Limor Raviv
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Limor Raviv has a total of 234 co-authors. Here are the top co-authors by number of shared publications:
Antje Meyer
Max Planck Institute for Psycholinguistics, The Netherlands; Radboud University Nijmegen, The Netherlands
4 shared publications- William Tecumseh Fitch
University of Vienna
4 shared publications Andrea Ravignani
3 shared publicationsBrenda Suyanne Barbosa
3 shared publications- Christina M. Vanden Bosch der Nederlanden
Department of Psychological & Brain Sciences, University of Toronto Mississauga , Mississauga, ON
3 shared publications - Fang Liu
School of Psychology and Clinical Language Sciences, University of Reading, Reading, UK
3 shared publications
Latest Publications
- Open Access
- 15 October 2026
- Published by Elsevier in SSRN Electronic Journal
Compositionality in emerging vocal communication can be linear, simultaneous, and iconic
Compositional structure and iconicity are key properties of human language. Compositionality allows efficient communication through the systematic recombination of established units, while iconicity facilitates mutual understanding through form-meaning resemblance. However, despite prominent theories linking their evolution (for instance, high iconicity potentially hindering the emergence of structure), iconicity and compositionality are typically studied independently. Moreover, most experimental work on the emergence of compositionality comes from modalities other than speech (e.g., gesture, written text, drawing), which may differ in their communicative and iconic affordances. Here, we fill in this gap by examining the co-evolution of compositional structure and iconicity in the vocal modality. In a communication game in a 3D virtual environment, pairs of participants invented non-linguistic vocalizations to refer to fantasy creatures varying in shape, size, movement, and speed. Vocalizations were analyzed for iconic form-to-meaning mappings and systematic structure. Results show that participants readily created stable iconic mappings between acoustic and semantic features (e.g., higher pitch for smaller creatures). Compositional structure emerged and increased significantly over time, with vocalizations showing both linear compositionality (smaller, word-like units recombined in a morphological structure) and simultaneous compositionality (iconic mappings recombined within a single vocalization). Beyond providing the first empirical evidence for the emergence of compositionality in the vocal modality, these findings challenge claims that iconicity erodes as structure emerges. Iconicity not only co-evolved with compositionality but also scaffolded it. This suggests that, in emerging communication, iconicity can serve as a resource for building compositional structure.
- 15 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessCompositionality in emerging vocal communication can be linear, simultaneous, and iconic
Compositional structure and iconicity are key properties of human language. Compositionality allows efficient communication through the systematic recombination of established units, while iconicity facilitates mutual understanding through form-meaning resemblance. However, despite prominent theories linking their evolution (for instance, high iconicity potentially hindering the emergence of structure), iconicity and compositionality are typically studied independently. Moreover, most experimental work on the emergence of compositionality comes from modalities other than speech (e.g., gesture, written text, drawing), which may differ in their communicative and iconic affordances. Here, we fill in this gap by examining the co-evolution of compositional structure and iconicity in the vocal modality. In a communication game in a 3D virtual environment, pairs of participants invented non-linguistic vocalizations to refer to fantasy creatures varying in shape, size, movement, and speed. Vocalizations were analyzed for iconic form-to-meaning mappings and systematic structure. Results show that participants readily created stable iconic mappings between acoustic and semantic features (e.g., higher pitch for smaller creatures). Compositional structure emerged and increased significantly over time, with vocalizations showing both linear compositionality (smaller, word-like units recombined in a morphological structure) and simultaneous compositionality (iconic mappings recombined within a single vocalization). Beyond providing the first empirical evidence for the emergence of compositionality in the vocal modality, these findings challenge claims that iconicity erodes as structure emerges. Iconicity not only co-evolved with compositionality but also scaffolded it. This suggests that, in emerging communication, iconicity can serve as a resource for building compositional structure.
- Open Access
- 12 October 2026
- Published by Walter de Gruyter GmbH in Cognitive Linguistics
Investigating the role of pragmatic visual signals in language emergence using communication games and agent-based models
Studying the emergence of language has often relied on communication games and agent-based models, with a strong focus on referential linguistic signals, often overlooking pragmatic signals that are frequently expressed visually, such as facial signals. These signals are a crucial component of natural interaction, supporting repair, coordination, and negotiation of meaning. We outline existing language emergence experiments, highlighting findings from communication games and research on pragmatic facial signals in interaction, demonstrating their relevance to the emergence of communication systems. Furthermore, computational models of language emergence, such as agent-based models, largely overlook that communication is layered across multiple communication channels. We argue that incorporating pragmatic visual signals into these approaches can help to provide a more nuanced understanding of the processes underlying language emergence in natural, multimodal interaction. Following these insights, we propose two complementary directions for future research. First, communication game paradigms can be adapted to systematically analyze visual pragmatic signals, linking them to existing commonly used measures such as convergence, communicative success, and adaptation of linguistic forms. Second, agent-based models can be extended to multi-channel frameworks in which agents produce and interpret both referential and pragmatic signals, enabling investigation of how layered multimodal information shapes language emergence.
- 12 October 2026
- Published by Walter de Gruyter GmbH in Cognitive Linguistics
Open AccessInvestigating the role of pragmatic visual signals in language emergence using communication games and agent-based models
Studying the emergence of language has often relied on communication games and agent-based models, with a strong focus on referential linguistic signals, often overlooking pragmatic signals that are frequently expressed visually, such as facial signals. These signals are a crucial component of natural interaction, supporting repair, coordination, and negotiation of meaning. We outline existing language emergence experiments, highlighting findings from communication games and research on pragmatic facial signals in interaction, demonstrating their relevance to the emergence of communication systems. Furthermore, computational models of language emergence, such as agent-based models, largely overlook that communication is layered across multiple communication channels. We argue that incorporating pragmatic visual signals into these approaches can help to provide a more nuanced understanding of the processes underlying language emergence in natural, multimodal interaction. Following these insights, we propose two complementary directions for future research. First, communication game paradigms can be adapted to systematically analyze visual pragmatic signals, linking them to existing commonly used measures such as convergence, communicative success, and adaptation of linguistic forms. Second, agent-based models can be extended to multi-channel frameworks in which agents produce and interpret both referential and pragmatic signals, enabling investigation of how layered multimodal information shapes language emergence.
- Open Access
- 12 September 2026
- Published by Center for Open Science in PsyArXiv
Synchronised group singing enhances social bonding more than group conversation or recitation does: A Registered Report across 30 languages
The evolution of music, language, and prosociality has been debated since before Darwin. The social bonding hypothesis proposes that these phenomena may be interlinked: musicality may have facilitated the evolution of social bonding beyond the possibilities of spoken language. Although dozens of experimental studies have argued that synchronised rhythms can promote bonding, methodological issues including publication bias, sample bias, experimenter effects, and appropriateness of experimental controls make it unclear whether synchronous singing reliably and generally enhances bonding relative to speaking. Following a peer-reviewed Stage 1 Registered Report protocol, we collected data from randomized controlled in-person experiments with 1,017 participants singing and speaking in groups of 5-10 in 30 diverse languages (N = 901 participants after planned quality control exclusions). Our results strongly supported all three of our pre-registered confirmatory hypotheses (all Bayes Factors > 30): (i) synchronised group singing enhances self-reported social bonding (20.9% mean increase), a significantly greater increase than (ii) group conversation (12.9% increase) or (iii) synchronous recitation (15.2% increase). Unregistered exploratory analyses suggested that perceived synchronisation and enjoyment are key predictors of social bonding, results are not artefacts of expectancy or group size, and singing’s bonding effects may be stronger in non-Western than Western contexts. Group singing may serve as a better bonding activity than speaking activities in settings such as school or workplace orientation events. Our study provides strong cross-cultural empirical evidence that music is not merely “auditory cheesecake” that is “useless” compared with language (Pinker, 1997). All forms of vocal communication increase social bonding, but group singing bonds us together over 1.5 times more on average than spoken conversation does.
- 12 September 2026
- Published by Center for Open Science in PsyArXiv
Open AccessSynchronised group singing enhances social bonding more than group conversation or recitation does: A Registered Report across 30 languages
The evolution of music, language, and prosociality has been debated since before Darwin. The social bonding hypothesis proposes that these phenomena may be interlinked: musicality may have facilitated the evolution of social bonding beyond the possibilities of spoken language. Although dozens of experimental studies have argued that synchronised rhythms can promote bonding, methodological issues including publication bias, sample bias, experimenter effects, and appropriateness of experimental controls make it unclear whether synchronous singing reliably and generally enhances bonding relative to speaking. Following a peer-reviewed Stage 1 Registered Report protocol, we collected data from randomized controlled in-person experiments with 1,017 participants singing and speaking in groups of 5-10 in 30 diverse languages (N = 901 participants after planned quality control exclusions). Our results strongly supported all three of our pre-registered confirmatory hypotheses (all Bayes Factors > 30): (i) synchronised group singing enhances self-reported social bonding (20.9% mean increase), a significantly greater increase than (ii) group conversation (12.9% increase) or (iii) synchronous recitation (15.2% increase). Unregistered exploratory analyses suggested that perceived synchronisation and enjoyment are key predictors of social bonding, results are not artefacts of expectancy or group size, and singing’s bonding effects may be stronger in non-Western than Western contexts. Group singing may serve as a better bonding activity than speaking activities in settings such as school or workplace orientation events. Our study provides strong cross-cultural empirical evidence that music is not merely “auditory cheesecake” that is “useless” compared with language (Pinker, 1997). All forms of vocal communication increase social bonding, but group singing bonds us together over 1.5 times more on average than spoken conversation does.
- 6 July 2026
- Published by Cambridge University Press (CUP) in Behavioral and Brain Sciences
- Vol. 49, e236
Young children do not change language
Language differs from other cultural domains in its high degree of normativity and conventionality; hence, many linguistic innovations by young children are considered acquisition errors that do not diffuse into language communities. Moreover, young children’s cognitive-pragmatic limitations hinder their ability to introduce structure into novel communication systems. Consequently, children are unlikely agents of language change or drivers of language emergence.
- 6 July 2026
- Published by Cambridge University Press (CUP) in Behavioral and Brain Sciences
- Vol. 49, e236
Young children do not change language
Language differs from other cultural domains in its high degree of normativity and conventionality; hence, many linguistic innovations by young children are considered acquisition errors that do not diffuse into language communities. Moreover, young children’s cognitive-pragmatic limitations hinder their ability to introduce structure into novel communication systems. Consequently, children are unlikely agents of language change or drivers of language emergence.
- Top Cited Papers
- 20 November 2025
- Published by American Association for the Advancement of Science (AAAS) in Science
- Vol. 390 (6775) , eadq8303
What enables human language? A biocultural framework
Explaining the origins of language is a key challenge in understanding ourselves as a species. We present an empirical framework that draws on synergies across fields to facilitate robust studies of language evolution. The approach is multifaceted, seeing language emergence as dependent on the convergence of multiple capacities, each with their own evolutionary trajectories. It is explicitly biocultural, recognizing and incorporating the importance of both biological preparedness and cultural transmission as well as interactions between them. We demonstrate this approach through three case studies that examine the evolution of different facets involved in human language (vocal production learning, linguistic structure, and social underpinnings).
- 20 November 2025
- Published by American Association for the Advancement of Science (AAAS) in Science
- Vol. 390 (6775) , eadq8303
Top Cited PapersWhat enables human language? A biocultural framework
Explaining the origins of language is a key challenge in understanding ourselves as a species. We present an empirical framework that draws on synergies across fields to facilitate robust studies of language evolution. The approach is multifaceted, seeing language emergence as dependent on the convergence of multiple capacities, each with their own evolutionary trajectories. It is explicitly biocultural, recognizing and incorporating the importance of both biological preparedness and cultural transmission as well as interactions between them. We demonstrate this approach through three case studies that examine the evolution of different facets involved in human language (vocal production learning, linguistic structure, and social underpinnings).
- 8 September 2025
- Published by American Psychological Association (APA) in Psychological Review
Children are not the main agents of language change.
The long-standing claim that young children are the main agents of language change is often presented as an established fact, and has tacitly guided research in developmental science and evolutionary linguistics. It rests on the assumption that language change arises from language acquisition errors predominantly committed by children. Here, we review whether arguments in support of this idea stand up to logical and empirical scrutiny. We conclude that while children's imperfect learning indeed leads them to produce input-divergent linguistic variants, there is no convincing evidence that it is these child-generated innovations that eventually spread through the language community, nor that language change is mainly driven by constraints and biases operating uniquely in children. By exposing the conceptual and empirical shortcomings of overemphasizing children as the agents of language change, we hope to rebalance the field toward a more nuanced understanding of how individual- and population-level processes shape language change. (PsycInfo Database Record (c) 2025 APA, all rights reserved).
- 8 September 2025
- Published by American Psychological Association (APA) in Psychological Review
Children are not the main agents of language change.
The long-standing claim that young children are the main agents of language change is often presented as an established fact, and has tacitly guided research in developmental science and evolutionary linguistics. It rests on the assumption that language change arises from language acquisition errors predominantly committed by children. Here, we review whether arguments in support of this idea stand up to logical and empirical scrutiny. We conclude that while children's imperfect learning indeed leads them to produce input-divergent linguistic variants, there is no convincing evidence that it is these child-generated innovations that eventually spread through the language community, nor that language change is mainly driven by constraints and biases operating uniquely in children. By exposing the conceptual and empirical shortcomings of overemphasizing children as the agents of language change, we hope to rebalance the field toward a more nuanced understanding of how individual- and population-level processes shape language change. (PsycInfo Database Record (c) 2025 APA, all rights reserved).
- Open Access
- 14 August 2025
- Published by Taylor & Francis in International Journal of Transgender Health
- Vol. 27 (5) , 2461-2486
The inescapable cage of a man and a woman: Experiences of non-binary people with gender identity (non-)affirmation in the context of a gendered language
- 14 August 2025
- Published by Taylor & Francis in International Journal of Transgender Health
- Vol. 27 (5) , 2461-2486
Open AccessThe inescapable cage of a man and a woman: Experiences of non-binary people with gender identity (non-)affirmation in the context of a gendered language
- Open Access
- 21 July 2025
- Published by Elsevier in Computers in Human Behavior Reports
- Vol. 19
No evidence for generational differences in the conventionalisation of face emojis
- 21 July 2025
- Published by Elsevier in Computers in Human Behavior Reports
- Vol. 19
Open AccessNo evidence for generational differences in the conventionalisation of face emojis
- 1 February 2025
- Published by American Psychological Association (APA) in Developmental Psychology
- Vol. 61 (2) , 336-350
A bidirectional association between language development and prosocial behavior in childhood: Evidence from a longitudinal birth cohort in the United Kingdom.
This study investigated a developmental cascade between prosocial and linguistic abilities in a large sample (N = 11,051) from the general youth population in the United Kingdom (50% female, 46% living in disadvantaged neighborhoods, 13% non-White). Cross-lagged panel models showed that verbal ability at age 3 predicted prosociality at age 7, which in turn predicted verbal ability at age 11. Latent growth models also showed that gains in prosociality between 3 and 5 years were associated with increased verbal ability between 5 and 11 years and vice versa. Theory of mind and social competence at age 5 mediated the association between early childhood prosociality and late childhood verbal ability. These results remained robust even after controlling for socioeconomic factors, maternal mental health, parenting microclimate in the home environment, and individual characteristics (sex, ethnicity, and special educational needs). The findings suggest that language skills could be boosted through mentalizing activities and prosocial behaviors. (PsycInfo Database Record (c) 2024 APA, all rights reserved).
- 1 February 2025
- Published by American Psychological Association (APA) in Developmental Psychology
- Vol. 61 (2) , 336-350
A bidirectional association between language development and prosocial behavior in childhood: Evidence from a longitudinal birth cohort in the United Kingdom.
This study investigated a developmental cascade between prosocial and linguistic abilities in a large sample (N = 11,051) from the general youth population in the United Kingdom (50% female, 46% living in disadvantaged neighborhoods, 13% non-White). Cross-lagged panel models showed that verbal ability at age 3 predicted prosociality at age 7, which in turn predicted verbal ability at age 11. Latent growth models also showed that gains in prosociality between 3 and 5 years were associated with increased verbal ability between 5 and 11 years and vice versa. Theory of mind and social competence at age 5 mediated the association between early childhood prosociality and late childhood verbal ability. These results remained robust even after controlling for socioeconomic factors, maternal mental health, parenting microclimate in the home environment, and individual characteristics (sex, ethnicity, and special educational needs). The findings suggest that language skills could be boosted through mentalizing activities and prosocial behaviors. (PsycInfo Database Record (c) 2024 APA, all rights reserved).
- Open Access
- 30 December 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-13
Deep neural networks and humans both benefit from compositional language structure
Deep neural networks drive the success of natural language processing. A fundamental property of language is its compositional structure, allowing humans to systematically produce forms for new meanings. For humans, languages with more compositional and transparent structures are typically easier to learn than those with opaque and irregular structures. However, this learnability advantage has not yet been shown for deep neural networks, limiting their use as models for human language learning. Here, we directly test how neural networks compare to humans in learning and generalizing different languages that vary in their degree of compositional structure. We evaluate the memorization and generalization capabilities of a large language model and recurrent neural networks, and show that both deep neural networks exhibit a learnability advantage for more structured linguistic input: neural networks exposed to more compositional languages show more systematic generalization, greater agreement between different agents, and greater similarity to human learners.
- 30 December 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-13
Open AccessDeep neural networks and humans both benefit from compositional language structure
Deep neural networks drive the success of natural language processing. A fundamental property of language is its compositional structure, allowing humans to systematically produce forms for new meanings. For humans, languages with more compositional and transparent structures are typically easier to learn than those with opaque and irregular structures. However, this learnability advantage has not yet been shown for deep neural networks, limiting their use as models for human language learning. Here, we directly test how neural networks compare to humans in learning and generalizing different languages that vary in their degree of compositional structure. We evaluate the memorization and generalization capabilities of a large language model and recurrent neural networks, and show that both deep neural networks exhibit a learnability advantage for more structured linguistic input: neural networks exposed to more compositional languages show more systematic generalization, greater agreement between different agents, and greater similarity to human learners.