Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
16504 papers
Cette thèse vise à présenter une méthode complète permettant d'extraire des règles grammaticales quantitatives et interprétables à partir de treebanks syntaxiques. Cet objectif s'inscrit dans le cadre des grammaires descriptives, qui nécessitent des analyses fines pour saisir des phénomènes linguistiques complexes tout en intégrant les propriétés fondamentales du langage. Pour y parvenir, nous proposons une formalisation générale des règles grammaticales issues de corpus, facile à généraliser et à mettre en œuvre à l'aide de méthodes automatisées. Les règles extraites sont concises, ont différents niveaux de granularité, permettent une sélection flexible et sont ordonnées par importance. Ces descriptions formelles sont extraites de corpus à l'aide de modèles de régression logistique parcimonieux, qui favorisent l'interprétabilité tout en produisant un ensemble concis de règles présentant les propriétés souhaitées. Les résultats de notre méthode sont évalués dans plusieurs langues afin d'examiner sa capacité à répondre aux besoins descriptifs et comparés à d'autres approches existantes. La méthodologie est ensuite étendue à la description contrastive des langues, mettant en évidence les différences et les similitudes entre les langues. Cette approche permet d'obtenir des signatures linguistiques, des ensembles de modèles communs et distinctifs qui profilent chaque langue. Les expériences portent sur plusieurs paires de langues, familles de langues et genres textuels. Une attention particulière est accordée à la nature des règles extraites et au rôle des connaissances linguistiques théoriques dans le processus d'extraction. Cette thèse vise en fin de compte à démontrer la faisabilité de l'extraction d'une grammaire guidée par le corpus pour la description des langues. Ainsi, elle étend les liens entre la linguistique descriptive, formelle et computationnelle dans le contexte plus large de la description des langues.
Arguments, unlike adjuncts, are typically understood as verb-specific dependents, which includes the fact that the morphosyntactic devices used for argument encoding are determined by individual verbs. Building on this observation, we operationalize arguments as dependents whose encoding device occurs with a given verb at a significantly higher-than-average frequency. We apply an argument extraction algorithm to a dataset of 132,221 verb dependents from Russian treebanks available in the Universal Dependencies (UD) platform. To evaluate the algorithm ’ s performance, we compare its results to a manually annotated subset, informed by The Active Dictionary and a detailed semantic understanding of argumenthood. The frequency-based algorithm achieves acceptable precision (approx. 0.83), with particularly few false positives, making it a promising tool for cross-linguistic applications in typologically diverse languages with UD treebanks. Theoretically, we argue that a quantitative distributional approach to valency—originally proposed in Ju. D. Apresjan ’ s early pioneering work—broadly aligns with the in-depth semantic analyses of individual verbs and their meanings found in his later works, including The Active Dictionary.
The article discusses the design and application of dependency treebanks for Biblical Hebrew, focusing on their potential for linguistic research. It highlights the importance of such treebanks in studying historical languages, where native speaker intuition is unavailable. The study compares dependency grammar frameworks, such as Universal Dependencies and Prague Dependencies, examining their suitability for different research goals, including syntax-semantics interface, word order analysis, and phonological-syntactic relationships. Specific criticisms of Universal Dependencies, particularly its hybrid nature prioritising semantic over surface-syntactic relations, are addressed alongside alternatives like the multilayered Prague Dependencies. The article emphasises the need for research-driven design, recommending adaptations based on the intended linguistic applications and underlying theoretical assumptions.
Researchers often assess processes underlying human perception by measuring participants’ judgements of image stimuli. However, traditional methods for quantifying subjective judgements, such as Likert scales, sliding scales, and pairwise comparisons, are vulnerable to biases or demand extensive time and resources from researchers and participants. The present study compared the efficiency, reliability, and validity of these established methods against the Fast Image Rating Experiment (FIRE), our force-choice-based paradigm for assessing perceptions of visual stimuli. When used to rate image preference and naturalness, the FIRE was five times faster than established methods, highly reliable, and valid. FIRE achieved high reliability in less than half the time required to reach equivalent reliability with the Likert or sliding scale, which could save researchers thousands of dollars. The scalability and cost-effectiveness of the FIRE make it a valuable resource for supporting large-scale behavioral science.
International audience
The future of healthcare delivery across the cancer continuum holds great promise and challenge. U.S. cancer mortality across all cancers combined decreased ~2% annually from 2015 to 2019 thanks to a range of factors including clinical and delivery innovations in cancer prevention, control, treatment, supportive care, and efforts to improve clinical trial access [1]. However, long-standing cancer health disparities remain. Accelerated progress is vital to reduce cancer deaths for all Americans and achieve National Cancer Plan goals [2-4]. Additionally, COVID-19 impacts on cancer are still emerging [5, 6] and the long-term cancer survivor population is growing—increasing sustained surveillance for recurrence, new cancers, late effects, and other long-term health concerns [7]. These and other individual, institutional, and societal trends are shaping cancer care delivery, treatment, and research. Cancer health services research has a role in tracking and understanding how such trends influence health service design, delivery, and outcomes, to inform care approaches, health system decisions, and policy innovation across the cancer continuum. Such trends also beg the question, what measurement and methodological innovations are needed for timely, valid evaluation of their impact on cancer-related health services and outcomes important to patients, caregivers, healthcare professionals, payers, and policy makers? Continued measurement and methodological development only stand to improve scientific quality, reproducibility, and practical impact. Therefore, in this commentary, we briefly summarize 10 trends in cancer care delivery, treatment, and research and explore potential implications for health services research measurement and methods. Our intent is not to comprehensively address all possible opportunities or ideas presented here, but to highlight pressing, foundational needs and promising directions for measurement and methods focused science. Our focus is cancer health services research. However, the challenges and opportunities discussed clearly have broader implications. Rededication to strengthening our research methods and measures is an investment in the foundational T0 basic science [8] of cancer health services research and, therefore, essential to generating and translating future evidence into practice and policy. A universe of trends influences health services at-large at any given moment, however, we highlight 10 trends elevating the necessity of methods and measures research in cancer-focused health services research, including: (1) precision oncology; (2) whole-person perspectives; (3) health technologies (e.g., artificial intelligence (AI), mobile health, telehealth); (4) expanding in-home and community-based services; (5) health system integration and efforts addressing care fragmentation; (6) workforce capacity and evolving roles; (7) population aging; (8) improving safety, quality, value and access while controlling costs and addressing financial toxicity; (9) addressing social drivers of health; and (10) leveraging data oceans with unstructured, semi-structured, and structured elements. Deep discussion of each is beyond our scope, but we discuss several of these trends with examples and then give focused attention to measurement and methodological implications. One trend with significant measurement and methodological implications is the rapid advancement of precision oncology paradigms. Precision paradigms are fundamentally changing the understanding of cancer risk, diagnosis, disease profiling, treatment monitoring, therapeutic development, trial eligibility, and have birthed new health services (e.g., genetic counseling, in-house molecular pathology) [9-11]. Precision approaches highlight potential pitfalls of analyses by organ site (e.g., lung, breast) that lump together variations in genetic or social risks, different genomic signatures, treatments, and implications for prognosis and quality of life (QOL) [12-14]. Population aging and movement toward whole-person health similarly underscore opportunities to assess and model a broader constellation of factors (e.g., multiple chronic conditions, functional status, degrees of caregiver support) and understand the effects of incentivizing wholistic care approaches on cancer-related outcomes [15, 16]. Paralleling rapid clinical advancements is increasing attention on controlling costs, including addressing financial toxicity and financial distress [17, 18]. Cancer care costs increasingly outpace other areas; for example, they comprised 43% of 2020 Medicare Part B spending [19]. Additionally, recent analyses found cancer survivors were nearly 4 times more likely to declare bankruptcy and experienced credit score declines persisting up to nearly 10 years post-diagnosis [20]. Financial distress is associated with higher symptom burden, worse QOL, and lower adherence to recommended care [21-23]. Challenges quantifying costs or balancing cost with access to high-quality care and clinical innovations are certainly not new [24]. However, the scope and duration of cancer care costs at individual, family, and population levels beget opportunities for multilevel measure development, innovative modeling approaches, and data linkages, as well as interventions that integrate financial considerations into goals of care discussions and financial navigation [25-27]. The speed of cancer care innovation is also matched by rapid transformations in the healthcare system landscape across the cancer continuum. For example, in 2017 more oncology physician practices (50%–55%) reported vertical integration with a hospital or health system compared with any other specialty, up from ~20% in 2007 [28, 29]. Trends toward greater health system integration, new affiliation models, expansion of non-traditional players into the care delivery sector, and pervasive care fragmentation underscore opportunities to develop and adopt richer measures of organizational structure, functioning, policies, norms, and coordination across the cancer continuum [30]. Similarly, evolving roles and approaches to care are arising from clinical innovation (e.g., home-based screening, oral anti-cancer agents) paired with patient volumes rapidly outpacing oncology workforce capacity. For example, some care delivery models, state policies, and billing guidelines are enabling Advanced Practice Professionals, community health workers, patient navigators, home care, and other care team members to practice at the top of their license or certification [16]. These trends challenge future research to more precisely assess where and who is delivering care and to advance methods suitable for evaluating contributions of a growing constellation of collaborators and settings to cancer-related outcomes of interest. Additionally, the field has seen increased focus on understanding and addressing adverse social drivers of health, financial hardship, and social risks (e.g., transportation, food, housing instability) and their influence on persistent cancer health disparities [31]. For example, eliminating cancer health disparities was one of eight goals in the 2024 National Cancer Plan [32, 33]. NCI has a long history of supporting efforts to improve cancer health disparity measurement [34] given quantifying heterogeneity in cancer incidence and mortality is part of the Annual Report to the Nation on Cancer. However, efforts to address social risks and drivers of health via the healthcare system—as well as related measures and approaches for tracing impact on cancer outcomes—are still nascent. Collectively, these trends and others noted at the opening of this section underscore numerous opportunities for cancer health services scientists to address persistent and emerging measurement and methodological challenges. We highlight several opportunities for future measurement and methods development or refinement below. We simultaneously encourage the field to identify and pursue numerous others not discussed here. Many trends above may necessitate new measurement paradigms (e.g., whole-person cancer care, measurement-based care). Others underscore the need for dedicated attention toward solving persistent, yet fundamental measurement challenges (e.g., evolving care delivery settings, usual care, organizational characteristics). Given these trends, we discuss five example areas for measurement-focused research attention below. Evidence exists for the benefit of whole-person care models, yet defining components of whole-person cancer care requires conceptual elaboration, refinement, and standardization [35]. Both new measures and novel person-centered methods are essential to designing and optimizing whole-person focused systems of cancer care. In 2024, building from work in primary care, the Integrative Oncology Leadership Collaborative (IOLC) defined whole-person cancer care as an approach that integrates conventional cancer treatments with evidence-based complementary therapies and/or lifestyle interventions, addresses the physical, emotional, social, and spiritual aspects of a person's life, and focuses on what matters most to the patient [36, 37]. The IOLC definition and related minimal-required elements are based on the Two-Circle Model of Whole-Person Care [38], which reframes current disease-focused approaches toward one that is person-centered, relationship-based, and recovery and health-promotion focused. An emphasis on person-centered care, coordination, continuity and integration, and relationships are distinguishing characteristics of the whole-person paradigm and are conceptualized as features most likely to improve population health, access, quality, and lower costs. The Two-Circle framework also highlights roles, services, and workforce changes needed to implement, scale up, and sustain this type of care. New payment models are also important to support and incentivize a transition to whole-person care. Many health services measurement and methodological approaches developed or operationalized around a single disease, organ system, specific health care setting, or payer will continue to be useful in evaluating models of whole-person cancer care (e.g., cancer registries, Consumer Assessment of Healthcare Providers & Systems [CAHPS]) [39-41]. However, person-centered measures of unmet needs, experiences of care involving larger care teams, well-being (physical, emotional, social), care costs, and medical financial hardship will require further conceptual, lexical, and methodologic development in the context of whole-person care [42]. Ensuring such measures are accessible and meaningful for all patients, and interpretable as predictors and moderators of whole-person health outcomes will require mixed methods studies that go beyond traditional psychometric approaches to establish validity and interpretation [43]. Measurement of whole-person outcomes also requires accommodating, sometimes simultaneously, for within-person and group-level change, and methods able to address differences between individuals on a collection of measures or scale dimensions (e.g., almost matching exactly methods) [44]. The NIH National Center for Complementary and Integrative Health's 2021 Workshop on Methodological Approaches for Whole Person Research discussed several such measurement and methodological opportunities [45]. Measurement-based care (MBC) is an emerging approach in chronic disease management, including cancer care [46], generally defined as “systematic evaluation of patient symptoms before or during an encounter to inform” [47](p324) care-related decisions. Patient-reported outcomes (PROs) form MBC's foundation, providing critical tools for assessing targeted needs for distinct populations (e.g., older adults, adolescents), tumor and treatment types, and care phase [48]. Well-validated instruments essential for MBC in oncology exist (e.g., needs assessment, symptoms, functional status, social risks) [49-51]. However, some domains remain underdeveloped, including measurement of treatment burden, patient engagement, and care experiences. Both existing and newly developed measures also require adaptation and validation to meet accessibility needs of all patients, including groups understudied in measure development research such as older adults and people with differing degrees of English language proficiency. Additionally, integrating these measures into feasible, efficacious MBC interventions and coupling them with evidence-based decision support necessary to prioritize and comprehensively address the constellation of needs identified, requires further development. Alternative settings of care beyond traditional inpatient-outpatient distinctions are also rapidly arising, including: telehealth, remote patient monitoring, home-based care (e.g., hospital-in-the-home, self-management support), distributed clinical trials, and consumer-oriented platforms (e.g., Amazon Care, CVS MinuteClinic). These settings offer new opportunities for interdisciplinary collaboration, improved efficiency, and access. However, the field lacks screening measures to match these settings to patient needs and resources, risk stratify, and predict clinical complications. Also needed are bespoke measures demonstrating solid measurement properties in these settings for care quality, safety, clinical outcomes, budget impact, costs, value, and experiences of patients, caregivers, and staff [52-54]. Measuring strategies and contextual factors contributing to the adoption and sustainment of delivery models employing alternative care settings are also important for dissemination and adaptation [54, 55]. Usual care is a frequent comparator in randomized trials and pragmatic trials (e.g., cluster randomized, stepped wedge), particularly those evaluating health interventions or healthcare delivery interventions studies or in health services also usual care as a comparator usual care is the is important to understand the to which any pragmatic exists in the comparator to treatment However, significant and challenges exist in the and measurement of usual care and of the heterogeneity in what usual care Usual care is a and and between settings usual care may and is context the of valid usual care is not well in studies and approaches to and modeling usual care, and related and between settings are important Cancer health services for example, approaches to to evidence-based practices and interventions in science characteristics of the care delivery and across the cancer continuum influence access, coordination, clinical outcomes, and costs. in new affiliation models, and expansion of non-traditional players in care delivery underscore opportunities to develop and adopt richer measures of organizational integration, structure, functioning, policies, norms, and coordination across the cancer continuum. For example, the promise of improved access, quality, and costs, studies the impacts of health system integration the cancer and studies from the broader field effects in to differing and measures across studies The field beyond measures of toward richer measures different and degrees of organizational integration to understand drivers of costs, quality, of guidelines and clinical and of or cancer care. on and approaches to and organizational health system characteristics For example, and of organizational integration of and integration, with for measures policy and other of inform at the population understanding of care, and health system efforts to improve health for all the COVID-19 and focus on addressing social risks underscore opportunities to develop richer measures of integration and coordination with health and community Additionally, are opportunities to develop measures in degrees of affiliation or integration to understand emerging affiliation and clinical trial and opportunities to advance development and access to measures of payer organizational and characteristics Care and other trends methodological advancements focus on approaches able to model factors simultaneously, and address heterogeneity across or levels of methods effects of a of factors on cancer outcomes will continue to have a particularly However, is great for cancer health services scientists to and develop methods and mixed methods. and modeling are critical for understanding and effects of factors at different levels (e.g., health system, on cancer outcomes, costs, care and understanding the of factors cancer Additionally, they are important for evaluating or interventions and related (e.g., multilevel modeling has understanding factors cost in cancer impacts of and policy and numerous other A of multilevel approaches exists in health services research and other however, opportunities remain. For example, and modeling methods and to work also focus on of for Additionally, with effects and models in health services opportunities exist to from multilevel methods to understand or other cluster heterogeneity Similarly, approaches for modeling or differences in models, be For example, such methods be useful for understanding and outcomes more and differences in the of or health system also be useful in understanding or other in of existing or advancement of new multilevel modeling approaches also focus on measurement of approaches, and related measurement methods is important cancer health services research is to address current and future challenges. these methods have important for understanding oncology workforce and outcomes addressing social risks care quality disparities and methods will also be to understanding health system and integration as well as the of new clinical technologies and changes research explore the adaptation of existing methods for understanding and the impacts of policy change, and to genomic and other data improve the of or disease care and care team toward modeling together with and is For example, from methods to social data on modeling patient clinical innovation and organizational changes and methods for in data also useful for effects on policy and and methods also be useful to or methods data modeling that for volumes of with are critical the field is to model the heterogeneity in cancer screening and disease characteristics and care and outcomes at the patient and population For example, data and (e.g., of such new methods address this while data to meaningful volumes and methods that or For example, modeling has to identify in care and and paired with models models, effects to predictors and with outcomes these and related methods (e.g., cluster of based on multiple or These approaches are critical for tracing care and patient and across care be useful and to modeling of usual care or other and differences (e.g., to understand populations to cancer with emerging approaches to improve and of data from medical and other intelligence and methods have rapidly emerging roles in clinical care and health services research. discussion of language and other methods is our scope, but their in and other aspects of cancer care are expanding rapidly However, significant exist in evaluating the characteristics of in clinical settings and the field lacks to in specific care and with specific services research has generally more some other the to which methods will address measurement and methodological challenges. However, methods have and be for numerous aspects of health services research, including: data and addressing and policy evaluation For example, an for and Research five methodological areas that benefit from including and for or health outcomes, and in and in analyses and other research areas methods with promise and that require methodological with methodological and for and modeling or data are of leveraging a long history in some areas of health services research modeling is in health policy development healthcare workforce and research However, a basic of the for found only exist for future research to adopt and advance particularly given the of new approaches for generating data and science platforms of health services models For example, NCI has the Cancer and to advance modeling of the history of disease and cancer including cancer and screening have in understanding that a in of screening reduce cancer deaths by and cancer deaths by the contributions of in screening, treatment, and other aspects of care on cancer mortality and in analyses of screening data from cancer registries, clinical trials, and other and offer a potential for other cancer health services scientists in For example, methods from or and These approaches are critical in cancer health services research given the of cancers and related care, particularly methods are also important for understanding value of care and more precisely understanding to usual care. However, to adoption we how existing mixed methods be and and develop more methods (e.g., rapid need further refinement to support adoption These methods have potential to inform measurement of care value, and in and understanding of variations in care delivery and outcomes the cancer continuum. focused on innovative data collection approaches and data integration is also particularly given declines in and increasing on patients, health professionals, and These ideas be important Our was to and offer some not all or promising We to focus on challenges and including with data social and other Additionally, we have to and the trends and opportunities noted have their and considerations not here. some may with our of measurement methodological is for the domains is across models and the need for and in addressing healthcare delivery measurement and methods challenges. has years of Research measurement and data challenges the is progress in some areas in a related the trends and challenges underscore that important opportunities remain. chronic cancer and to and new measures and methods for and modeling of health services and outcomes Given trends discussed and in the we that is for future cancer health services research to focus on measurement while simultaneously the of cancer care For example, some measurement discussed are fundamental (e.g., usual care, with However, attention to these fundamental measurement the field opportunities to assess and from the of increasingly cancer care. Many of the methodological opportunities discussed are in system science paradigms and are to address this heterogeneity and (e.g., multilevel they toward modeling the and between the factors shaping care delivery, clinical outcomes, costs, and other important Similarly, from science (e.g., organizational health or multilevel research be useful for and or as in several factors continue to influence the approach to measurement and methodological including: data and and evolving data refinement of approaches able to (e.g., health system, while simultaneously enabling data and are For example, opportunities remain for innovation in data mixed and approaches for populations Additionally, continuity of data and innovation around measurement and methodological are critical given declines in in and toward research, and increasing on patients, caregivers, health professionals, and health system The NCI Healthcare Research research in of the areas discussed For example, Cancer Research in 2020 and were in a of the National Cancer including an of cancer health research also data of to the field including and Additionally, has on (e.g., and the Healthcare Research is to improve health for individuals and populations by innovative research to understand and improve care delivery across the cancer continuum science to develop and measures and methods to cancer care delivery, related interventions, and impacts on cancer-related focused on measurement and methodological to cancer health services and healthcare delivery research be via NIH and/or NCI opportunities In this several opportunities for future measurement or methodological may be or existing innovative linkages, or primary data collection with existing or clinical Others will require measurement development. be for in or will benefit from and in data and science policy that address data and Additionally, all will benefit from and with and data from other patients, caregivers, patient health system, and policy these measurement and methodological opportunities is critical to understanding and improving care, outcomes, and costs across the cancer continuum. cancer health services research is to be at this and into the we a focus on by attention to addressing in our methods and our Both to and design, and of and of the We from several and was as part of the as of the U.S. and are those of the and this not be as the of the U.S. of and the National of or the National Cancer The declare of interest. not new data or the research.
Characterizing false information based on linguistic analysis is important to understand the factors that affect the proliferation of fake news in the media. Previous work has identified some linguistic regularities that suggest a trend towards decreased complexity, polarization and sentiment in false information. This study is aimed at identifying linguistic differences between real and fake news using a corpus of annotated media news in Spanish via the automatic analysis of linguistic cues using dictionaries of lexical norms. We focus on lexical aspects of complexity, familiarity and sentiment. Consistent with previous results, we found that fake news are associated with lower cognitive loads, reflected by reduced sentence complexity, and increased lexical familiarity and imaginability. Moreover, and consistently with previous results, the analysis revealed that fake news are associated with more polarized emotional content.
Abstract Previous research on syntactic complexity is primarily focused on the synchronic distribution of clausal and phrasal features and the diachronic shift from clausal elaboration to phrasal compression. However, the interrelationship between clause complexity and phrase complexity remains unexplored. This study investigated syntactic complexity at different linguistic levels across three disciplinary groups (Social Sciences, Humanities and Natural Sciences) using a corpus of research article abstracts. Sentence complexity was measured by the number of clauses per sentence, clause complexity by the number of clausal constituents per clause, and nominal group (NG) complexity by the number of words per NG. The results show that: (1) sentences are the least complex in Natural Science (NS) texts; (2) clauses are also the least complex in NS, despite having the highest average number of clausal constituents; (3) NGs are the most complex in NS texts. Furthermore, the study found that NG complexity could be more accurately measured by the number of premodifiers of the head noun (HN) of the NG. These findings have important implications for instructing English as a Foreign Language (EFL) learners in discipline-specific academic writing.
In recent years, syntactic and semantic analysis tools have become increasingly important in various subfields of Natural Language Processing (NLP). These tools enable automatic parsing of large-scale sentences in language corpora, allowing researchers to uncover syntactic structures and statistical regularities of a given language. This study focuses on the development and evaluation of syntactic parsing models for the Uzbek language, employing two widely used approaches: constituency parsing and dependency parsing. For constituency parsing, a rule-based system was developed to identify noun and verb phrases along with their internal constituents. For dependency parsing, a set of hand-crafted linguistic rules was created and applied to syntactically analyze simple Uzbek sentences. As a result of this work, a dependency-based syntactic treebank for Uzbek-Named UzTreebank was constructed. The treebank includes 20,000 automatically parsed simple sentences, of which 10,000 were manually annotated. Additionally, 36 syntactic templates of simple sentences were identified, and 50 linguistic rules were formalized and integrated into the system. The suboptimal performance of the system at its current stage is primarily attributed to the absence of hybrid modeling approaches and the limited size of the training corpus. The paper presents an overview of the rule-based architecture, parsing results, and the current stage of syntactic resource development for the Uzbek language.
The present study extends recent work on Universal Dependencies annotations for second-language (L2) Korean by introducing a semi-automated framework that identifies morphosyntactic constructions from XPOS sequences and aligns those constructions with corresponding UPOS categories. We also broaden the existing L2-Korean corpus by annotating 2,998 new sentences from argumentative essays. To evaluate the impact of XPOS-UPOS alignments, we fine-tune L2-Korean morphosyntactic analysis models on datasets both with and without these alignments, using two NLP toolkits. Our results indicate that the aligned dataset not only improves consistency across annotation layers but also enhances morphosyntactic tagging and dependency-parsing accuracy, particularly in cases of limited annotated data.
International audience
Recent experimental studies have examined GOODNESS IS BRIGHTNESS and a host of other primary metaphors. However, complex mappings such as INTELLIGENCE IS BRIGHTNESS have been largely ignored, nor has there been any attempt to distinguish their effects from those of primary metaphors such as GOODNESS IS BRIGHTNESS. The current study assesses both the nonprimary metaphoric mapping INTELLIGENCE IS BRIGHTNESS and the well-documented primary metaphor GOODNESS IS BRIGHTNESS in a visual priming task. The study finds that a bright background encourages photos of faces to be rated as both more intelligent and wellintentioned, though the background does not significantly affect either attribute alone. This suggests that two metaphors with the same source domain can reinforce each other. The study also underscores the difficulty in assessing a non-primary mapping in isolation from other factors.
One of the fundamental tasks in natural language processing (NLP) is dependency parsing, which involves analyzing the grammatical structure of sentences by establishing relationships or dependencies between words. In this paper, we examine the difficulties and methods for performing dependency parsing for Bangla text, a language with complex morphology and distinctive syntactic properties. The article addresses the value of dependency parsing in capturing the linguistic subtleties of Bangla sentences and their applications in various NLP tasks. A method has been proposed to develop dependency parsing on Bangla text using a graph-based approach. A parsing tree is generated from a directed graph using Bangla input. The proposed system is achieved overall 68% accuracy which is evaluated using Bangla dependency corpus. This method enriches Bangla language linguistic resources and annotated corpora, facilitating the language’s global use. The resultant tree is also evaluated using evaluation metrics.
Cross-domain constituency parsing is still an unsolved challenge in computational linguistics since the available multi-domain constituency treebank is limited.We investigate automatic treebank generation by large language models (LLMs) in this paper.The performance of LLMs on constituency parsing is poor, therefore we propose a novel treebank generation method, LLM back generation, which is similar to the reverse process of constituency parsing.LLM back generation takes the incomplete cross-domain constituency tree with only domain keyword leaf nodes as input and fills the missing words to generate the cross-domain constituency treebank.Besides, we also introduce a span-level contrastive learning pretraining strategy to make full use of the LLM back generation treebank for cross-domain constituency parsing.We verify the effectiveness of our LLM back generation treebank coupled with contrastive learning pre-training on five target domains of MCTB.Experimental results show that our approach achieves state-of-theart performance on average results compared with various baselines.
Cross-domain constituency parsing is still an unsolved challenge in computational linguistics since the available multi-domain constituency treebank is limited. We investigate automatic treebank generation by large language models (LLMs) in this paper. The performance of LLMs on constituency parsing is poor, therefore we propose a novel treebank generation method, LLM back generation, which is similar to the reverse process of constituency parsing. LLM back generation takes the incomplete cross-domain constituency tree with only domain keyword leaf nodes as input and fills the missing words to generate the cross-domain constituency treebank. Besides, we also introduce a span-level contrastive learning pre-training strategy to make full use of the LLM back generation treebank for cross-domain constituency parsing. We verify the effectiveness of our LLM back generation treebank coupled with contrastive learning pre-training on five target domains of MCTB. Experimental results show that our approach achieves state-of-the-art performance on average results compared with various baselines.
Abstract Morphological analysis is a foundational task in natural language processing (NLP) and is particularly challenging for low-resourced and morphologically rich languages such as Kangri. Despite substantial numbers of speakers, Kangri lacks annotated corpora, computational tools, and lexicons, making linguistic analysis and downstream processing difficult. This paper presents a hybrid morphological analyzer for the Kangri language that integrates rule-based suffix analysis, lexicon extraction, and efficient machine learning models. A lexicon and suffix transformation rules were automatically induced from the Universal Dependencies (UD) Kangri Treebank. The rule-based morphological analyzer achieved an accuracy of 59\% on the UD test set. A machine learning baseline using TF--IDF character n-grams with Logistic Regression achieved 64.61\% accuracy, while an enhanced model incorporating POS tags improved performance to 67.40%. The results demonstrate that combining linguistic heuristics with statistical learning substantially improves lemma prediction and morphological interpretation for Kangri. This work establishes an initial computational morphology framework for Kangri and provides a foundation for further NLP tool development.
Stella Markantonatou, Vivian Stamou, Stavros Bompolas, Katerina Anastasopoulou, Irianna Linardaki Vasileiadi, Konstantinos Diamantopoulos, Yannis Kazos, Antonios Anastasopoulos. Proceedings of the 21st Workshop on Multiword Expressions (MWE 2025). 2025.
Peer reviewed: True
Cette thèse présente le corpus NaijaSynCor-Prosody, une ressource innovante qui intègre des annotations phonétiques détaillées dans un corpus syntaxique existant. Chaque token du corpus est associée à des annotations décrivant la hauteur, la durée, l’intensité et d’autres attributs prosodiques de chaque syllabe. Ces annotations incluent notamment des contours prosodiques stylisés produits à l’aide du modèle SLAM 3, développé au cours de cette thèse. Cette fusion de données morphosyntaxiques et prosodiques permet des analyses quantitatives de l’interaction entre intonation et syntaxe dans le Naijá, ou pidgin nigérian. À l’aide de ce corpus, nous étudions la différenciation prosodique des unités lexicales qui remplissent également une fonction grammaticale, par exemple des auxiliaires préverbaux marqueurs de TAM. Les contrastes prosodiques entre usages lexicaux et grammaticaux ne sont pas uniformes selon les éléments, bien que notre corpus révèle deux grandes classes prosodiques d’auxiliaires. Un groupe (dey, go, bin) présente une hauteur faible et une courte durée, tandis que les autres se caractérisent globalement par une hauteur plus élevée et une durée plus longue. Deux d’entre eux, make et no, sont particuliers car ils combinent une hauteur élevée et une faible durée. Une analyse plus approfondie de leurs environnements syntaxiques montre qu’ils ont également des distributions atypiques parmi les auxiliaires. Par ailleurs, plusieurs adverbes annotés présentent une distribution syntaxique analogue à celle des auxiliaires et un profil prosodique similaire aux auxiliaires à haute hauteur et longue durée. Ce travail soulève des questions plus générales sur l’utilisation des annotations prosodiques pour remettre en question les étiquettes morphosyntaxiques existantes. La thèse fournit ainsi à la fois une nouvelle ressource linguistique et un cadre méthodologique pour intégrer l’analyse prosodique dans l’étude syntaxe–prosodie à partir de corpus.
Treebanks are critical resources in Natural Language Processing (NLP), supporting parser development, linguistic research, and the evaluation of large language models. While Tamil has seen progress in Universal Dependencies (UD) treebanking, existing corpora have been restricted to prose texts, leaving its vast poetic tradition underrepresented. This paper presents the first effort to be made to construct a syntactic treebank for Tamil poetry, specifically focussing on the ThirukkuRaḷ, which is composed in kuRaḷ veṇpā form. A central challenge in this work is posed by the treatment of multiword tokens (MWTs) and elliptical constructions, both of which are observed to occur frequently in Tamil verse due to its agglutinative morphology and metrical constraints. An annotation strategy is proposed within the Enhanced UD (EUD) framework to systematically address five major types of ellipsis—casal, verbal, adjectival, comparative/simile, and cumulative—alongside complex MWT patterns. These annotations not only enhance the representation of Tamil poetic syntax but also broaden the applicability of UD guidelines to underrepresented genres. The contribution is shown to underscore the linguistic and computational importance of capturing the structural specificities of Tamil poetry, while establishing a foundation for future cross-linguistic and literary treebanking efforts.
Thermal imaging technology, known for its noncontact and noninvasive nature, offers distinct advantages in computerized emotion sensing. In the literature, a decrease in nose-tip temperature has been associated with dynamic subjective arousal. However, these studies were limited by their focus on a few regions of interest, neglecting a comprehensive analysis of the entire face, and not accounting for the temporal dynamics of thermal changes. To overcome these limitations, we propose an analytical method for facial thermal images using statistical parametric mapping (SPM), which was developed for functional brain image analysis. We developed semiautomated preprocessing protocols to effectively realign and standardize facial thermal images. To validate these analyses, we recorded the thermal images of participants’ faces and assessed dynamic valence and arousal ratings while they observed emotional films. The proposed SPM analyses revealed significant negative associations with dynamic arousal ratings at the nose tip and forehead. The analyses incorporating temporal disparity revealed more forehead clusters than the analyses assuming no delay. These findings validate the proposed pixel-based facial thermal image analysis method using SPM. The results suggest that computerized pixel-based analysis of facial thermal images can be used to estimate dynamic emotional states, with potential applications in various human behavioral fields, including mental health diagnosis and marketing research. • We developed a pixel-based facial thermal image analytical method using SPM. • The method for emotion sensing was tested with an experiment showing emotional films. • Facial thermal images and dynamic valence/arousal ratings were measured. • Negative associations with arousal ratings were detected at the nosetip and forehead. • The data validate the pixel-based facial thermal image analysis for emotion sensing.
We expand the second language (L2) Korean Universal Dependencies (UD) treebank with 5,454 manually annotated sentences. The annotation guidelines are also revised to better align with the UD framework. Using this enhanced treebank, we fine-tune three Korean language models and evaluate their performance on in-domain and out-of-domain L2-Korean datasets. The results show that fine-tuning significantly improves their performance across various metrics, thus highlighting the importance of using well-tailored L2 datasets for fine-tuning first-language-based, general-purpose language models for the morphosyntactic analysis of L2 data.
The present study extends recent work on Universal Dependencies annotations for secondlanguage (L2) Korean by introducing a semiautomated framework that identifies morphosyntactic constructions from XPOS sequences and aligns those constructions with corresponding UPOS categories.We also broaden the existing L2-Korean corpus by annotating 2,998 new sentences from argumentative essays.To evaluate the impact of XPOS-UPOS alignments, we fine-tune L2-Korean morphosyntactic analysis models on datasets both with and without these alignments, using two NLP toolkits.Our results indicate that the aligned dataset not only improves consistency across annotation layers but also enhances morphosyntactic tagging and dependency-parsing accuracy, particularly in cases of limited annotated data.
The quantity and variety of Old Irish text which survives in contemporary manuscripts, those dating from the Old Irish period, is quite small by comparison to what is available for Modern Irish, not to mention better-resourced modern languages.As no native speakers have existed for more than a millennium, no more text will ever be created by native speakers.For these reasons, text surviving in contemporary sources is particularly valuable.Ideally, all such text would be annotated using a single, common standard to ensure compatibility.At present, discrete Old Irish text repositories make use of incompatible annotation styles, few of which are utilised by text resources for other languages.This limits the potential for using text from more than any one resource simultaneously in NLP applications, or as a basis for creating further resources.This paper describes the production of the first Old Irish text resources to be designed specifically to ensure lexical compatibility and interoperability.
This study investigates the sentiment polarity (positive, negative, neutral) and specific emotions (joy, sadness, anger, surprise, trust, anticipation, disgust, and fear) expressed by Generation Z in digital platform comments regarding seven female duets with famous male singer. A dataset of 500 digital comments (250 from YouTube, 125 from Twitter, 125 from Instagram) was collected. The sample was then refined to include comments from 100 individuals (50 men, 50 women) affiliated with a private university in Mexico City, ensuring gender balance. Sentiment polarity was classified using a Bidirectional Encoder Representations from Transformers (BERT) model, with its hyperparameters (learning rate, epochs, batch size) optimized via a Particle Swarm Optimization (PSO) metaheuristic, leading to a 4% accuracy improvement over default settings. Emotion detection was performed concurrently using the NRC Emotion Lexicon, a lexical database mapping terms to eight emotional categories. Results indicate a clear correlation between musical tone and expressed sentiment: melancholic duets elicited predominantly negative sentiments, whereas more energetic collaborations generated a higher proportion of positive comments and the emotion 'joy.' Furthermore, significant differences using $\chi^2$ ($p < 0.01$) were observed in the distribution of 'anger' and 'sadness' between intimate and collaborative duets. These findings offer valuable insights into how Generation Z, segmented by gender, emotionally interprets this singer's musical productions. This research has significant implications for developing targeted music marketing strategies and content production for digitally native audiences.
This manuscript proposes the S M Nazmuz Sakib Dependency-Focus Principle for Bengali sentence structure and introduces a derived scalar quantity, the Sakib constant, defined over dependency treebanks. Informally, the principle states that in attested Bengali usage, core arguments (subjects and objects) cluster closer to the verbal head than peripheral modifiers (adverbials and clausal adjuncts), and that the ratio between these average distances is numerically stable across corpora. Using real statistics from the UD Bengali-BRU treebank and the Bengali section of the Bengali-Magahi PUD treebank, we define the Sakib constant K Sakib as the ratio between average dependency lengths of core versus peripheral relations, and compute its value for UD Bengali-BRU. Ten figures based on genuine counts and averages illustrate tense and case distributions, relation frequencies, and core versus non-core dependency lengths for Bengali and, for comparison, Magahi. The proposal is presented as a precise hypothesis, mathematically well-defined and empirically grounded in existing treebank data, but still requiring broader testing for confirmation and cross-linguistic generalisation.
The article addresses the concept of linguistic purism in the context of globalization, when language barriers weaken and borrowings become commonplace. Purism, as an ideology, focuses on preserving the purity of the language, its stability and protection from external influences. The basic principles of purism, such as protecting the national language from foreign borrowings, preserving traditional norms, and countering linguistic changes are considered in the article. Two types of purism can be singled out, such as gustatory, based on subjective criteria, and scientific, which requires further clarification. The typology of linguistic purism is considered in terms of orientation, goals, and the nature of relations to linguistic facts. The experience of puristic activity in different languages and, accordingly, different societies, cultures and historical contexts are analysed. In general, linguistic purism is becoming the subject of topical discussions in the context of standardization and codification of the Russian literary language. Key words: language, vocabulary, linguistic norm, codification of linguistic norm, socio-cultural functions of language, sociolinguistics, psycholinguistics, linguistic (linguistic) purism.
This study investigated the neurophysiological and affective responses elicited by nature-inspired indoor design elements, including curvilinear forms (CL), nature views (N), and wooden interiors (W), in a virtual environment, and their effects on cognitive performance. Thirty-six participants experienced one control and three experimental conditions in a within-subject design. Electroencephalography (EEG) was used to record neural activity, relaxation and valence ratings assessed affective states, and standardized tasks measured cognitive performance. The W condition elicited EEG patterns indicative of relaxed attentional engagement, including increased alpha-to-theta (ATR) and alpha-to-beta (ABR) ratios, and a decreased theta-to-beta (TBR) ratio. These neural patterns were associated with higher self-reported relaxation and positive affect, and with enhanced cognitive performance relative to the control condition. In contrast, the CL and N conditions did not improve cognitive performance, and the N condition showed elevated physiological arousal, likely due to heightened visual stimulation. Regression analysis identified ATR and relaxation as significant predictors of cognitive performance, emphasizing the role of emotional stability and neural balance in supporting task engagement. Overall, the findings highlight the potential of nature-inspired design to foster a synergy between psychological relaxation and cognitive attention, though further research is needed across diverse spatial typologies to isolate specific design parameters.
The increasing availability of cross-linguistic databases dedicated to documenting morphosyntactic, lexical and phonological features has proliferated the use of such data for studies on language evolution and human history. However, most of these databases were not designed to ensure independence of features, such that it is not valid to jointly use all their features in large-scale statistical analyses assuming independence of inputs. Here, we curate published data from five large linguistic databases to generate two global-scale cross-linguistic datasets: GBI (from the Grambank dataset), and TLI (using inputs from the World Atlas of Language Structures, AUTOTYP, PHOIBLE and Lexibank). The datasets minimize logical dependencies of features and forms of strong statistical dependencies that go beyond phylogenetic and geographical signal. They are also made available in densified form, reducing the proportion of missing data. We document our curation principles and workflows to ensure reusability of this framework with other inputs or thresholds of independence. Our curation steps on both datasets reveal robust and comparable global patterns of structural linguistic diversity.
BACKGROUND: Emotion dysregulation is a central feature in trauma-associated disorders such as posttraumatic stress disorder (PTSD) and borderline personality disorder (BPD). However, it remains unclear whether emotion dysregulation is a transdiagnostic phenomenon closely linked to childhood trauma, or if disorder-specific alterations in emotion processing exist. Following a multimethodological approach, we aimed to assess and compare the reactivity to and regulation of emotions between patients with BPD and PTSD, as well as healthy controls, and identify associations with childhood trauma. METHODS: A total of 135 women, 43 healthy controls, 43 with BPD and 49 with PTSD, took part in a multimethodological assessment of emotional reactivity and regulation. Self-report measures were used to assess childhood trauma and emotion dysregulation. Additionally, participants performed a classic emotion regulation (ER) paradigm. Subjective emotional valence ratings and neurophysiological responses (P3 and late positive potential, LPP) were measured in response to negative, positive, and neutral pictures (emotional reactivity) and during active regulation vs. passive viewing of negative pictures (ER). RESULTS: Regarding emotional reactivity, during the experimental paradigm both patient groups reported lower emotional valence after viewing positive or neutral pictures compared to healthy controls. Furthermore, P3 amplitudes in response to neutral pictures were reduced in both patient groups and in response to negative pictures, specifically in patients with PTSD. Regarding ER, while both patient groups self-reported significant disturbances in ER, neither valence ratings nor neurophysiological responses assessed during the ER task (P3, LPP) differed from healthy controls. Across groups, childhood trauma was related to decreased emotional valence ratings on neutral and positive pictures and higher self-reported emotion dysregulation. CONCLUSIONS: Patients with BPD and PTSD exhibited a reduced emotional reactivity in response to positive and neutral information. Specifically, patients with PTSD demonstrated hypo-reactivity to neutral and trauma-unrelated negative stimuli, which might be due to altered attentional resource allocation following trauma. Although patients reported using adaptive ER strategies less frequently in daily life, they effectively implemented them when instructed to, highlighting important clinical and theoretical implications.
Formulaic Networks as Prototypical Categories: Combining the Ancient Greek Dependency Treebank with the Ancient Greek WordNet for a Pilot Study on the Iliad was published in Advances in Ancient Greek Linguistics on page 737.
Emotional experiences involve dynamic multisensory perception, yet most EEG research uses unimodal stimuli such as naturalistic scene photographs. Recent research suggests that realistic emotional videos reliably reduce the amplitude of a steady-state visual evoked potential (ssVEP) elicited by a flickering border. Here, we examine the extent to which this video-ssVEP measure compares with the well-established Late Positive Potential (LPP) that is reliably larger for emotional relative to neutral scenes. To address this question, 45 participants viewed 90 matched pairs of realistic videos and scenes. Consistent with prior work, reduced 7-8 Hz ssVEP amplitude was evident during emotional relative to neutral videos. However, this reduction in power was not specific to the driving frequency of 7.5 Hz, and in fact, Fourier transformation analyses limited to 7.5 Hz were not modulated by video content. Still, at the group level, the video-driven reductions in 7-8 Hz power and LPP modulation by scenes produced similarly large valence effects, and both measures strongly correlated with arousal ratings. Consistent with previous research, the scene-LPP was sensitive to specific emotional contents (erotica and gore) somewhat inconsistent arousal ratings. In contrast, the video-driven oscillation modulation did not show this content sensitivity and was better explained by individual arousal ratings per video clip. In sum, these results show that the 7.5 Hz flickering-border paradigm does not index emotional engagement with video stimuli, yet emotional videos do evoke robust decreases in 3-10 Hz oscillatory power that is somewhat distinct from emotional modulation of the scene-evoked LPP. Matched emotional video and scenes evoke large EEG responses compared with neutral content within-participant. Our findings align with previous research indicating that video modulation of power around the evoked 7.5 Hz ssVEP frequency (7-8 Hz) serves as a reliable emotional measure. However, further analyses reveal that this effect is attributable to a general decrease in power across the 3-10 Hz frequency range.
Decision confidence is a prototypical metacognitive representation that is thought to approximate the probability that a decision is correct. The perception of being correct has also been associated with affective valence such that being correct feels more positive and being mistaken more negative. This suggests that, similarly to confidence, affective valence reflects the probability that a decision is correct. However, both fields of research have seen very little interaction. Here, we test if affect, similarly to confidence reflects probability that a decision is correct in two perceptual decision-making experiments where we compare the relationships of theoretically relevant variables (e.g. evidence, accuracy, and expectancy) with both confidence and affect ratings. The findings indicate that confidence and affect ratings are similarly sensitive to changes in accuracy, evidence, and expectancy, indicating that both track the subjective probability that a decision is correct. We identify various mechanisms that can explain these results. We also envision future research for clarifying the role of cognitive and affective aspects of metacognition relying on deeper integration of the respective research fields.
Over the past two decades, quantitative models and statistical methods have been increasingly applied to syntax research, with treebanks emerging as a vital resource for this field. However, tools specifically designed for calculating syntactic metrics remain relatively scarce compared to traditional word frequency measures. This gap hinders the deeper integration of quantitative methods into syntax research. In response, we have developed a tool called QuanSyn, which is tailored for quantitative syntax analysis. Based on treebanks, QuanSyn can calculate metrics such as dependency distance, hierarchical distance, dependency direction, and valency-related syntactic metrics. Additionally, it facilitates the construction of linguistic networks and rapid fitting of functions. QuanSyn can help lower the barriers to accessing syntactic quantitative analysis, thereby promoting the broader application of quantitative methods in syntax research.
Detection of out-of-distribution (OOD) samples is cru-cial for safe real-world deployment of machine learning models. Recent advances in vision language foundation models have made them capable of detecting OOD sam-ples without requiring in-distribution (ID) images. How-ever, these zero-shot methods often underperform as they do not adequately consider ID class likelihoods in their detection confidence scoring. Hence, we introduce CLIPScope, a zero-shot OOD detection approach that normalizes the confidence score of a sample by class likelihoods, akin to a Bayesian posterior update. Furthermore, CLIPScope incor-porates a novel strategy to mine OOD classes from a large lexical database. It selects class labels that are farthest and nearest to ID classes in terms of CLIP embedding distance to maximize coverage of OOD samples. We conduct ex-tensive ablation studies and empirical evaluations, demon-strating state of the art performance of CLIPScope across various OOD detection benchmarks. Code is available at https://github.com/ful001hao/CLIPScope.
Background and aims: Despite a previously reported connection between compulsive sexual behaviors (CSB), such as problematic pornography use, and heightened cue-reactivity, empirical evidence of the alteration of processes responsible for increased salience attribution to erotic cues remains sparse. Drawing on similarities with addiction models, this study explores the neuronal mechanisms of CSB through the use of appetitive conditioning and extinction with erotic and monetary rewards. Methods: Thirty-two heterosexual males struggling with CSB (age: 28.9 ± 7.1), and 31 healthy matched participants (age: 27.8 ± 5.6) underwent active appetitive conditioning and extinction tasks in fMRI. The effects of conditioning and extinction towards cues of erotic and monetary rewards were measured via self-assessment (valence and arousal rating towards cues), behavior (reaction times), and brain reactivity. Results: In conditioning, subjective ratings increased, and reaction times were faster for both erotic and monetary cues among participants with CSB, along with altered activity in ventral striatum (vStr), dorsal anterior cingulate cortex (dACC), and anterior orbitofrontal cortex (aOFC). In extinction, self-assessment ratings remained elevated in the CSB group for both cues in a non-reward-specific fashion, accompanied by altered activity of dACC and vStr. Discussion and conclusions: These findings suggest enhanced incentive salience attribution to conditioned cues, highlighting a generalized motivational and value-related transfer from rewards to the cues in participants with CSB. Additionally, despite the absence of rewards, the persistence of arousal and valence towards cues underscored the maladaptive extinction process. These insights advance the understanding of CSB's neurobiological underpinnings and its relation to addiction frameworks.
Abstract This study investigates the relationship between literature and poetry reading frequency and participants’ ratings of metaphors on key features: quality, aptness, familiarity, and comprehensibility. Using a set of Serbian poetic metaphors, we explored two main questions: how reading habits correlate with metaphor feature ratings, and whether the type of reading material (i.e., literature vs poetry) influences sensitivity to these features. The sample consisted of 140 native Serbian-speaking students from varied academic disciplines. Participants rated metaphors based on reading frequency (literature and poetry) using a 7-point Likert scale. Analysis showed that frequent readers generally gave higher overall metaphor ratings than infrequent readers, with significant differences noted particularly in familiarity and comprehensibility. Specifically, familiarity ratings yielded the most substantial differences between infrequent and frequent readers, which can indicate the influence of reading experience on the perceived recognition and understanding of metaphors. Aptness and quality ratings showed no significant differences, which suggests that familiarity and comprehensibility are more sensitive to variations in reading habits.
Acceptability judgments are one of the major tools for (psycho)linguists to assess speakers’ preferences for specific utterances in a given language, shedding light on the grammar of the language under study. However, it is well known that factors that are not related to grammaticality, such as frequency of exposure, cognitive constraints, and others, can influence the perceived acceptability of an utterance. We will use the system of wh-interrogatives in French as an example to study the impact of linguistic norms on what is considered “good” French. In three experiments, we show that adult L1 French speakers have internalized the dichotomy between variants that are considered “good French”, according to the norms, and those that are suited to more informal daily life situations. Speakers can express these differences when given the appropriate tools, but not with a unique general acceptability scale. In line with previous work, we argue that acceptability judgments are a useful task, but that they need to be refined to account for sociolinguistic factors that constrain speakers’ assessments (i.e., linguistic norms, but also speaker group and formality of the context of interaction).
We address the challenge of syntactic parsing for Urdu, a morphologically rich language, and present state-of-the-art results for both constituency and dependency parsing. This paper offers four major contributions: 1) the conversion of the CLE-UTB phrase structure treebank into a dependency treebank by developing language-specific head-word and phrase-to-dependency label mapping rules; 2) a novel sequence labeling scheme that transforms the parsing task into a unified representation; 3) the training of contextualized word representations on a large 220 million tokens Urdu corpus collected from the web; and 4) development of parsing framework using two learning paradigms, single-task and multi-task learning. Several post-processing rules are applied to improve the quality of the automatically converted dependency structure treebank. The proposed sequence labeling scheme enables the use of a shared architecture that learns the syntactic structures from both grammatical structures simultaneously and hence improves generalization. Experiments show that the multi-task learning setup significantly enhances parsing performance, achieving an F1 score of 91.39 for constituency parsing (an improvement of 3.29 points) and a labeled attachment score of 85.69 for dependency parsing (an improvement of 1.49 points). These results demonstrate that learning cross-task representations provides measurable benefits and advances the state of syntactic parsing for Urdu.
Music- and distraction-induced pain reduction have been investigated extensively, yet the main mechanism underlying music-induced analgesia remains unknown. In this study, to assess whether music-induced analgesia primarily operates through cognitive modulation, we used the cold pressor task and objectively compared the pain tolerances of participants in a four-group between-subjects design: a music group that listened to a music piece in the absence of any tasks, a music-and-attention-to-music group that listened to the same piece while also rating the arousal levels in the music, a music-and-attention-to-pain group that rated their pain levels while listening to the same piece, and a silence group as control. The group passively exposed to music playback did not show significantly higher pain tolerance compared to the silence group. However, pain tolerances in the music group negatively correlated with participants' self-reported arousal ratings of the music at the end of the experiment. The groups that engaged in an active task - whether evaluating the arousal levels in the music or reporting their experienced pain levels - demonstrated similarly higher pain tolerances compared to the silence group. These findings suggest that engaging in a task, regardless of whether it involves exteroceptive or interoceptive attention, can enhance pain tolerance.
Abstract In this paper, we analyze question-answer pairs and stand-alone questions within the spoken discourse of TED Talks, specifically focusing on the TED-Multilingual Discourse Bank. Our aim is to reveal various characteristics of questions through an annotation approach. We have developed a taxonomy, referred to as the TAQ-TED, to categorize types of questions and examine their information transfer and dialogue control functions, drawing on the Dynamic Interpretation Theory++ taxonomy of dialogue acts and their attribution in accordance with the Penn Discourse Treebank annotation guidelines. We outline the taxonomy, present our annotation results, and provide a preliminary cross-linguistic analysis comparing English questions with their Turkish and Portuguese translations. The TAQ-TED represents a promising initial framework for annotating questions in monologic discourse across multiple languages.
Abstract Standardized quantitative measurement of texts lies at the heart of digital approaches to humanities. Structure-based textual measures are known to be influenced by the choice of syntactic annotation schemes. Building on previous research, the present article further explores the relation between annotation schemes and the index of mean dependency distance (MDD) by comparing the treebanks of seventeen languages, respectively, within a tree representation (basic universal dependencies, BUD) and within a graphic representation (enhanced universal dependencies, EUD). Following the idea of decomposing annotation schemes into the combinations of analyses of specific constructions (coordinate structures, control constructions, and relative clauses), we design algorithms to identify them in the CoNLL-U format treebanks and explore their influences. It is found that the overall MDD of the EUD representation is statistically higher than that of BUD at corpus level, primarily affected by the coordinate structure due to its high frequency. At sentence level, all three constructions might contribute to either increased or decreased MDD, with stochastically intervening words and word order being two important determinants of the values of the measure. Finally, we propose and argue for the view that MDDs calculated under different annotation schemes should be regarded as different textual measures in nature. In sum, the present study provides another case study to deepen our understanding of the nature of syntactic annotation schemes and its relation with textual indices, which paves the way for standard measurement of texts in future humanities research.
This study compares AI-generated texts (via ChatGPT) and student-written essays in terms of lexical diversity, syntactic complexity, and readability. Grounded in Communication Theory—especially Grice’s Cooperative Principle and Relevance Theory—the research investigates how well AI-generated content aligns with human norms of cooperative communication. Using a corpus of 50 student essays and 50 AI-generated texts, the study applies measures such as Type-Token Ratio (TTR), Mean Length of T-Unit (MLT), and readability indices like Flesch–Kincaid and Gunning-Fog. Results indicate that while ChatGPT produces texts with greater lexical diversity and syntactic complexity, its output tends to be less readable and often falls short in communicative appropriateness. These findings carry important implications for educators seeking to integrate AI tools into writing instruction, particularly for second-language (L2) learners. The study concludes by calling for improvements to AI systems that would better balance linguistic complexity with clarity and accessibility.
Information on the relationship between facial thermal responses and emotional state is valuable for sensing emotion. Yet, previous research has typically relied on linear methods of analysis based on regions of interest (ROIs), which may overlook nonlinear pixel-wise information across the face. To address this limitation, we investigated the use of machine learning (ML) for pixel-level analysis of facial thermal images to estimate dynamic emotional arousal ratings. We collected facial thermal data from 20 participants who viewed five emotion-eliciting films and assessed their dynamic emotional self-reports. Our ML models, including random forest regression, support vector regression, ResNet-18, and ResNet-34, consistently demonstrated superior estimation performance compared to traditional simple or multiple linear regression models for the ROIs. To interpret the nonlinear relationships between facial temperature changes and arousal, saliency maps and integrated gradients were used for the ResNet-34 model. The results show nonlinear associations of arousal ratings in nose = tip, forehead, and cheek temperature changes. These findings imply that ML-based analysis of facial thermal images can estimate emotional arousal more effectively, pointing to potential applications of non-invasive emotion sensing for mental health, education, and human-computer interaction.
This article explores the role of five expressive punctuation marks – multiple exclamation marks (!!!), exclamation mark (!), full stop (.), ellipsis (...), and null punctuation (ø) – as cues to writer attitudes in CMC. Specifically, it investigates how the underlying meaning of expressive punctuation influences the perceived emotional valence of discourse referents within exclamative constructions. In a 1x5 between-subjects repeated measures design, valence ratings were collected for 120 discourse referents embedded in exclamative constructions manipulated by message finale punctuation mark (e. g., What a view!!!/!/./…/ø) on a 1 (negative) to 9 (positive) scale. For inherently positive discourse referents, a clear positivity hierarchy in the overall valence of embedded discourse referents emerges, indicating a differential influence of punctuation on perceived valence: multiple exclamation marks > exclamation marks > null punctuation > full stop > ellipsis. For inherently negative discourse referents, the differences between the conditions are less distinct. Notably, only multiple exclamation marks yield a significantly lower valence rating within that range of values. While the findings for inherently positive referents align with prior assumptions on the expressive meaning of different punctuation marks in CMC, the observed pattern for inherently negative referents cannot be readily explained by existing literature on expressive punctuation in CMC.