ABSTRACT This second paper in a two‐part methodological guide demonstrates how Generalised Linear Mixed Model (GLMM) tree analysis can be used to explore linguistic conditioning in sociolinguistic variation. Building on Part 1, which introduced the dataset and illustrated how GLMM trees reveal social patterning in (ING) variation, Part 2 focuses on the internal linguistic factors governing the alternation between [ɪn] and [ɪŋ] in Canadian Maritime English (CME). Using over 10,000 tokens from more than 300 speakers—the largest single‐dialect (ING) dataset to date—this study shows how GLMM trees accommodate overlapping or collinear predictors such as grammatical category, phonological context, and lexical frequency. The analysis confirms the dominant morphological constraint (verbs favouring [ɪn], non‐verbal forms favouring [ɪŋ]) and identifies additional phonological and frequency effects that emerge only in specific prosodic and lexical environments. These results demonstrate that GLMM trees offer a straightforward, statistically robust way to model complex sociolinguistic data while maintaining interpretability. The paper concludes by comparing hierarchical tree‐based and ‘flat’ regression models, validating the stability of the GLMM trees through cross‐validation, and highlighting how this approach can clarify the interplay of linguistic and social constraints in variationist research.

ABSTRACT Knowledge of public attitudes towards language‐based diversity can help uncover wider societal prejudices. Nonetheless, sociolinguists are increasingly aware of the limitations of using traditional explicit self‐report measures to gauge attitudes towards linguistic variation. Accordingly, a growing number of studies have adapted more recently developed socio‐psychological techniques—most particularly the implicit association test (IAT)—to access more deeply embedded, and durable, implicit evaluations of linguistic stimuli. This article provides a critical review of the methodological approaches undertaken and findings of existing prior explicit and implicit language attitude research, and examines the nature of the relationship between the two levels of linguistic evaluation. The methodological and theoretical merits of employing both implicit and explicit attitudinal measures in future language attitude research is discussed, in particular regarding more fine‐grained assessment of language attitude change in progress and to better understand the relationship between any changes underway in linguistic attitudes and patterns of language use and language change within the speech community under investigation.

ABSTRACT Inter‐turn silences are silences that occur at turn transitions in conversation. They mark the gap between the end of one speaker's turn and the beginning of the next speaker's turn. We provide a brief review of the literature on inter‐turn silences. The literature suggests that inter‐turn silences tend to be short (< 1 s in duration), with a large proportion of the silences being less than half a second in length. A range of cultural and contextual factors affect the length of inter‐turn silences. Interlocutors have expectations about the lengths of inter‐turn silences, and silences that are longer than expected tend to be given a negative interpretation (e.g., assumptions of dishonesty, disinterest, or disagreement). We suggest that prediction‐based accounts of language processing can provide a reasonable starting point for understanding both the variation that exists in the duration of inter‐turn silences and the way that the silences are interpreted. We also suggest that theories of time perception could be used to better understand the interpretation of inter‐turn silences as well as potential brain regions involved in both timing and turn transitions.

ABSTRACTHow can modern neural networks like language models be useful to the field of language acquisition, and more broadly cognitive science, if they are not a priori designed to be cognitive models? As developments towards natural language understanding and generation have improved leaps and bounds, with models like GPT‐4, the question of how they can inform our understanding of human language acquisition has re‐emerged. As such, it is critical to examine how in practice linking hypotheses between models and human learners can be safely established. To address these questions, we propose a model taxonomy, including four modelling approaches, each having differing goals, from exploratory hypothesis generation to hypothesis differentiation and testing. We show how the goals of these approaches align with the overarching goals of science and linguistics by connecting our taxonomy to the realist versus instrumentalist approaches in philosophy of science. We survey recent work having adopted each of our modelling approaches and address the importance of computational modelling in language acquisition studies.

Abstract Phonetic data are used in several ways outside of the core field of phonetics. This paper offers the perspective of one such field, sociophonetics, towards another, the study of acoustic cues to clinical depression. While sociophonetics is interested in how, when, and why phonetic variables cue information about the world, the study of acoustic cues to depression is focused on how phonetic variables can be used by medical professionals as tools to diagnosis. The latter is only interested in identifying phonetic cues to depression, while the former is interested in how phonetic variation cues anything at all. While the two fields fundamentally differ with respect to ontology, epistemology, and methodology, I argue that there are, nonetheless, possible avenues for future engagement, collaboration, and investigation. Ultimately, both fields need to engage with Crip Linguistics for any successful intervention on the relationship between speech and depression.

Abstract Deepfakes, particularly audio deepfakes, have become pervasive and pose unique, ever‐changing threats to society. This paper reviews the current research landscape on audio deepfakes. We assert that limitations of existing approaches to deepfake detection and discernment are areas where (socio)linguists can directly contribute to helping address the societal challenge of audio deepfakes. In particular, incorporating expert knowledge and developing techniques that everyday listeners can use to avoid deception are promising pathways for (socio)linguistics. Further opportunities exist for developing benevolent applications of this technology through generative AI methods as well.

Abstract With a focus on the post‐2015 period in the western and northern regions of Europe, the research examined here shows prominent media discourses of othering, threat and deservedness of migrants. This spatial and temporal frame lends itself to the study of how discourses reflect the impact of the so‐called ‘refugee crisis’ in these regions of Europe. Since there was also continued immigration related to increased opportunities for work, education, quality of life and family togetherness which have long brought migrants to these European countries, examining the research in this period allows us to discover how these discourses might distinguish between different migrant experiences. There is some evidence for the differentiation of certain types of people of migration background in the media discourses, despite a strong tendency to stereotype and essentialise regardless of the actual background of migrants or their descendants. Another key aspect in the research to date is how professional versus participatory media can be compared in the discourses of migration they reproduce, and how these different types of media play a role in society. The article ends with a call for a more intersectional perspective on migration which incorporates critical perspectives on racialisation, and further examination of the voices of migrants in the media.

No abstract is available for this article.