used with learners of English as a second language (L2) to explore what sorts of information L2 learners use during online comprehension compared to native speakers, and how task (reading for comprehension vs. translation) and proficiency affect L2 comprehension. Thirty-six Korean native speakers of English and 32 native English speakers read plausible and implausible subject relative clauses and object relative clauses. Reading times, comprehension accuracy, and translations were analyzed. Results showed that L2 learners were able to use syntactic information similarly to native speakers during comprehension, and that online L2 processing and offline comprehension were modulated by reading goals and proficiency. Results are interpreted as showing that L2 processing is quantitatively rather than qualitatively different from first language processing, i.e. strategically “good enough”.

Muysken argues for four general “strategies” that characterize language contact phenomena across several levels of description. These strategies are (A) maximize structural coherence of the first language (L1); (B) maximize structural coherence of the second language (L2); (C) match between L1 and L2 patterns where possible; and (D) use universal language processing principles. These strategies are seen as choices that bilingual speakers make, individually and collectively, and that are influenced by multiple social, individual, and linguistic factors. This account has the clear advantage of unifying a seemingly very diverse set of language contact phenomena using a limited set of principles. One such phenomenon is cross-linguistic structural priming, the tendency of bilingual speakers to copy grammatical structures from a language recently used to another language (e.g., Hartsuiker, Pickering & Veltkamp, 2004), which Muysken considers an example of “bilingual interference”. In this domain, I will explore how these strategies can be realized in terms of a psycholinguistic processing model, and whether these strategies can be reduced to even more basic principles.

Cross-language priming is a widely used experimental paradigm in psycholinguistics to study how bilinguals’ two languages are represented and organized. Researchers have observed a number of interesting patterns from the priming effects of both translation equivalents and semantically related word pairs across languages. In this study, we implement a self-organizing neural network model, DevLex–II, to simulate these two types of priming effects across Chinese and English. Specifically, our model incorporates a computational mechanism for simulating spreading activation based on the distance between bilingual words in the semantic space. The model also considers additional factors that modulate priming effects, such as the initial activation level of the prime words and the degree to which the target word can be recognized. Our model reveals differences with respect to the priming effects as a function of bilingual type (early versus late L2 learners), directions of priming (L1 to L2 versus L2 to L1), and types of priming (translation versus semantic priming). These simulated differences are compared with empirical findings from previous studies and discussed in the light of interactive and developmental theories of bilingual lexical representation.

Pieter Muysken's keynote paper, “Language contact outcomes as a result of bilingual optimization strategies”, undertakes an ambitious project to theoretically unify different empirical outcomes of language contact, for instance, SLA, pidgins and Creoles, and code-switching. Muysken has dedicated a life-time to researching, rather successfully, various sub-fields of language contact, so I am very pleased to see him develop a synergistic model that reduces the complexities of different bilingual contact phenomena to four optimization strategies, the specific permutations of which yield the different, linguistically significant, generalizations. Such attempts are necessary, certainly, if the field of language contact has to make progress, theoretically. The success of such a theoretical unification, however, depends to a large extent on (i) the empirical mileage such unification receives; (ii) how well the assumptions underlying the logic of unification are theoretically motivated, to yield precise predictions about the orderliness of bilingual behavior; and (iii) the conceptual clarity required to understand the various links among the outcomes of language contact. On all these counts, Muysken's paper comes close to achieving success, though one notices several areas of fuzziness that need to be addressed for a competent model to fully emerge. In this short essay, I will point out two areas that need theoretical attention so that subsequent revisions of the present version of the model can address them. I will restrict my comments to code-switching, an extremely productive area of language contact, with which I am most familiar.

This study compares mastery of gender assignment and agreement in Italian by adult Italian–German bilinguals who have acquired two languages simultaneously (2L1), and by adult German highly proficient second language learners (L2ers) of Italian. Our data show that incompleteness in bilingual acquisition and in second language (L2) acquisition primarily affects gender assignment: the categorization of nouns and the interpretable gender feature are subject to vulnerability in the two modalities of acquisition. Overall, mastery of morpho-syntax (i.e., gender agreement) was nearly native-like for both groups of speakers, suggesting that uninterpretable features are unlikely to be subject to vulnerability in the heritage language of adult bilingual speakers and can be acquired in adult L2 acquisition. Deviances from the target in gender assignment and, to a lesser extent, in gender agreement are attributed to both language-internal (i.e., language) and language-external factors (i.e., amount of input).

Spanish [l] is characterized as clear, and is associated with a high second formant (F2) frequency and a large difference between F2 and the first formant (F1) frequencies. In contrast, English [l] is darker (with a lower F2 and a relatively smaller F2–F1 difference) and also exhibits contextual variation due to an allophonic velarization rule that further darkens [l] postvocalically. We aimed to determine if Spanish–English bilingual children evidence these differences productively, in a manner comparable to that of monolinguals, or if they produce an [l] that is intermediate to that of Spanish and English monolinguals. We acoustically analyzed [l] productions of seven Spanish–English bilingual, seven Spanish monolingual, and seven English monolingual children. Results showed that the bilinguals had similar prevocalic F2 and F2–F1 values for [l] in both languages, comparable to those of Spanish monolinguals, but significantly higher than those of English monolinguals. The bilinguals also produced English (but not Spanish) [l] with significantly lower postvocalic F2 and F2–F1 values. We assume that the bilinguals have a merged phonetic category for prevocalic [l] but not postvocalic [l], and further, that they maintain separate grammars, allowing the allophonic velarization rule to apply in English but not Spanish.

During speech comprehension, bilinguals co-activate both of their languages, resulting in cross-linguistic interaction at various levels of processing. This interaction has important consequences for both the structure of the language system and the mechanisms by which the system processes spoken language. Using computational modeling, we can examine how cross-linguistic interaction affects language processing in a controlled, simulated environment. Here we present a connectionist model of bilingual language processing, the Bilingual Language Interaction Network for Comprehension of Speech (BLINCS), wherein interconnected levels of processing are created using dynamic, self-organizing maps. BLINCS can account for a variety of psycholinguistic phenomena, including cross-linguistic interaction at and across multiple levels of processing, cognate facilitation effects, and audio-visual integration during speech comprehension. The model also provides a way to separate two languages without requiring a global language-identification system. We conclude that BLINCS serves as a promising new model of bilingual spoken language comprehension.