Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
18625 papers
The Junction Grammar of English A Reproducible Corpus-Wide Analysis of Morpheme-Boundary Statistics and Consonant–Vowel Information Asymmetry Boicho Dimitrov Temelakiev Saxon Ventura Research Ltd 28th of May,2026 CC BY Abstract This paper reports a reproducible, corpus-wide statistical analysis of English word structure derived entirely from a single public word list of 455,246 entries, computed in a spreadsheet with no specialized tooling. The morpheme boundary—the junction between a stem and an affix—is treated as the primary object of measurement, and the distribution of the letters that may occupy each side of a junction is read across the corpus. Three results are established. First, the junction carries a stable structure: a consonant backbone (T, L, N, R, S, I) admitted by nearly all suffixes, an absolute floor (J, Q) admitted by none, and a sparsity that scales inversely with a suffix’s productivity. Second, the relationship between any two affixes is quantified by the correlation of their boundary distributions, which measures the degree to which they share a stem population; this correlation ranges from ~0.99 for etymological doublets to 0.57 for productivity-asymmetric near-twins. Third, the written word decomposes into a consonant skeleton carrying lexical identity (53.0% of the vocabulary is uniquely recoverable from consonants alone) and a vowel tissue carrying grammatical form (1.8% recoverable from vowels alone), a ~29-fold information asymmetry. Multiple independent measures—consonant recoverability, derivational classmarking, and free-stem fraction—partition the lexicon at a single boundary separating a transparent Germanic core from a bound Latinate superstructure. The method, its corrections, and its limits are reported in full, and a program of remaining work is stated. 1. Corpus and Method All results derive from one corpus analysed by one elementary procedure, and the reproducibility of that procedure is treated as part of the contribution. The corpus is the dwyl/english-words list (words_alpha.txt), comprising 455,246 alphabetic entries with a total of 4,254,354 letter occurrences and a mean word length of 9.345 letters. Each letter is assigned its ordinal value (A=1 through Z=26). Words bearing a given suffix are isolated by end-anchored matching and aligned on their final letter, so that each suffix position returns its exact ordinal value as a sanity check; the first stem letter preceding the suffix—the linker—is then read as a full A–Z frequency distribution rather than as a mean. The governing methodological constraint is that distributions are read in full and never collapsed to a mean prematurely, that no numerical coincidence is treated as a finding until tested across many cases, and that every claim is backed by a count. Page 1 Method: a derived column applies the ordinal map; end-anchored COUNTIF and MID/CODE formulas extract suffix positions and the linker; per-letter tallies at the linker yield the boundary distribution. Nesting among suffixes (for example ‑MENT within ‑ENT, or ‑ATION within ‑TION within ‑ION) is controlled by excluding longer relatives before counting. No statistical software, machine-learning library, or external lexical database is used at any stage of the core analysis. The constraint that the analysis remain computable by elementary means is not incidental; it ensures every figure in this paper can be independently reconstructed from the public corpus with a spreadsheet alone. 2. The Junction and Its Backbone The investigation began as a survey of unconditioned letter bigrams, which returned no morphological signal; structure appeared only when letter distributions were conditioned on a single morpheme boundary, and that conditioning is the method’s foundation. A preliminary tabulation of adjacent letter pairs across the corpus—which letters follow which, without regard to position within the word—yielded frequency patterns reducible to general orthographic regularities and carrying no isolable morphological content. The signal emerged only when a specific suffix was fixed and the letters preceding it were read as a population: conditioning on the junction, rather than measuring adjacency as such, is what renders the boundary structure visible. The set of letters that may legally occupy the stem side of a morpheme boundary then proves narrow, structured, and stable across suffixes. Reading the linker distribution across the mapped suffix inventory reveals a three-tier structure. A backbone of six letters—T, L, N, R, S, and I—is admitted at high frequency by nearly every suffix; T is the single most frequent linker across the inventory and recurs as the dominant boundary letter in suffix after suffix. An absolute floor of two letters—J and Q—is admitted by no suffix at measurable frequency, a prohibition confirmed corpus-wide and consistent with their status as the two rarest letters overall (J at 0.18%, Q at 0.19% of all letter occurrences). Between backbone and floor lies a selective middle whose composition varies by suffix and in which each suffix’s identity resides. Two descriptive terms are used throughout. Because each letter carries an ordinal value (A=1 through Z=26), a linker distribution may be summarized by where its mass falls on that scale: a distribution concentrated on early-alphabet letters (low ordinal values, A through roughly M) is termed cold, and one concentrated on late-alphabet letters (high ordinal values, roughly N through Z) is termed warm. The terms refer solely to ordinal position on the A–Z scale and carry no semantic content; the backbone letters, for instance, span both ends (cold I and L against warm N, R, S, T). The degree of selectivity is itself a measurement. The count of forbidden letters at a junction—its sparsity—scales inversely with the suffix’s productivity: derivational suffixes that attach choosily to a constrained stem class forbid many letters, whereas inflectional or highly productive suffixes forbid few. Sparsity is therefore not noise but signal: the pattern of exclusion characterizes the suffix as informatively as the pattern of admission. Page 2 Method: forbidden-letter counts are read directly from the linker distribution (frequency below 0.5% taken as floor); cross-checks against COUNTIF totals confirm the populations; J/Q absence is verified against the bare ‑S population of 84,208 words and against corpus-wide letter frequencies. The junction thus behaves as a filter whose admitted and forbidden letters together encode the morphological role of the boundary, with the consonant backbone bearing the structural load and the rare letters marking its limits.
Abstract Cognitive reappraisal is an emotion regulation strategy that involves changing one’s interpretation of a situation to change its emotional impact. Cognitive reappraisal studies typically assess changes in valence ratings following reappraisal as the primary dependent variable, but scholars have recently documented that valence ratings can take two forms: affective valence (i.e., the positivity/negativity of one’s hedonic internal feelings) and semantic valence (i.e., the positivity/negativity of one’s cognitive evaluation of the stimulus). Because typical rating instructions do not distinguish between these forms of valence, it remains unknown how strongly cognitive reappraisal shifts either one. We address this gap through two preregistered studies. In Study 1, N = 155 participants completed a classic cognitive reappraisal task in which they either responded naturally to or cognitively reappraised negative images. Critically, we manipulated whether participants rated either affective valence, semantic valence, or “default” valence (conventional instructions). Reappraisal significantly improved valence ratings in all conditions, but reappraisal most strongly influenced semantic valence and least strongly affective valence. Interestingly, default valence behaved most similarly to semantic valence, suggesting that existing reappraisal studies largely document changes in semantic evaluations, more than affective experiences. Study 2 ( N = 105) replicated these effects within-subjects. These results extend research on affective and semantic valence by demonstrating that they vary in their responsiveness to regulation. Additionally, they suggest that existing research likely overestimates the impact of reappraisal on felt emotional experiences, prompting new perspectives on past findings and more focused methods for assessing the hedonic impact of emotion regulation.
ABSTRACT Emotional episodes involve partially coordinated changes in experience and autonomic activity, and subjective–physiological coherence appears to vary with aging and physiological stress. Physiological accounts of emotional aging propose that age‐related changes in autonomic and interoceptive function reduce the informativeness of bodily signals, weakening the coupling between feelings and bodily responses. We test whether subjective–physiological coherence is lower in older vs. younger normoxic adults, lower in hypoxic vs. normoxic younger adults, and similar in hypoxic younger adults and older normoxic adults; in secondary mechanism‐oriented analyses, we examine whether heart‐rate variability and cardiac interoceptive accuracy are statistically related to group differences in coherence. We will test these questions in three groups ( N = 120): younger adults in normoxia, younger adults under acute hypoxia (FiO 2 ≈ 12%), and older adults in normoxia. Participants will view emotional images and provide trial‐wise ratings of affect and approach motivation while heart rate and skin conductance are recorded. Emotional coherence will be indexed primarily by within‐person correlations between subjective arousal ratings and autonomic reactivity, with approach motivation and valence examined as complementary subjective dimensions.
ABSTRACT This study examines how Arabic language teachers in Indonesian madrasahs negotiate transnational Islamic authority and epistemic dependency within curriculum practices. Drawing on qualitative interviews, the research explores how global centres of Islamic knowledge—particularly Middle Eastern institutions, textbooks and linguistic norms—shape local understandings of legitimacy, authenticity and pedagogical authority. Rather than indicating structural domination, the findings show that epistemic dependency operates primarily as symbolic and referential alignment with perceived centres of authority. Teachers retain curricular autonomy and actively mediate these influences through contextual adaptation, aligning instruction with Indonesian cultural values, national educational frameworks and local religious traditions. These practices demonstrate negotiated forms of authority in which transnational standards function as reference points rather than mechanisms of control. The study contributes to discussions of language, curriculum and epistemic hierarchy in Global South contexts by highlighting how authority is symbolically structured yet locally mediated in practice.
AI safety systems are increasingly deployed in high-stakes domains such as mental health, yet they are built on implicit assumptions about language that do not hold in practice. We argue that AI safety systems implicitly decide whose language counts as interpretable, and that risk expressed outside those norms is systematically overlooked. At the same time, AI systems can learn new language faster than their safety mechanisms can adapt to it. This creates a structural gap: while models may recognize emerging expressions, safety guardrails remain anchored to static linguistic norms. We introduce the concept of linguistic lag to describe the delay between the emergence of meaning within a community and its recognition by safety systems. Through examples and empirical observations, we show that semantically equivalent expressions of distress yield divergent safety outcomes when expressed in adolescent digital dialects. We argue that this is not a data coverage problem but a consequence of mismatched adaptation dynamics: language evolves through rapid, decentralized social diffusion, while safety systems evolve
This study examines how everyday interactions in public spaces shape belonging, social cohesion, and health equity in Holligen, a culturally diverse neighbourhood in Bern, Switzerland. Drawing on focused ethnographic observation and brief in-situ interviews, we identify subtle indicators of inclusion and exclusion: small visual, behavioural, spatial, and affective cues through which residents come to feel recognised, entitled to participate, or marginalised. The findings show that social cohesion is asymmetrical. While parks, playgrounds, food events, and community spaces enabled co-presence and everyday cross-cultural recognition, interaction often remained clustered by language, generation, and familiarity. Exclusion was produced not only through physical barriers, but also through cultural meanings attached to place, linguistic norms, and informal gatekeeping within community networks. We argue that health equity depends not only on access to public space, but also on whether spaces are emotionally legible, symbolically welcoming, and socially navigable. Methodologically, the study demonstrates the value of ethnography for making visible the everyday cues of belonging that are often missed by access-based urban health metrics.
This article examines digital transformation of scientific discourse from the perspective of new models in information and communication space. The study aims to identify structural, linguistic, and multimodal characteristics of scientific texts in digital format using content analysis of ten scientific articles published in Kazakh in humanities and social sciences, published in 2023-2025 online in journals listed by the Committee for Quality Assurance in Science and Higher Education of Kazakhstan. The study coded parameters such as thematic area, annotation and keyword system, compositional structure of the text, terminological consistency, geography of sources, frequency of use of visual elements (tables, diagrams, illustrations) and hyperlinks, and conducted a quantitative and qualitative analysis. The study found that the IMRAD-based model is preserved in digital scientific discourse in Kazakh, but the text acquired a networked and interactive character due to open access, DOIs, online platform capabilities, and multimodal components. A number of articles noted incomplete use of hypertext links and visual tools in the digital format, inconsistency in presentation of sources, and violations of linguistic norms. The results helped to assess the level of scientific discourse digital modernization in the national linguistic space and describe new models of scientific communication in information and communication space.
With the increasing popularity and feasibility of implementing Ecological Momentary Assessment (EMA), research on affective dynamics has expanded considerably (1–3). In parallel, advances in artificial intelligence (AI) have enabled scalable extraction of rich multimodal features (e.g., facial, vocal, and linguistic) from video data, which may serve as adjunct complementary behavioural indicators to subjective ratings of affective states typically captured in EMAs. Emerging research has demonstrated the promising utility of these features in predicting the diagnostic status of mental health problems and the presence of transdiagnostic symptomology (4–6). However, how these behavioural indicators relate to momentary affect in naturalistic settings, and whether they explain unique variance in mental health symptoms beyond self-reported mood, remains unexplored. Given the recent rise in perinatal depression and anxiety (7), exploring these questions in the perinatal period will inform the potential clinical value of implementing naturalistic video screeners of postpartum mental health symptoms in perinatal care provision. The present study therefore has two primary aims: 1) to examine the extent to which multimodal behavioural indicators map onto momentary affect ratings across a one-week EMA protocol (3 pings daily) in birthing parents who are 1-6 months postpartum; 2) to test whether behavioural indicators explain variance in mental health symptoms above and beyond self-reported momentary mood, thereby evaluating their incremental validity as markers of emotional functioning.
Negative biases in emotional processing are central to cognitive models of Major Depressive Disorder (MDD), yet it remains unclear whether negatively biased valence attribution covaries with depressive symptoms or represents a more stable vulnerability. In this longitudinal study, we examined emotional valence ratings during a subliminal affective priming task in n = 232 MDD patients and n = 496 healthy controls (HC's) across multiple follow-up assessments, yielding 1395 observations. Multilevel models tested effects of depressive symptom severity, diagnostic group (MDD vs. HC), affective prime (happy, sad, neutral), and their interactions. To separate within-person fluctuations from between-person differences, depressive symptom severity was decomposed (Mundlak within-between specification). Exploratory analyses examined whether long-term symptom trajectory clusters were associated with emotional bias. Greater depressive symptom severity was associated with more negative valence ratings. Within-between analyses indicated that this association was primarily driven by within-person symptom fluctuations, indicating more negative ratings when symptom severity exceeded individual average. In contrast, diagnostic group, prime condition, and symptom severity × prime interactions did not significantly predict valence ratings, and no group differences emerged between HC's and remitted MDD patients. Exploratory trajectory analyses suggested that patients with recurrent symptom patterns may show more negative valence ratings, although cluster stability was limited and findings should be interpreted cautiously. Overall, the results suggest that negative valence attribution in MDD is more closely linked to current depressive symptom severity than to diagnostic status alone. Longitudinal approaches separating within- and between-person processes may help clarify when negative biases emerge and persist during depression progression.
In the contemporary world, social media have become a key space for communication, information exchange, and self-presentation. However, communication within this environment is not uniform across all age groups; rather, it varies depending on age-related characteristics, which are reflected in language choice, forms of interaction, and communicative strategies. The article examines the communicative behavior of X, Y, Z, and Alpha age cohorts within the Kazakh-speaking social media landscape (Instagram, TikTok, Facebook). The study aims to identify differences in language choices and communicative strategies among representatives of different generations. To collect empirical data, a specially designed survey method was employed, and the results were subjected to statistical and linguistic analysis. The survey results indicate that the older generations (X, Y) primarily prioritize informational and professional content, whereas the younger generations (Z, Alpha) lean towards multimodal and expressive-emotional communication. The study describes the development of skills for concise and brief expression of thoughts under the influence of social media, as well as the transformation of linguistic norms (slang, emojis, hybrid style). The findings of the research contribute to the theory of Kazakh-language digital discourse in the fields of media linguistics and sociolinguistics.
Automated item generation (AIG) using large language models (LLMs) is emerging as a promising solution for scalable creativity assessments, yet the potential of AIG to create psychometrically rigorous creativity items has thus far received little attention. In the present study, we evaluated the psychometric properties of LLM-generated items for the classic Consequences task of divergent thinking, comparing them to human-written items. A sample of 192 undergraduates completed six Consequences tasks—two human-written, two standard LLM-generated, and two genre-inspired LLM-generated, drawing on sci-fi and fantasy themes—along with measures of personality, cognitive ability, and creative behavior. We found that LLM-generated items elicited significantly higher fluency and flexibility. Genre-inspired items (fantasy and sci-fi) additionally elicited higher originality, enjoyment, and positive valence ratings; LLM-generated items were similar to human-written items. Genre-inspired items did not unfairly advantage participants with more exposure to genre content, demonstrating measurement invariance. Together, these findings suggest that LLMs can generate consequences items with psychometric properties comparable to those of human-written items. We discuss the implications for creativity assessment, item bank development, and best practices for AIG in creativity research, and we provide the prompts we used for item generation to enable researchers to generate new Consequences items at scale.
The Darbest dataset is a Universal Dependencies (UD) treebank dataset for Standard Sorani Kurdish written in the Perso-Arabic script. It contains 69,000 annotated sentences and 1,205,855 tokens collected from nine textual domains. The corpus was collected from seven Kurdish online news websites and supplemented with texts from published books. Before preprocessing, the collected corpus contained 1,250,275 words from 5,627 web pages together with book-based texts and was preprocessed using a Python-based pipeline involving text cleaning, Unicode and punctuation normalization, sentence segmentation, and tokenization. The dataset was developed to provide a large-scale syntactically and morphologically annotated resource for Standard Sorani Kurdish. A separate 100-sentence gold-standard set was manually annotated according to the Universal Dependencies v2 guidelines. Sorani Kurdish linguistic experts supported the selection of sentences representing diverse and linguistically complex structures and reviewed LLM-generated annotations for errors. The gold-standard set was used to construct few-shot prompts for annotating the remaining corpus. The resulting annotations were represented in the standard CoNLL-U format and validated using the official Universal Dependencies validation tool, followed by manual correction and quality review. The released treebank is divided into training, development, and test sets and includes lemmas, Universal Part-of-Speech (UPOS) tags, morphological features, syntactic heads, and dependency relations. The dataset can be used to train, evaluate, and benchmark NLP models for part-of-speech tagging, lemmatization, morphological analysis, dependency parsing, and related computational linguistics tasks. The accompanying repository also contains the separate 100-sentence gold-standard set, plain-text corpus splits, README documentation, and a dataset statistics spreadsheet.
PURPOSE: This study aimed to provide normative affective ratings for middle-aged to older adults with normal hearing using the Marcell database. The database includes 120 diverse environmental sounds representing various real-world acoustic events, such as those produced by animals, humans, musical instruments, tools, signals, and liquids. Secondary aims were to examine the relationship between valence and arousal, and whether factors such as age, gender, hearing-related lifestyle, hearing demands, and self-reported hearing difficulties influence these affective evaluations. STUDY SAMPLE: Ninety participants (mean age: 54 ± 9 years) with self-reported normal hearing. DATA COLLECTION AND ANALYSIS: The study was conducted online. Participants rated each sound token in the Marcell database in terms of valence (pleasantness) and arousal (intensity). They also completed a hearing-related lifestyle questionnaire assessing the frequency, importance, and difficulty of hearing in everyday situations. Normative data for valence and arousal were calculated as means and standard deviations across participants for each individual sound token. To explore the relationship between valence and arousal, correlation analyses were performed. Furthermore, mixed-effects linear regression was used to examine whether any of the measured variables could predict participants' affective responses to the sounds. RESULTS: Normative data for each individual sound token were obtained and reported. A moderately strong negative correlation was found between valence and arousal ratings, indicating that pleasant sounds were generally perceived as calming, while unpleasant sounds were seen as more arousing. Regression analyses revealed that age, gender, hearing-related lifestyle, hearing demands, and self-reported hearing difficulties did not significantly predict valence or arousal ratings in the current dataset of middle-aged to older adults. CONCLUSIONS: The normative data presented in this study offer a benchmark for the hearing research community when examining emotional responses across middle-aged to older adult populations.
Abstract This study examines associations between brain levels of amyloid-β and tau with representational pattern similarity in amygdalar reactivity to negative and neutral images in older adults without dementia. 81 participants viewed affective images during functional magnetic resonance imaging (fMRI). Participants rated the images on valence and arousal outside the scanner. Amyloid-β and tau were measured with 11 C-Pittsburgh compound B and 18 F-MK-6240 positron emission tomography (PET) imaging. Representational pattern similarity analyses compared amygdalar responses to negative and neutral stimuli, while preserving the voxel-wise pattern of fMRI activation. Greater differentiation in the pattern of responding across the amygdala was indicated by lower similarity between negative and neutral stimuli. Greater tau levels in the entorhinal cortex and amygdalae were associated with less pattern similarity in left amygdalar reactivity to negative and neutral images in participants whose tau level was below standard positivity thresholds. Less pattern similarity in right amygdalar reactivity to negative and neutral images was also associated with the participants’ valence and arousal ratings of the stimuli. No associations were found with global amyloid levels and only the association between entorhinal tau and amygdalar similarity remained significant after Bonferroni correction. Greater amygdalar pattern separation in response to negative and neutral stimuli with higher levels of entorhinal tau suggest greater amygdalar sensitivity to negative compared to neutral information as tau accumulates in the brain, suggesting a potential underlying mechanism for the disrupted emotional processes often observed in preclinical Alzheimer’s Disease and other Related Dementias.
The Darbest dataset is a Universal Dependencies (UD) treebank dataset for Standard Sorani Kurdish written in the Perso-Arabic script. It contains 69,000 annotated sentences and 1,205,855 tokens collected from nine textual domains. The corpus was collected from seven Kurdish online news websites and supplemented with texts from published books. Before preprocessing, the collected corpus contained 1,250,275 words from 5,627 web pages together with book-based texts and was preprocessed using a Python-based pipeline involving text cleaning, Unicode and punctuation normalization, sentence segmentation, and tokenization. The dataset was developed to provide a large-scale syntactically and morphologically annotated resource for Standard Sorani Kurdish. A separate 100-sentence gold-standard set was manually annotated according to the Universal Dependencies v2 guidelines. Sorani Kurdish linguistic experts supported the selection of sentences representing diverse and linguistically complex structures and reviewed LLM-generated annotations for errors. The gold-standard set was used to construct few-shot prompts for annotating the remaining corpus. The resulting annotations were represented in the standard CoNLL-U format and validated using the official Universal Dependencies validation tool, followed by manual correction and quality review. The released treebank is divided into training, development, and test sets and includes lemmas, Universal Part-of-Speech (UPOS) tags, morphological features, syntactic heads, and dependency relations. The dataset can be used to train, evaluate, and benchmark NLP models for part-of-speech tagging, lemmatization, morphological analysis, dependency parsing, and related computational linguistics tasks. The accompanying repository also contains the separate 100-sentence gold-standard set, plain-text corpus splits, README documentation, and a dataset statistics spreadsheet.
Emotional responses to sounds are often described along two dimensions: arousal (exciting–calming) and valence (pleasant–unpleasant). Whereas sound level is a robust cue for arousal, the acoustic cues supporting valence are less well defined, particularly across listeners with different hearing abilities. Valence ratings were obtained from 236 adults (with and without hearing loss) for 75 sounds from the International Affective Digitized Sounds corpus (IADS-2). These data were pooled from several studies. For each sound, acoustic characteristics reflecting spectral shape and its time-varying changes were extracted. For each listener, age, gender, pure-tone average hearing threshold (0.5–2 kHz), and anxiety/depression symptom scores were collected. Valence ratings for previously unseen listeners were predicted using an interpretable machine-learning model (gradient-boosted trees with a mixed-effects structure). Out-of-fold performance was moderate (R² = 0.458; Spearman ρ = 0.679). The strongest predictors included variability in spectral “peakiness” (spectral crest), mid-frequency spectral contrast near 1–2 kHz (mean and variability), and loudness variability. Hearing loss also contributed, suggesting audibility and/or altered sound representation affects pleasantness judgments. Overall, listeners appear to rely on time-varying spectral structure, particularly in mid frequencies, when forming pleasant–unpleasant judgments, providing candidate acoustic cues for affective sound processing in listeners with and without hearing loss.
Form-to-lemma lookup tables for Greek (568,918 corpus-attested forms with counts, joined to Liddell-Scott-Jones) and Latin (1.4M paradigm-generated forms joined to Lewis & Short), built from and for the Source Library corpus of early modern books. Validated against the Universal Dependencies gold treebanks: Greek 88.8% coverage / 96.6% lemma agreement (Perseus, 132,156 tokens); Latin 91.4% / 95.1% (Index Thomisticus, 333,281 tokens). See README.md for schema, normalization, method, and attribution. Built on the Perseus Project's Morpheus analyzer (Perseids distribution, MPL-2.0) and the lsj9 digitization of LSJ (CC BY 4.0).
Language change is continuous and vital for the sociocultural adaptation and status development of language varieties. Perhaps the most important linguistic feature of Nigerian English language change is semantic extension which refers to the process whereby terms already existing in English (standard English) acquire new meanings. This paper investigated the usage of semantic extension in Nigerian English (NIE), the social motivations responsible for such semantic innovations and the relevance of linguistic innovations for the description and teaching of English in Nigeria. Using descriptive qualitative research design, data for the research were collected using naturally occurring language forms while other information sources included print and social media, educational, political and religious institutions. It was found that semantic extension in Nigerian English is rule-governed, based on culture, multilingualism, technological innovations, socio-economic experiences and indigenous conceptual framework. The data further revealed that the semantic extensions are sociolinguistically recognized by educated users of Nigerian English and linguistic norm and not mere linguistic deviation from Standard English. It was observed that semantic extension promotes lexical reduction, language creativity and the institutionalization of Nigerian English as a variety of the World Englishes paradigm. It was recommended that Nigerian English new lexical items should be incorporated into English textbooks and dictionaries, Nigerian English lexical innovations should form the basis for the teaching of English in Nigeria.
Front mid-vowel phonemes /e/ and /ɛ/ are contrastive in Gallego, while Spanish contains the phoneme /e/ and English contains the phoneme /ɛ/. This study investigates how Language Mode, Language Dominance, and L3 proficiency in Spanish-Gallego early bilinguals affect front mid-vowel production in L3 English. Language Mode (Grosjean 1985) represents the real-time activation of a speaker’s languages and the associated processing frameworks, heavily influenced by environmental and psychosocial factors. We manipulate participants' Language Modes using story-retelling and sentence-reading tasks across their languages. The resulting Euclidean distances between vowel tokens will be measured and analyzed against Bilingual Language Profile (Birdsong, Gerken, and Amengual 2012) Language Dominance ratings and self-reported English proficiency. We expect to find that participants with higher proficiency in English will have greater distinction between /e/ and /ɛ/ phonemes across all three languages, that stronger Language Dominance of Gallego over Spanish will correlate with greater distinction between these phonemes, and that activation of Gallego or English Language Mode will also correlate with a greater distinction on average for speakers. We anticipate that participants with low self-reported English proficiency will see greater distinction between these phonemes in the Gallego Language Mode conditions, but not within English Language Mode, following Escudero's (2005) Second Language Linguistic Perception model and it's extension to phonological production as proposed by Casillas and Simonet (2018).
This record provides disclosure-controlled derived analytical data, analysis code, archived model outputs, and figures supporting the study “Beyond street imagery: A multi-source, multilevel study of spatial conditions and in situ emotional valence in public spaces.” The package reproduces the reported analyses from 378 valid in-situ valence ratings collected across 40 route sections within 10 purposively selected sites. Direct identifiers, exact site linkage, the raw workbook, source route photographs, and manual annotation materials are not included because of privacy, consent, and image/location-rights considerations.
Abstract This chapter analyzes the deliberate modification of English language accents as a tactic in professional career development. This chapter examines how individuals modify their speech patterns to conform to established linguistic norms in professional environments. This chapter further explains these by utilizing linguistic theories and professional communication principles. This chapter incorporates language profiling, cultural impacts, and geographical impacts of accent modification on career development. This chapter highlights the dual influence of accent modification on individual identity and career prospects.
Emotion perception in speech supports speech comprehension and depends on listeners’ integration of acoustic and lexico-semantic cues. The computational principles underlying this integration remain unclear. Here, we model the perceived speaker emotion as an adaptive internal state that updates continuously during speech perception, revealing temporally evolving weights on acoustic and lexico-semantic features. We presented 90 natural conversational speech excerpts (30–90 s each) from the Buckeye Corpus (Pitt et al. 2007) to 173 listeners, who rated speaker emotional valence continuously over time from most positive to most negative. Participants also rated excerpts without lexico-semantic cues (via low-pass filtering) and without acoustic cues (via reading time-locked transcripts). We find that time-resolved emotional valence ratings for some excerpts are more affected by removing lexico-semantic cues, and others are more affected by removing acoustic cues, implying that listeners flexibly utilize different features to infer speaker emotion at any given moment. We then explicitly map the temporal relationship between acoustic/lexico-semantic features and emotional valence ratings by fitting time-lagged random forest regressors across a range of lag values, identifying the relative weighting between features that drive perceived speaker emotion and resolving the timing of their integration during speech perception.
Nick Joaquin (1917-2004), one of the most celebrated Filipino writers in English, masterfully appropriates the language of the colonizer to create a distinctive style of English known as Joaquinesque. His works challenge Western linguistic norms by integrating indigenous and Hispanic influences into English, thereby resisting the erasure of Filipino identity. Through his creative manipulation of English, Joaquin engages in what Joy Harjo describes as "reinventing the enemy’s language" (Harjo et al., 1998, p. 22), transforming it into a medium of cultural assertion and artistic defiance.
This is the Adult–Child Touching Hands Dataset (AC-THD): a dataset of 59 image stimuli depicting hand-to-hand touch between an adult and a child, systematically categorized into three emotional valence classes: negative (n = 15), neutral (n = 29), and positive (n = 15). The tool emphasizes the central role of the hands in conveying emotional information within an adult–childinterpersonal context.The images and the hand-to-hand positions were developed by four trained psychologists, and were first assigned into three emotional valence categories: negative (e.g., one person tightly pinching or forcefully gripping the other person’s hand); neutral (e.g., the two individuals’ fingers lightly overlapping or their wrists touching), and positive (e.g., one person gently stroking the back of the other person’s hand or the two individuals interlocking their hands). Images were initially assigned provisional alphabetical labels, which were subsequently refined through a percentile-based procedure in accordance with validation data. The AC-THD acquisition phase involved a mother (Caucasian, aged 48, right-handed) and her child (Caucasian male, aged 10, right-handed). Images were acquired by a professional photographer using a Canon EOS 1100D digital SLR camera with a 12.2 megapixel APS-C CMOS sensor and a resolution of 4272 × 2848 pixels, equipped with an EF-S 18–55 mm f/3.5–5.6 lens. The built-in flash (Guide Number 9.2 at ISO 100) was manually triggered for each shot to standardize illumination and minimize shadows. The camera was positioned on a tripod to maintain a fixed 90° angle. All images were captured in color in sRGB mode, with framing restricted to the hands and forearms. In the validation study, images were rated by 321 participants on valence, using a numeric rating scale ranging from 0 to 10, where 0 indicates a strongly negative image, 5 a neutral image, and 10 a strongly positive image, with anchors at 0 (“strongly negative”) and 10 (“strongly positive”). Mean emotional valence ratings were calculated for each image, along with standard deviations (SDs), mode, minimum and maximum values, and percentiles. Stimuli were finally classified as negative, neutral, or positive based on the 25th and 75th percentiles of the distribution of mean valence ratings. Accordingly, images with mean valence scores ≤ 4.25 were classified as negative, those with mean valence scores > 4.25 and < 7.23 as neutral, and those with mean valence scores ≥ 7.23 as positive. Therefore, the final dataset is comprised of: N = 15 images in the negative category; N = 29 images in the neutral category; N = 15 images in the positive category. Study validation of the dataset also examined potential differences in emotional valence ratings as a function of participants’ socioeconomic status (SES) and gender, which are reported as Supplementary Materials. These images can be used across a range of research fields, including emotion elicitation, neuroscience, and psychophysiological studies, as well as for assessing affective responses in individuals with histories of supportive or adverse tactile caregiving, given their potential to evoke autobiographical memories.
Visually evoked affective states serve as a representative method for investigating human emotion, which is recognized as a foundational substrate for the coordination between behavioral and physiological systems. Prior research has endeavored to establish correlations between subjective affective ratings of visual stimuli and corresponding hemodynamic responses in the human brain, thereby exploring the brain's responses to affective scenes. Existing findings indicate the presence of distributed emotional representations throughout various regions of the cerebral cortex and cortical affective maps that connect categorical and dimensional affective spaces. Nevertheless, these studies are constrained by (1) the exclusion of subcortical and cerebellar regions from analyses, (2) an insufficient understanding of how different brain regions are interconnected in representing diverse emotions, and a lack of explainability methodologies to validate brain decoding outcomes. In the present study, we have devised an affective decoding framework. An atlas encompassing all cortical, subcortical, and cerebellar regions was utilized to incorporate the entire brain into the analysis. By employing a linear decoding approach, we decomposed participants' complex emotional representations into 34 discrete categories and examined activation patterns across brain regions for each category. Ultimately, leveraging generative AI models, we conducted an additional analysis to elucidate the decoding results and thereby enhance interpretability.
Elementary students develop sensitivity to English sound-spelling patterns through instruction and practice. However, it is unclear whether this extends to subtle patterns and whether sensitivity varies by age or reading skill. One such pattern involves the pronunciation of a single-letter vowel (V) in the first syllable of bisyllabic words. A vowel letter is more likely to have a long (tense) sound when followed by a single consonant letter (C) and vowel letter (VCV, e.g., labor) than a short (lax) one (Kearns, 2020). However, short-vowel pronunciations (e.g., habit) are common). This study examines whether early readers exhibit sensitivity to this long-vowel tendency and whether this varies by grade or reading skill. Grade 2 (n = 18), Grade 3 (n = 28), and Grade 4 (n = 29) students with and without reading difficulties (RD) read 48 five-letter bisyllabic CVCVC pseudowords (e.g., losal). First-vowel pronunciations were recorded. Long-vowel pronunciation frequencies were compared across grades and with long-vowel frequencies in a lexical database (Zeno et al., 1995). Students used long vowel pronunciations less frequently than expected based on the frequency of polysyllabic words in the database, pronouncing them at a rate closer to that found in monosyllabic words. Grade 3 and 4 students used long vowel pronunciations more often than younger students and those with RD. Long vowel usage was notably higher for the letter O than other vowels, consistent with the more frequent occurrence of long pronunciations for that letter. Students show limited sensitivity to the subtle VCV long-vowel pattern, indicating that while acquiring this knowledge, it is not quickly consolidated, possibly due to variability in the pattern or differing learning rates.
A single-session fMRI study in which participants breathed with a visual pacer whose rate increased, decreased, or held constant on each trial (with abrupt vs gradual change salience), interleaved with free-breathing periods, and reported the perceived direction of change with confidence and arousal ratings. Respiration was recorded with a chest belt and expired CO2 with a capnograph. This registration covers the functional imaging analysis only: all behavioural and physiological processing is complete and disclosed; no functional image statistic has been computed. Hypotheses include direct replications of Farb, Zuo & Price (2023, eNeuro): insula/opercular deactivation under exogenous respiratory pacing, and MAIA-moderated sparing of anterior cingulate activity.
This research aims to study the stylistic choice of words and trace their semantic and argumentative movement by shedding light on the precision of the stylistic choice of Quranic words, which carry argumentative potential and persuasive semantic dimensions through their context. The recipient has no choice but to submit and acquiesce. The study examines these words in terms of their lexical and semantic content, not as a departure from the linguistic norm. The study examines examples from the Holy Quran, where the word is evaluative, lively, and charged with semantic dimensions and persuasive argumentative potential. These words compel the recipient to submit and acquiesce to what is presented to them. They possess semantic features that connect them to the context in which they appear, directing them in an influential argumentative direction.
The thesis investigates phraseological transformation, contamination and occasional paradigm expansion as sources of semantic ambivalence. The analysis is based on English, Russian and Uzbek examples in which a stable expression is reinterpreted through context. The study shows that phraseological ambivalence appears when an idiom is understood both as a fixed expression and as a free combination of words. Contamination creates a hybrid form in which two phraseological prototypes are simultaneously recognizable. Occasional paradigm expansion occurs when a text temporarily includes a foreign unit into an existing semantic series on the basis of phonetic, graphic or morphological similarity. These mechanisms reveal the semantic productivity of the tension between linguistic norm and contextual innovation.
Introduction: the Ludic supradialect of the Karelian language remains the least documented among the Karelian supradialects. The present article describes the verbal inflectional system of the South Ludic dialect: the types of lexical stems are classified, the verb paradigm is derived, and the morphological regularities that distinguish this dialect from the other supradialects of Karelian are identified. The regularities thus established are formalized as a system of rules underlying a module for the automatic generation of verb word forms in the VepKar corpus. Objective: to describe and systematize the verbal inflectional system of the South Ludic dialect of the Ludic supradialect of the Karelian language, and to formalize the identified regularities as rules for the automatic generation of verb word forms. Research materials: verb lemmas and word forms from the Open Corpus of Veps and Karelian Languages (VepKar), together with field materials collected by the author. Results and novelty of the research: the absence of a systematic description of the verbal morphology of the South Ludic dialect has hitherto impeded both its linguistic study and the corpus annotation of Ludic texts in VepKar. The research comprised: (1) the derivation of a complete inflectional paradigm of verbs of the South Ludic dialect and the identification of its distinctive features in comparison with the Livvi and Proper Karelian supradialects; (2) the construction of a system of lexical stems, auxiliary stems, and pseudo-stems for verbal inflection; (3) the development of word-form generation rules covering all identified inflectional types, illustrated in tables with paradigm exemplars; (4) the implementation, on the basis of these rules, of a word-form generation module in the Dictorpus corpus manager, which enabled the addition of over 100,000 Ludic verb word forms to the VepKar dictionary and raised the annotation coverage of the subcorpus from 60 % to 75%. A quantitative analysis of the distribution of verbal stem types in the VepKar dictionary and corpus confirmed the representativeness of the lexical database and the predominance of single-stem verbs (approximately 77 % of lemmas; 83 % of tokens). Scientific novelty: the first comprehensive system of rules for the automatic generation of verb word forms for the South Ludic dialect of the Karelian language.
Respect plays a crucial role in successful interpersonal and intercultural communication. However, differences in linguistic norms and pragmatic conventions often lead to misunderstandings between speakers of different languages. This article examines pragmatic failures in expressing respect in English-Russian cross-cultural communication. The study aims to identify linguistic and cultural factors that cause misinterpretations of respect and to analyze how respect is pragmatically encoded in both languages. The findings suggest that pragmatic failures frequently arise from divergent politeness strategies, speech act realizations, and sociocultural expectations embedded in English and Russian communicative practices. The study emphasizes the importance of pragmatic awareness in developing intercultural communicative competence.
Digital communication has become a key site of sociolinguistic variation, identity performance, ideological contestation, and inequality, while AI now actively shapes linguistic norms. This chapter develops an integrated account of platformed language, media discourse, and AI-mediated communication as interdependent domains. It examines how (1) platform affordances reorganise interactional norms and community formation; (2) media logics shape representations of language and identity; and (3) algorithmic systems function as governance infrastructures that regulate linguistic visibility and legitimacy. Drawing on theories of variation, identity, language ideology, and communities of practice, the chapter proposes the Platformed Sociolinguistic Ecology (PSE), structured across interactional, representational, and infrastructural layers. It introduces algorithmic linguistic authority and algorithmic normativity as key constructs, and adopts a conceptual methodology based on critical synthesis of interdisciplinary scholarship.
This is a retrained Slovenian model for the Trankit v1.1.1 library for multilingual natural language processing (https://pypi.org/project/trankit/), trained on the concatenation of the SSJ UD treebank of written Slovenian (featuring fiction, non-fiction, periodicals and Wikipedia texts) and the SST UD treebank of spoken Slovenian (featuring transcriptions of spontaneous speech in various settings). It is able to predict sentence segmentation, tokenization, lemmatization, language-specific morphological annotation (MULTEXT-East morphosyntactic tags), as well as universal part-of-speech tagging, morphological features, and dependency parses in accordance with the Universal Dependencies annotation scheme (https://universaldependencies.org/). In comparison to its counterpart models trained on SSJ (http://hdl.handle.net/11356/1963) or SST datasets only, this model yields a significantly better performance on spoken transcripts and an identical state-of-the-art performance on written texts. The model can therefore be recommended as the default, 'universal' Trankit model for processing Slovenian, regardless of the data type. To utilize this model, please follow the instructions provided in our github repository (https://github.com/clarinsi/trankit-train) or refer to the Trankit documentation (https://trankit.readthedocs.io/en/latest/training.html#loading). This ZIP file contains models for both xlm-roberta-large (which delivers better performance but requires more hardware resources) and xlm-roberta-base. Version 1.3 was trained on the same data as version 1.2, except that spoken SST data (UD v2.15) was augmented by colloquial (non-standardized) transcriptions of spoken Slovenian alongside the standardized ones. The resulting model achieves state-of-the-art performance on both standardized (e.g. "včasih govorimo takole") and colloquial speech transcriptions (e.g. "včas govorimo tkole"), without affecting the performance on written data.
Closely related languages written in different scripts expose little surface overlap to multilingual models, limiting cross-lingual transfer. We compare two approaches to script unification: the general-purpose uroman romanizer and the family-specific Common Turkic Script (CTS). We train matched fastText models on transliterated Wikipedia corpora from 11 Turkic languages and evaluate them on WikiANN named entity recognition and Universal Dependencies part-of-speech tagging. CTS and uroman show no significant difference on NER, while both substantially outperform the official monolingual fastText baselines. POS results reveal no universal winner: language-specific differences are associated with the cross-lingual character n-gram coverage induced by each representation, while within-language coverage becomes more important when target-language supervision is available. Although CANINE-c achieves higher overall POS averages, the substantially simpler fastText-based systems remain competitive on several treebanks. Overall, the effectiveness of script unification depends on the language, the induced subword overlap, and the available supervision.
The intricate relationship between social identity and linguistic variation presents a persistent challenge in sociolinguistics, particularly regarding how speech communities evolve over time. Central to this challenge is the conceptual distinction between age-grading, where individuals systematically alter their linguistic habits across their lifespan, and actual language change in progress, which reshapes the fundamental linguistic norms of an entire community. This paper investigates how slang and specialized linguistic markers serve as crucial indicators of identity, facilitating the intergenerational transmission of cultural and social norms. By bridging established sociolinguistic theories with contemporary computational models of cultural evolution and network dynamics, we propose a comprehensive analytical framework. This framework aims to rigorously disambiguate cyclical lifespan variations from permanent evolutionary shifts in a population's linguistic repertoire. Ultimately, this research provides a methodological foundation for understanding how overlapping generations negotiate identity through the transmission, adoption, and eventual abandonment of localized linguistic markers.
This article examines how a historically marginalized Chinese Indonesian community negotiates belonging in post-Suharto Indonesia, drawing on ethnography with the Indonesian Jinjiang Hometown Association. It argues that the association functions as an agentive institutional site rather than a passive vessel of cultural preservation, simultaneously curating collective memory and regulating members’ identity performances. As ‘memory curator’, it manages historical trauma through heroization, securitization and legitimization; as ‘identity regulator’, it orchestrates national, ethnic and class performances through mechanisms including donation-based status hierarchies and linguistic norms. These are not two separate functions but interlocking stages of a single institutional cycle, here named ‘Connection-Translation-Empowerment’, in which memory work authorizes status claims and the resulting capacity sustains further rounds of memory work. Fieldwork evidence complicates any celebratory account of this cycle: the empowerment it generates is unevenly distributed across gender, generation and ethno-cultural lines, revealing the limits of the belonging the association claims to produce.
The increasing use of artificial intelligence (AI) in hiring raises significant ethical concerns, particularly regarding language-based biases. Although AI systems are promoted as tools to reduce human prejudice, research shows they can reproduce and even intensify existing inequalities. This qualitative study, based on secondary data, explores how resume language influences AI hiring decisions, focusing on linguistic patterns and the ethical implications of automated screening. Drawing on case studies and prior literature, it examines how algorithms interpret tone, action verbs, and culturally coded expressions. Findings indicate that AI tools often favor resumes aligned with dominant linguistic norms, such as industry jargon and assertive or masculine-coded language, putting underrepresented groups and non-native speakers at a disadvantage. The study also highlights how keyword filtering and biased training data reinforce systemic inequalities. Overall, it emphasizes the need for transparent, inclusive, and linguistically aware AI systems, calling for interdisciplinary efforts to ensure fair and equitable hiring practices.
This study examines the phenomenon of stylistic deviation (déviation stylistique) in the celebrated Dāliyya poem of the Moroccan scholar and Sufi mystic Abū ʿAlī al-Yūsī (d. 1102 AH/1691 CE). Drawing on Jean Cohen’s theory of poetic language, it explores how al-Yūsī departs from conventional linguistic norms to generate multiple layers of aesthetic, semantic, and spiritual meaning. The study adopts an analytical approach encompassing the phonological, morphological, syntactic, semantic, and pragmatic levels of the text. It demonstrates that stylistic deviation in al-Yūsī’s poetry is not merely a formal rhetorical device but also a cognitive and mystical act that embodies the Sufi vision of transcendence, unity, and spiritual transformation. The findings reveal that al-Yūsī’s Dāliyya represents a distinctive model of Moroccan Sufi poetry, in which language transcends its representational function to become a medium of revelation, illumination, and spiritual insight. Consequently, stylistic deviation emerges as a dual process of linguistic innovation and spiritual liberation, reflecting the profound interconnection between poetic expression and mystical experience.
Le French Treebank: une ressource lexicale et syntaxique richement annotée (et validée manuellement) pour les linguistes, utilisable en TAL, dans sa version 2.0 Projet initié en 1997, avec le soutien de l'IUF, du CNRS et du CNRTL21 550 phrases (environ 664 500 tokens) du journal Le Monde (1990-1993)Métadonnées: auteur, date, domaine (par article)Annotations lexicales (catégories, sous-catégories, flexion, mots composés avec composants) et syntaxiques (constituants majeurs, fonctions grammaticales) validéesPlusieurs formats disponibles: XML, PTB, CoNLL, codage UTF-8 (ligature œ notée oe)Nouveautés de la version 2.0: plusieurs erreurs d'annotation ont été corrigées depuis la publication en 2016 de la version 1.0
Language is a living organism that evolves alongside technological and social advancements. This paper examines the phenomenon of neologisms—newly coined words or expressions—and their pervasive role in contemporary English mass media. The study categorizes recent neologisms based on their morphological formation processes, such as blending, compounding, and functional shift. Furthermore, it analyzes how mass media acts as a primary catalyst for the popularization of these terms. By investigating digital journals, social media platforms, and news broadcasts, the research highlights the pragmatic functions of neologisms in creating concise, engaging, and culturally relevant communication. The findings provide insights into the current trends of English lexicology and the impact of the digital age on linguistic norms.
Recent advances in Vision Language Models (VLMs) have prompted growing interest in their use as scalable proxies for human affective evaluation of urban and architectural environments. However, existing studies largely rely on non-immersive 2D imagery, inconsistently defined affective dimensions, and environmental appraisal measures rather than subjective emotional responses. Consequently, the extent to which VLMs reproduce human affect under ecologically valid immersive conditions remains unclear.,This study evaluates VLMs as affective proxies within immersive Virtual Reality (VR) environments. Human participants assessed a curated set of urban scenes using the Positive and Negative Affect Schedule (PANAS), while multiple state-of-the-art VLMs generated corresponding affective ratings for the same stimuli.,Across models, VLMs achieved moderate-to-high agreement with human judgments on aggregate affective ratings and preserved substantial similarity in the representational structure of affective responses (RSA). However, this alignment was not uniform: Bland–Altman and item-level analyses revealed systematic calibration asymmetries, characterized by compression of negative affect intensity and localized errors in high-salience descriptors related to threat and hostility. Together, these findings indicate that VLMs reproduce the relational geometry of human affective experience while diverging in the scaling of affect intensity.,These results suggest that contemporary VLMs are well-suited for comparative and exploratory assessment of architectural environments, but remain limited for precise estimation of absolute emotional intensity, particularly for negative affect. By integrating immersive VR experimentation with psychometrically grounded evaluation and multi-level model analysis, this work advances the methodological foundations for AI-assisted affective assessment in architecture and urban design.
Abstract Recognizing others’ emotions is central to social interaction. Traditional biological psychology infers emotional responding via laboratory measures, whereas contemporary computer vision algorithms claim to identify emotions unobtrusively from facial video. However, the validity of such algorithms for classifying spontaneous emotional responses occurring without explicit communicative intent remains debated. We compared established psychophysiological measures (EEG, facial EMG, EDA activity) with the open-source facial behavior toolkit OpenFace for classifying participants’ spontaneous responses during free viewing of happiness-inducing, disgust-inducing, and neutral pictures. Participants provided valence and arousal ratings and later selected the basic emotion that best matched their reaction which served as the classification criterion. Using within-participants single-trial support vector machine (SVM) classification, EEG achieved the highest accuracy (40%), followed by facial EMG (37%); OpenFace reached 36%. All methods except EDA exceeded chance performance (33.3%) and were lower compared to human raters (48%). Predictions declined slightly for across-participants SVMs, being at chance for OpenFace and EDA. The results indicate that in principle both, psychophysiological measures and video-derived facial action units, can capture diagnostically relevant aspects of emotional responding during picture viewing, but that their performance is limited when expressions are spontaneous and not produced for communicative purposes. Inter-individual variability in expressivity and physiological responding likely contributes to these limitations and should be considered when deploying automatic emotion recognition in research or applied settings.
There is a consensus among researchers and educators that teachers’ language register (LR) influences learners’ cognitive development. While previous studies have examined how teachers’ professionalism in relation to LR is evaluated by educational experts and researchers, this study was motivated by the need to understand the perceptions of parents, teachers, and teacher-trainees regarding the contribution of LR to teachers’ professionalism in intercultural environments. In this mixed-methods study, 308 participants were asked to rate teacher professionalism based on one of three written reflections, identical in content but differing in LR level. The study was conducted in Israel, a multilingual society in which Hebrew is the official language of instruction but is a second language for many teachers and learners. This context, shaped by ongoing immigration, adds an important cultural lens to the investigation. Unlike previous studies, the findings in this research showed that very few participants considered LR an important component of teacher professionalism. Instead, they emphasised proficiency in other areas, particularly teacher knowledge, cultural awareness, and responsiveness to learners’ needs. This study offers global implications for teacher education in increasingly multicultural and multilingual societies, suggesting that perceptions of professionalism prioritise cultural competence and pedagogical adaptability over formal linguistic norms.
Considering the trends towards universalization and differentiation in procedural law, the paper hypothesises that similar patterns exist in the language of procedural law. It identifies the need to study the linguistic and textological distinctiveness of the texts of the Criminal Procedure Code, Civil Procedure Code, Arbitration Procedure Code, and Administrative Procedure Code. The study reveals certain features of the language of these laws in terms of semantics, grammar, vocabulary, syntax, and textology. Examples of homonymy, synonymy, oxymoron, violations of linguistic norms in terms of spelling and correct word order, the use of identical headings, combining heterogeneous elements into structural elements of a normative legal act, and lexical patterns are given. The normative meaning of headings is revealed; examples reflecting the problems of the structure of procedural laws are given. The author names the striking textual feature of the Criminal Procedure Code of the Russian Federation, namely Art. 5, which includes the definition of the main terms used in the law. The syntactic complication of the language of procedural laws has been noted in recent years. The lexical features of individual procedural laws are revealed, demonstrating their substantive originality. An assumption is made about the influence of the quality and features of the normative legal language on the language of judicial acts and the language of legal science.
Abstract This study investigates the multifunctionality of the Turkish discourse connective ve ‘and’ in simultaneous interpreting from Turkish into Turkish Sign Language (TİD) in broadcast news. Using a corpus of interpreted news texts, it examines how this polyfunctional connective is transferred across modalities. Drawing on Halliday and Hasan’s cohesion model and the Penn Discourse Treebank typology, the study identifies discourse relations expressed by ve — such as elaboration, temporality, causality, and comparison — and analyzes how they are conveyed in TİD through explicit signs, implicit marking, or omission. Most relations are preserved, though often implicitly, reflecting the economy principle in signed languages. Manual signs used to express ve, including ve -1 and palm-up, vary systematically with relation type. This points to a cross-modal tendency for explicit realization under increased processing demands. As the data derive from hearing interpreters rather than native Deaf signers, the findings reflect interpreter-mediated strategies rather than monolingual Deaf discourse norms.
Static concreteness ratings are widely used in NLP, yet a word's concreteness can shift with context, especially in figurative language such as metaphor, where common concrete nouns can take abstract interpretations. While such shifts are evident from context, it remains unclear how LLMs understand concreteness internally. We conduct a layer-wise and geometric analysis of LLM hidden representations across four model families, examining how models distinguish literal vs figurative uses of the same noun and how concreteness is organized in representation space. We find that LLMs separate literal and figurative usage in early layers, and that mid-to-late layers compress concreteness into a one-dimensional direction that is consistent across models. Finally, we show that this geometric structure is practically useful: a single concreteness direction supports efficient figurative-language classification and enables training-free steering of generation toward more literal or more figurative rewrites.
<p>For Arabic-English bilingual students, writing is a particularly challenging task, as it requires learning how to produce meaning through written scripts, distinctively different from those of the mother tongue. In this study, subjective familiarity ratings and vocabulary knowledge scores of the printed words of the Peabody Picture Vocabulary Test (PPVT) were collected from native Arabic speakers whose English proficiency was at least at the competent-user level (as per International English Language Testing System [IELTS] criteria). This level of competency is generally the precondition for admission of second-language learners to English-medium universities. The aim of the study was two-fold: (a) to determine the information upon which participants’ vocabulary knowledge relies through an examination of the extent to which such knowledge is predicted by key subjective and objective word properties; and (b) to assess the degree to which participants’ vocabulary knowledge, estimated from printed word comprehension (WC), is related to second-language students’ writing performance, as well as writing anxiety. In this study, objective word properties (e.g., frequency counts and the number of semantic neighbors), as well as subjective familiarity ratings of printed words, notably contributed to vocabulary knowledge (as indexed by WC scores). Furthermore, participants’ vocabulary knowledge was related to writing performance, as well as writing anxiety. Thus, the printed words of the PPVT could be used to predict not only the vocabulary knowledge of Arabic-English speakers admitted to university-level courses but also writing difficulties, thereby informing selective preemptive interventions.</p>