As a conceptual domain, time is a salient part of human life. Describing time and its relation to human thought is challenging yet essential for understanding human nature. In Indonesia, prior studies have concentrated on the structure of time expressions and grammatical time markers, leaving perceptions of time comparatively understudied. This preliminary study addresses this gap by investigating how a non-representative sample of 141 Indonesian participants, predominantly Javanese (75.9
Research has demonstrated reduced emotional reactivity in a second- (L2) compared to a first language (L1). One explanation for this states that L2 use is more cognitively demanding and thus leaves fewer resources available for emotion processing. Here, we specifically investigated this notion of differences in cognitive load during L1 and L2 use. EEG was used to investigate mismatch negativity during an auditory oddball paradigm. Native Swedish speakers with L2 English were asked to read text excerpts in their L1 and L2, whilst simultaneously listening to standard and deviant auditory tones to be ignored. The results showed significant oddball effects (i.e., mismatch negativity) in L1 but not in L2. Together with a lower accuracy in the L2 condition, these results imply a higher cognitive load during L2 reading. These findings are explained within a resource allocation framework and are suggested as a possible explanation of the underlying mechanisms of the foreign language effect.
Seasonal and other festivities with dance and music shape individuals’ aesthetic sensitivity to their cultural heritage. Yet, there is little empirical research about the influence of culture on dance perception. To fill the gap, this paper presents a pre-registered experiment with 120 participants (Iranian, English, Southeast Asian), that investigated the influence of cultural group (categorical variable) and enculturation (continuous variables) on aesthetic judgment to emotionally expressive Iranian classical dance stimuli. We also explored how culture and enculturation modulated responses to the question ‘how important is the cultural heritage of ‘Persian dance’ for you?’. Education, age, motion energy, and aesthetic responsiveness served as covariates across all models. Our data served to explore an important question in cross-cultural empirical aesthetics: which type of variable best predicts the influence of culture on questions about a cultural heritage like dance—categorical or continuous culture variables? Surprisingly, there were no overall differences in aesthetic judgment between the groups, although Iranians and English participants were most sensitive to joyful emotional expressivity of the dancer. Aesthetic judgment was modulated by English, but not Iranian enculturation. Conversely, Iranians found Persian dance more important than did the other two groups. In all analyses, aesthetic responsiveness was a significant covariate, while education, motion energy and age were not. Effect sizes (partial η2s and βs) were higher for continuous enculturation measures, than for the categorical variable cultural group. We recommend using continuous enculturation variables in cross-cultural empirical aesthetics research to allow higher granularity when exploring the impact of culture on perception.
Intelligence measures, such as the Raven's Coloured Progressive Matrices (CPM), rely on norm-referenced scores to interpret individual performance and guide educational or clinical interventions. However, when norm comparisons result in a severely skewed distribution, the validity of the norm sample must be scrutinized rather than accepting implausible interpretations. In this study, we challenge existing norms and improve the suitability of the CPM for Indian children attending government schools by developing new norms that better reflect their socio-demographic and educational contexts. Our sample includes 1,752 children in Delhi, Hyderabad, and Patna from two large-scale projects. Using existing CPM norms for India, we found that 78.5% of our sample scored in the bottom 10th percentile with an IQ score of 80 or lower, highlighting a heavily skewed distribution. Upon reviewing the existing norm sample, we identified several limitations: its size was five times smaller than our sample, the sample's linguistic background was underspecified, and it predominantly represented children from higher socio-economic backgrounds attending private schools. This norm sample fails to represent the majority of Indian pupils, particularly those in government schools, who constitute approximately two-thirds of the pupil population. In response, we developed alternative norms across four age categories (8, 9, 10, 11) that better align with the demographic realities of this group. These norms demonstrate an expected distribution of scores and provide a better benchmark for evaluating intelligence, specifically visuo-spatial reasoning, of Indian children attending government schools. We encourage researchers, educators and clinicians to use these norms as appropriate.
Generated within Cultural Model Theory, we present and discuss the results of a research project about the mental and linguistic representation of quality distinctions in Tongan, a Polynesian language. Previous research had discovered the Tongan foundational cultural model of Radiality—whose major trait is the backgrounding of ego and the foregrounding of other-than-ego—in a number of ontological primes, such as space, time, and relationship (kinship, social and with supernatural). Then, Radiality was suggested as being at the root of Tongan and Polynesian sociocentrism. The results of this new research indicate that the foundational cultural model of Radiality also contributes to the construction of another ontological prime, quality. An innovative methodology, based on the expression of primary and secondary quality distinctions by adjectives, was used to ascertain the contribution of Radiality to this underexplored domain.
Based on previous work reporting differences between monolinguals and bilinguals in cognitive and linguistic processing, this study investigates relations between cognitive and linguistic conflict resolution in L1 German adolescents by exploring participants’ cognitive control ability and the revision of grammatical analysis of (object-initial) wh-questions and relative clauses in German and English. Sixty-eight 12- to 14-year-old learners of English (monolinguals or heritage language bilinguals from birth) were matched on German and English proficiency and socio-economic status. Participants performed a Flanker task measuring non-linguistic conflict resolution and cognitive control as well as two sentence-picture matching experiments measuring linguistic conflict resolution: one in L1 German and one in L2/L3 English. In corroboration with previous work, the Flanker task results yielded a significantly smaller Flanker effect magnitude for bilinguals than monolinguals, which constitutes evidence of enhanced non-linguistic conflict resolution for bilinguals. As for linguistic conflicts, in the German experiment, bilinguals exhibited smaller differences between subject and object word orders than monolinguals, with only monolinguals showing a subject over object processing advantage. In contrast, in the English experiment, the groups performed similarly, and both displayed the subject over object processing advantage. We interpret these findings to indicate that differences in non-linguistic conflict resolution were only mirrored in resolving linguistic conflict in the more proficient and dominant language.
In the digital age, where human experience is increasingly shaped by algorithms, some scholars have appealed to Clark and Chalmers’s Extended Mind hypothesis to account for human–algorithm relations. These accounts suggest that algorithmically generated content functions as an extension of human cognition. This paper argues that while certain forms of online cognition may indeed be extended—either literally, by adhering to the core criteria of Clark and Chalmers’s thesis, or metaphorically, by satisfying only some of them—algorithmic systems violate key principles of the Extended Mind hypothesis. Drawing on post-humanist insights and AI literature, the paper shows that algorithms do not merely support or extend preexisting cognitive processes and beliefs. Rather, they often impose content based on past usage that leads to transformations of preexisting beliefs or to the formation of new ones. By contrasting stable analogue media, such as the notebook in Clark and Chalmers’s canonical example, with the dynamic and adaptive nature of contemporary algorithmic systems, the paper proposes a cognitive spectrum of human-technology relations, ranging from genuine cases of cognitive extension to cases of cognitive entanglement in which algorithmic agency influences the human user to such an extent that it nearly supervenes on their cognition.
As large language models (LLMs) continue to scale, emerging evidence suggests diminishing performance gains from further pretraining on high-quality, high-resource datasets, particularly those dominated by English-language corpora. This Perspective interprets this trend not only as a limitation of data quantity, but as a manifestation of semantic redundancy and epistemic narrowing within prevailing training distributions. We argue that contemporary LLMs increasingly function as proto world-models, implicitly encoding assumptions about meaning, causality, and knowledge that are disproportionately shaped by dominant linguistic and cultural sources. In response, this paper proposes a conceptual reframing of data strategy in LLM development. Rather than persistently extracting value from saturated corpora, we advocate for the systematic inclusion of culturally grounded and underrepresented knowledge systems, including low-resource and Indigenous languages, oral traditions (such as folklore, proverbs, and oral histories), and religious, philosophical, and historical texts. These sources introduce not only novel syntactic and semantic structures, but distinct epistemic frameworks that remain largely absent from conventional datasets. We suggest that such cultural corpora should be understood not merely as instruments of representational diversity, but as carriers of alternative world-models that can expand the epistemic capacity of language models. Their integration offers a potential pathway for mitigating redundancy, reducing reliance on synthetic data feedback loops, and enhancing generalization across diverse semantic domains.
Art museums welcome visitors with varying levels of prior knowledge, making interpretive tools such as labels essential for encouraging engagement and improving understanding. In partnership with the Rijksmuseum in Amsterdam, this study explored how visitors respond to Formal–Descriptive versus newly designed labels accompanying paintings that address the Dutch colonial past. Formal–Descriptive labels were designed to guide the visitors view, while the Contextual–Historical labels aimed to raise historical awareness by focusing on colonial histories and their implications. As visitors have different motivations for coming to the museum, we examined two visitor profiles: “Pleasure Seekers”, who primarily visit for enjoyment and a visual experience, and “Art Lovers”, who tend to seek deeper contextual or historical meaning. Using a mixed-methods approach including eye tracking, interviews, and self-report measures, we examined visual attention, emotional responses, and interpretation across three paintings. Pleasure Seekers showed larger pupil diameters and spent more time reading labels than Art Lovers, possibly reflecting greater cognitive effort. Viewers in the Contextual–Historical label condition spent more time initially engaging with label content and made fewer saccades, indicating more focused attention. Interviews revealed that Art Lovers used more historically specific language. Questionnaire data showed that Formal–Descriptive labels were linked to more negative reactions to sensitive topics. These findings demonstrate that museums can integrate historically sensitive narratives without causing discomfort and underscore the importance of tailoring interpretive texts to diverse audiences to support reflection and inclusivity.
Pragmatic competence, the ability to infer implied meaning, recognize presuppositions, and interpret speech acts, has traditionally been tied to embodied experience and social interaction. Whether large language models (LLMs) can approximate these abilities has become an increasingly pressing question. This study compares the performance of 70 Uzbek undergraduate students and two state-of-the-art LLMs (GPT-4 and Gemini 2.0) across three domains: conversational implicature, presupposition, and speech acts, using a controlled set of 60 stimuli. While the models handled some conventionalized pragmatic cues, they consistently underperformed relative to human participants, particularly in tasks requiring contextual inference, presupposition accommodation, and recognition of indirect communicative intent. Error patterns revealed systematic tendencies toward literal interpretation, accommodation failures, and misidentification of illocutionary force. The results indicate that current LLMs approximate the surface form of pragmatic behavior without reliably engaging the inferential and social mechanisms that underlie it in human communication.
In the ever-evolving realm of artificial intelligence, Chat-generative pre-trained transformer (ChatGPT) stands out as a significant achievement in the advancement of scientific research. This paper aimed to evaluate the potential of ChatGPT in the examination of autobiographical memory through two studies. In Study 1, ChatGPT was presented with a scale designed to measure specificity along with various autobiographical memories. It was then requested to analyze the memories for specificity, and its evaluation was subsequently compared to that of an independent rater. The results of ChatGPT’s analysis were correct, matching the assessment provided by the independent rater by 100
The study examined the relationships between emotion comprehension and language development among 989 children aged 61–80 months (506 boys) from families of middle socio-economic status (SES). Children were asked to create a story based on a series of pictures and were tested on their active vocabulary (verbal fluency test) and with the test of emotion comprehension (TEC). We analyzed the difference between effect of such language skills measures as verbal fluency and narrative skills on emotion comprehension development, and how they were interrelated. Significant correlations between emotion comprehension and language development aspects were analyzed with regression model. The results revealed that age and non-verbal intelligence were significant factors in both language and emotion comprehension development; furthermore, emotion comprehension development is more strongly interrelated with narratives’ production than with active vocabulary volume. The paper concludes with implications for the development of narrative skills and provides fresh avenues for future research that implies the analysis of the theory of mind as an intermediate factor between emotional and language development.
This study explores the current linguistic debates over the generative performance of large language models (LLMs) from both philosophical and empirical perspectives. We argue that the linguistic profiles cannot be constructed for LLMs without considering their linguistic performance, vulnerabilities to jailbreaks and architectural properties. The present review synthesizes the core debates over the abilities of LLMs, which are frequently described as both “stochastic parrots” from a philosophical perspective and, “human-like” from computational linguistic perspectives. We offer an overview of the explorations of LLMs in linguistic studies and the contradictory results regarding their multilingual outputs. Different cognitive and psycholinguistic assessments of LLMs are also examined from the perspective of language generation and jailbreaking. Connections between the successful natural language generation and failures to withstand simple jailbreak prompts are established with reference to the transformer architecture of LLMs.
This paper took the Yichun dialect in Jiangxi Province, China as a case study and used a mixed method to explore the factors that affected the inheritance and use of the Yichun dialect, as well as their relationship with local identity. As one of the endangered dialects in China, the Yichun dialect provides a valuable case for examining how the inheritance and use of dialects are influenced by social development and national language policies. Our findings revealed that family education played a crucial role in helping Gen Z individuals inherit Yichun dialect, while community support positively promoted actual dialect use. In contrast, school education negatively affected dialect proficiency, and migration had a negative impact on dialect use. Besides, dialect proficiency positively influenced both local identity and use behavior. Use behavior in turn, strengthened local identity. This article revealed the unique roles of family education in preserving Yichun dialect and social support in promoting its use, and uncovered the relationships among dialect proficiency, use behavior, and local identity. It further explained the impacts of the Mandarin promotion policy and economic factors on the decline of the dialect, and proposed measures to preserve the declining dialect.
The numeral systems of Indo-Aryan languages such as Hindi, Gujarati, and Bengali are highly unusual in that unlike most numeral systems (e.g., those of English, Chinese, etc.), forms referring to 1–99 are highly non-transparent and cannot be constructed using straightforward rules for forming combinations of tens and digits. Importantly, this non-transparency is not, as in some languages tolerating a mild degree of irregularity (e.g., Spanish catorce ‘14’, cuarenta ‘40’, which are not transparent combinations of cuatro ‘4’ and diez ‘10’), limited to the teens and decade forms. As an example, Hindi/Urdu ikyānve ‘91’ is not decomposable into the composite elements ek ‘one’ and nave ‘ninety’ in the way that its English counterpart is. This paper further clarifies the position of Indo-Aryan languages within the typology of numeral systems, and explores the linguistic and non-linguistic factors that may be responsible for the persistence of complex systems in these languages. Using data from multiple databases, we develop and employ a number of cross-linguistically applicable metrics to quantify the complexity of languages’ numeral systems, and demonstrate that Indo-Aryan languages have decisively more complex numeral systems than the world’s languages as a whole, though individual Indo-Aryan languages differ from each other in terms of the complexity of the patterns they display. We investigate the factors (e.g., religion, geographic isolation, etc.) that underlie complexity in numeral systems, with a focus on South Asia, in an attempt to develop an account of why complex numeral systems developed and persisted in certain Indo-Aryan languages but not elsewhere. Finally, we demonstrate that Indo-Aryan numeral systems adhere to certain general pressures toward efficient communication found cross-linguistically, despite their high complexity. We call for this somewhat overlooked dimension of complexity to be taken seriously when discussing general variation in numeral systems.
This article presents a newly developed instrument for measuring the perceived difficulty of Latin texts. A 13-item questionnaire was constructed to assess how smoothly readers process a given Latin passage. Text difficulty was defined here as the ease of text comprehension. The items were designed in accordance with Kintsch’s model of text comprehension and target distinct cognitive processes and representational levels involved in understanding a text. Additional items addressed translation difficulty more broadly. An exploratory factor analysis (EFA) revealed a two-dimensional structure, with one factor representing text comprehension and the other translation. Only the items measuring comprehension were retained for the final scale. The analysis indicated that most hypothesized processes and levels from Kintsch’s model could be applied to Latin texts. To control for individual differences, additional items were included to assess participants’ background knowledge, especially vocabulary, grammar, and—most importantly—knowledge of author and genre. These personal predictors proved to be highly influential. The final instrument is unidimensional, statistically reliable, and suitable for estimating the overindividual difficulty of Latin texts when administered to a sufficient number of readers. It thus provides an important empirical foundation for future research on Latin readability and the assessment of text difficulty in historical languages more broadly.
We explored language switching (code-switching) in spontaneous inner speech in multilingual native Arabic speakers living in Israel. The aim of this study was to test the effects of both long-term variables (education and proficiency in Hebrew and English as non-native languages) and short-term variables (linguistic context and mood). We created a smartphone application which sent prompts twice a day to 49 female Arabic native speakers for 2 weeks. Participants also filled a questionnaire including questions about their linguistic background and use. Results show that the short-term variable linguistic context affected language use in inner speech: a more mixed linguistic context contributed to more code-switching. Mood however, did not affect code-switching. Hierarchical Regression showed that the long-term variables: education and proficiency in Hebrew, but not English, explained most of the variability in code-switching in inner speech. These findings are discussed in the context of language status in the minds of multilinguals, control and inhibition theories, and the characteristics of inner speech.
Adult language use varies substantially across speakers, with literacy experience emerging as a crucial but understudied factor in creating this variation. While written language exposes speakers to broader, more diverse vocabulary than speech alone, most psycholinguistic research focuses on highly literate populations, leaving gaps in our understanding of how literacy shapes oral production. This study therefore addresses a critical question: Does literacy acquisition affect lexical diversity in spontaneous oral narrative production in Turkish? We compared lexical diversity results between illiterate and literate adult Turkish speakers during a structured storytelling task. Using Root Type-Token Ratio analyses across six parts of speech, we found that literate speakers consistently demonstrated significantly higher lexical diversity than illiterate speakers (d = 1.18–2.08 for most categories). Crucially, this occurred without increased word production, indicating that literacy enhances vocabulary variation rather than quantity. The largest effects emerged for elaborative categories: conjunctions, adverbs, and adjectives. These findings reveal that literacy fundamentally affects lexical organization and deployment in oral productions.
In recent years, global awareness of mental health issues has increased significantly. However, many counseling and support approaches still reflect Western perspectives, which may not resonate with diverse cultural contexts. This paper aims to explore three key areas: (1) the practices of Afrocentric mental health counseling services; (2) the historical context, effectiveness, and future directions of Afrocentric counseling; and (3) its implications for culturally responsive mental health care in Ethiopia. To conduct this analysis, a comprehensive search was performed across six bibliographic databases, including PsycINFO, Cochrane Library, Scopus, Elsevier, and Google Scholar. The inquiry explores counseling, mental health, psychotherapy, spirituality, indigenous Africa, and local mental health services, focusing on Afrocentric approaches, traditional counseling, and African psychology history A critical review of the literature consistently highlights the necessity of incorporating cultural identity, traditional healing methods, and spirituality into therapeutic approaches designed specifically for African and African-descended populations. The findings indicate that culturally adapted counseling techniques such as storytelling, community rituals, drumming, and spiritual practices are essential for enhancing client engagement, fostering empowerment, and improving psychological outcomes. Overall, this review emphasizes the urgent need for culturally sensitive mental health strategies that respect the unique experiences and strengths of African and African-descended populations. The ultimate goal is to reduce disparities and promote mental health equity.
Children’s language understanding has been studied as resulting in a dichotomy, with children either fully capturing meaning or failing to capture it entirely. However, it does not seem to reflect in daily interactions. This study examined children’s good-enough comprehension, this level of comprehension being defined as the formation of an imprecise and shallow yet mostly satisfying representation of the meaning of an utterance, used in communicative settings. Ten-year-old children were presented a test created for this study, in which they had to answer falsified questions under two conditions (speed answering vs precise answering). Mixed model analyses showed that children displayed good-enough comprehension by not noticing the falsification in the speed condition. Further analyses also showed that the time children took to answer falsified questions was significantly longer than non-falsified questions, even when the falsification went unnoticed. The consideration of good-enough comprehension in children dictates a reevaluation of existing literature on children’s comprehension, as good-enough comprehension may precede good comprehension during development.
