Bilinguals engage in qualitatively different code-switching patterns (alternation, insertion, and congruent lexicalization) to different degrees, according to their engagement in different types of interactional contexts (single-language context, dual-language context, and dense code-switching context). Drawing on the adaptive control hypothesis, we examined whether bilinguals’ code-switching patterns would differentially shape multiple aspects of cognitive control (interference control, salient cue detection, and opportunistic planning). We found that a dense code-switching context, which predominantly involves insertion and congruent lexicalization, was positively associated with verbal opportunistic planning but negatively associated with interference control and salient cue detection. In contrast, a dual-language context, which predominantly involves alternation, was not associated with interference control or salient cue detection, but with significantly reduced response times for opportunistic planning. Our findings partially corroborate the theoretical predictions of the adaptive control hypothesis. Altogether, our study illustrates the importance of bilinguals’ disparate code-switching practices in shaping cognitive control outcomes.

In two experiments, we examine how proficient second language speakers integrate verb bias and plausibility information during online sentence comprehension. Spanish–English speakers and native English speakers read sentences in English in which a post-verbal noun phrase (NP) could be interpreted as a direct object or a sentential subject. To examine the role of verb bias, the post-verbal NP was preceded by a verb that is preferentially followed by a direct object (DO-bias verbs) or a sentential complement (SC-bias verbs). To assess the role of plausibility, the semantic fit between the verb and the post-verbal NP was either congruent or incongruent with the direct object interpretation. The results show that both second language speakers and native speakers used verb bias information to assign a grammatical role to the post-verbal ambiguous NP with small differences. Syntactic revision of an initially incorrect DO interpretation was facilitated by the presence of an implausible NP.

In the current study, we examined the role of first-language (L1) influence on the additional-language development of grammatical gender marking in Spanish. The participants were L1 speakers of English or French (N = 215), who were learning Spanish and who were at three instructional levels. The data came from their use of gender marking in noun-modifier pairs in an argumentative essay. We adopted the unified methodological framework developed by Scott Jarvis and we applied insights from variationist second language acquisition to contribute to the discussion about whether learners’ L1 impacts variability in targetlike gender marking in additional-language Spanish. Specifically, we designed our study to investigate four types of evidence that Jarvis identified (intragroup homogeneity, intergroup heterogeneity, cross-language congruity, and intralingual contrasts), and we used variationist methods to account for other factors that are known to impact variable use of gender marking. The quantitative analyses supported each type of evidence, consequently demonstrating that these learners’ L1 influenced their variable use of gender marking in Spanish. We concluded by reflecting on the contributions that the current study has made to the understanding of gender marking in additional-language Spanish and to research on L1 influence more generally.

Children's remarkable ability to generalize beyond the input and the resulting overregularizations/ irregularizations provide a platform for a discussion of whether morphology learning uses analogy-based, rule-based, or statistical learning procedures. The present study, testing 115 children (aged 3 to 10) on an elicited production task, investigated the acquisition of the irregular distribution in the Turkish causative. Results showed that in early acquisition, to pin down the four causative suffixes, children engaged in comparisons between analogous exemplars. Thereafter to tackle the irregularity in two of the suffixes, children entertained competing hypotheses that yielded overregularizations and irregularizations. Overregularizations were instances of abstraction across the input based on type frequency; irregularizations were attempts to default to erroneous micro-generalizations. Negative correlation between errors and verb frequency suggested that recovery from errors was sensitive to token frequency. The overgeneralize-then-recover pattern that emerged in the acquisition of causative supported an integrated account of the roles of analogy, abstraction, and frequency in morphology learning.

In this study, we investigate the process of generating single-sentence representations for the purpose of Dialogue Act (DA) classification, including several aspects of text pre-processing and input representation which are often overlooked or underreported within the literature, for example, the number of words to keep in the vocabulary or input sequences. We assess each of these with respect to two DA-labelled corpora, using a range of supervised models, which represent those most frequently applied to the task. Additionally, we compare context-free word embedding models with that of transfer learning via pre-trained language models, including several based on the transformer architecture, such as Bidirectional Encoder Representations from Transformers (BERT) and XLNET, which have thus far not been widely explored for the DA classification task. Our findings indicate that these text pre-processing considerations do have a statistically significant effect on classification accuracy. Notably, we found that viable input sequence lengths, and vocabulary sizes, can be much smaller than is typically used in DA classification experiments, yielding no significant improvements beyond certain thresholds. We also show that in some cases the contextual sentence representations generated by language models do not reliably outperform supervised methods. Though BERT, and its derivative models, do represent a significant improvement over supervised approaches, and much of the previous work on DA classification.

Most people love metaphor, but we still sometimes find ourselves resisting their presence or meanings for various reasons. We resist metaphors both as a general strategy (e.g., “Metaphors are meaningless” or “Mixed metaphor are incoherent”), and as a response to some metaphors in very specific situational and discourse contexts (e.g., “I do not like the idea that my cancer treatment is seen as a war against my body”). People resist metaphors they have produced, metaphors imposed on them by others, and metaphors that they find to be offensive or that negatively stigmatize other individuals, or groups of people. But metaphors are also resisted for their lack of explanatory power in, for instance, scientific communities. There are also many ironies associated with metaphor resistance, such as consciously resisting some metaphor while still being governed by that same metaphor in our unconscious thinking and actions. Most generally, though, metaphor resistance is its own kind of metaphorical action. Taking a dynamic systems approach to resistance to metaphors, we discuss several implications of these observations for theories of metaphorical thought and language.

The last three decades have witnessed increasing interest in discourse-pragmatic markers (DPMs), both with regards to their high frequency in spoken discourse and their multifunctionality in interaction. Most studies have centered on English, with studies on Danish restricted to a handful of previous interactional discourse analyses. This paper is a preliminary investigation of the Danish word sådan (commonly glossed as ‘such’ or ‘like this/that’). A qualitative, form-based, discourse analytic approach is undertaken on over 40 minutes of naturally occurring Danish talk to argue that sådan qualifies as a DPM. In service of textual, subjective, and intersubjective macro-functions, sådan illustrates; exemplifies; marks hesitation; approximates a quantity; mitigates, hedges, or softens; and allows self-correction or self-repair. These findings are discussed in terms of their implications for sådan’s place in the Danish DPM system and our understanding of DPMs across languages.

The sociodemographic typology of sign languages classifies them based on the characteristics and configurations of their users. When considering homesign and sign languages in rural areas, this typology needs further refinement. Here, I present new concepts to enable this. The study is based on fieldwork with twelve deaf people in Western Highlands, Papua New Guinea, and review of studies worldwide. Sign language communities can be mapped as sign networks. Using this mapping, I propose a new typological category for languages with one central deaf user and many fluent hearing signers: nucleated network sign language. I use sign base analysis to determine lexical consistency between unconnected deaf signers in Western Highlands. The high level of consistency among largely unconnected deaf people is explained by a regional sign network connecting deaf and hearing signers. This research emphasises the role of both deaf and hearing signers in sign language emergence and maintenance. (Sign languages, social networks, sign networks, typology, homesign, rural sign languages, Papua New Guinea)*

A compelling ecological theory of movement and vigor must explain why humans and other animals spend so much time not moving. When we rest, our somatic maintenance systems continue to work. When our somatic maintenance requirements increase, we place greater subjective value on resting. To explain variation in movement and vigor, we must account for the subjective value of resting.

In active agents, sensory and motor processes form an inevitable bond. This wedding is particularly striking for saccadic eye movements – the prime target of Shadmehr and Ahmed's thesis – which impose frequent changes on the retinal image. Changes in movement vigor (latency and speed), therefore, will need to be accompanied by changes in visual and attentional processes. We argue that the mechanisms that control movement vigor may also enable vision to attune to changes in movement kinematics.

Modelling semantic similarity plays a fundamental role in lexical semantic applications. A natural way of calculating semantic similarity is to access handcrafted semantic networks, but similarity prediction can also be anticipated in a distributional vector space. Similarity calculation continues to be a challenging task, even with the latest breakthroughs in deep neural language models. We first examined popular methodologies in measuring taxonomic similarity, including edge-counting that solely employs semantic relations in a taxonomy, as well as the complex methods that estimate concept specificity. We further extrapolated three weighting factors in modelling taxonomic similarity. To study the distinct mechanisms between taxonomic and distributional similarity measures, we ran head-to-head comparisons of each measure with human similarity judgements from the perspectives of word frequency, polysemy degree and similarity intensity. Our findings suggest that without fine-tuning the uniform distance, taxonomic similarity measures can depend on the shortest path length as a prime factor to predict semantic similarity; in contrast to distributional semantics, edge-counting is free from sense distribution bias in use and can measure word similarity both literally and metaphorically; the synergy of retrofitting neural embeddings with concept relations in similarity prediction may indicate a new trend to leverage knowledge bases on transfer learning. It appears that a large gap still exists on computing semantic similarity among different ranges of word frequency, polysemous degree and similarity intensity.

Motivated by a series of interconnected studies on simultaneous attention to form and meaning, we revisit L2 learners’ real-time processing of text by using eye-tracking as an unobtrusive method to provide concurrent data on attention allocation. Seventy-five L2 Spanish learners were instructed to attend to an assigned form in a reading passage and to press a button when they noticed it. After reading the passage, the learners answered 10 multiple-choice comprehension questions. The participants’ responses to the comprehension questions and their reading behaviors reflected in eye-movement data suggest that attention to grammatical form may hinder L2 learners’ simultaneous attention to form and meaning. However, individual differences in global text processing contributed to the differences in the participants’ text-comprehension scores over and above the task instruction to attend to form: Slower L2 readers who read the passage more carefully showed better text comprehension.

Normative ratings are a means to control for the effects of confounding variables in psycholinguistic experiments. This paper introduces a new dataset of normative ratings for Swedish encompassing 111 concrete nouns and the corresponding picture stimuli in the MultiPic database (Duñabeitia et al. 2017). The norms for name agreement, category typicality, age of acquisition and subjective frequency were collected using online surveys among native speakers of the Finland-Swedish variety of Swedish. The paper discusses the inter-correlations between these variables and compares them against available ratings for other languages. In doing so, the paper argues that ratings for age of acquisition and subjective frequency collected for other languages may be applied to psycholinguistic studies on Finland-Swedish, at least with respect to concrete and highly imageable nouns. In contrast, norms for name agreement should be collected from speakers of the same language variety as represented by the subjects in the actual experiments.

Previous work has shown that stance—the way speakers position themselves with respect to what they are talking about and who they are talking to—provides powerful insights into why speakers choose certain linguistic variants, beyond correlations with macro-social categories such as gender, ethnicity, and social class. However, as stancetaking moves are highly context-dependent, they have rarely been explored quantitatively, making the observed variable patterns difficult to generalize. This article seeks to contribute to this methodological gap by proposing a formal guide to coding stance and demonstrating how it can be operationalized quantitatively. Drawing on a corpus of eight individuals, self-recorded in three situations with varying levels of social distance, we apply this method to variation between English complementizers that and zero (i.e. no overt complementizer), providing a replicable and theoretically grounded protocol that incorporates both quantitative and qualitative analyses in a variationist sociolinguistic study. (Stance, complementizers, that, English)*

Cet article analyse, dans une perspective comparative, deux classes de marqueurs de reformulation en français et en russe. La première est construite autour d’un indéfini en autre (rus. inoj) : fr. autrement dit, en d’autres termes ; rus. inače govorja, inymi slovami ; la seconde est historiquement dérivée d’une clause verbale finie ce + est (rus. to + estʹ) : fr. c’est-à-dire, rus. to estʹ. Il sera démontré que le français et le russe partagent bon nombre de propriétés dans le fonctionnement de ces marqueurs, qu’il s’agisse de leur fonctionnement grammatical ou sémantico-pragmatique. Il sera aussi argumenté que les marqueurs de la deuxième classe font l’objet d’un stade de grammaticalisation, mais aussi de pragmaticalisation, plus avancé et sont, par conséquent, plus appropriés à des interactions spontanées que les marqueurs construits autour d’un indéfini en autre.

Au sein de l’enseignement de l’oral, il est souvent difficile de convaincre les didacticiens du français qu’un recours à la phonologie et à la phonétique faisant place à la variation et la diversité des usages est possible et souhaitable. Pourtant, dès la fin du 19ème siècle, une autre pratique était défendue par des jeunes phonéticiens dont Paul Passy, auteur des Sons du français (1887) et coordinateur de la mise en place de l’API. Passy, à cause de ses engagements socio-politiques et religieux, de sa défense souvent mal comprise de l’API et de son combat pour la réforme de l’orthographe, est fréquemment relégué à une place relativement mineure dans l’histoire de la phonétique du français. Nous démontrons ici que les nombreuses révisions des Sons du français fourmillent de descriptions fines et de remarques sur des changements en cours qui, dans bien des cas, seront redécouverts dans la deuxième partie du 20ème siècle sans mention de son travail. Nous abordons alors la question de la norme et l’attachement de Passy à la variation et montrons sa modernité à travers son traitement de la prosodie, des unités distinctives, des réalisations allophoniques et de la transcription phonétique.