Although the mapping between sound and meaning in human language is assumed to be largely arbitrary, research in cognitive science has shown that there are non-trivial correlations between particular sounds and meanings across languages and demographic groups, a phenomenon known as sound symbolism. Among the many dimensions of meaning, sound symbolism is particularly salient and welldemonstrated with regards to cross-modal associations between language and the visual domain. In this work, we address the question of whether sound symbolism is reflected in vision-and-language models such as CLIP and Stable Diffusion. Using zero-shot knowledge probing to investigate the inherent knowledge of these models, we find strong evidence that they do show this pattern, paralleling the well-known kiki-bouba effect in psycholinguistics. Our work provides a novel method for demonstrating sound symbolism and understanding its nature using computational tools. Our code will be made publicly available1.

That which we call a rose by any other name would smell as sweet."

Although the mapping between sound and meaning in human language is assumed to be largely arbitrary, research in cognitive science has shown that there are non-trivial correlations between particular sounds and meanings across languages and demographic groups, a phenomenon known as sound symbolism. Among the many dimensions of meaning, sound symbolism is particularly salient and well-demonstrated with regards to cross-modal associations between language and the visual domain. In this work, we address the question of whether sound symbolism is reflected in vision-and-language models such as CLIP and Stable Diffusion. Using zero-shot knowledge probing to investigate the inherent knowledge of these models, we find strong evidence that they do show this pattern, paralleling the well-known kiki-bouba effect in psycholinguistics. Our work provides a novel method for demonstrating sound symbolism and understanding its nature using computational tools. Our code will be made publicly available.

Advances in artificial intelligence (AI), particularly large language models (LLMs) [1], have achieved remarkable success. This progress stems from "scaling laws" -- performance improves with greater computation, data, and model size [2]. They now excel at mathematics, medical, legal, and coding exams and competitions. Y et, this paradigm has a crucial vulnerability: scaling laws are effective only when data is abundant. Human reasoning, which relies on logical operations and abstractions rather than brute-force pattern matching on vast data, proves critical in tackling complex frontier domains, where usable data is often inherently scarce.

Large Language Models (LLMs) exhibit strong linguistic capabilities, but little is known about how they encode psycholinguistic knowledge across languages. We investigate whether and how LLMs exhibit human-like psycholinguistic responses under different linguistic identities using two tasks: sound symbolism and word valence. We evaluate two models, Llama-3.3-70B-Instruct and Qwen2.5-72B-Instruct, under monolingual and bilingual prompting in English, Dutch, and Chinese. Behaviorally, both models adjust their outputs based on prompted language identity, with Qwen showing greater sensitivity and sharper distinctions between Dutch and Chinese. Probing analysis reveals that psycholinguistic signals become more decodable in deeper layers, with Chinese prompts yielding stronger and more stable valence representations than Dutch. Our results demonstrate that language identity conditions both output behavior and internal representations in LLMs, providing new insights into their application as models of cross-linguistic cognition.

Hallucinations in large language models (LLMs) present a growing challenge across real-world applications, from healthcare to law, where factual reliability is essential. Despite advances in alignment and instruction tuning, LLMs can still generate outputs that are fluent yet fundamentally untrue. Understanding the cognitive dynamics that underlie these hallucinations remains an open problem. In this study, we propose a prompt-based framework to systematically trigger and quantify hallucination: a Hallucination-Inducing Prompt (HIP), which synthetically fuses semantically distant concepts (e.g., periodic table of elements and tarot divination) in a misleading way, and a Hallucination Quantifying Prompt (HQP), which scores the plausibility, confidence, and coherence of the output. Controlled experiments across multiple LLMs revealed that HIPs consistently produced less coherent and more hallucinated responses than their null-fusion controls. These effects varied across models, with reasoning-oriented LLMs showing distinct profiles from general-purpose ones. Our framework provides a reproducible testbed for studying hallucination vulnerability, and opens the door to developing safer, more introspective LLMs that can detect and self-regulate the onset of conceptual instability.

Although the mapping between sound and meaning in human language is assumed to be largely arbitrary, research in cognitive science has shown that there are non-trivial correlations between particular sounds and meanings across languages and demographic groups, a phenomenon known as sound symbolism. Among the many dimensions of meaning, sound symbolism is particularly salient and well-demonstrated with regards to cross-modal associations between language and the visual domain. In this work, we address the question of whether sound symbolism is reflected in vision-and-language models such as CLIP and Stable Diffusion. Using zero-shot knowledge probing to investigate the inherent knowledge of these models, we find strong evidence that they do show this pattern, paralleling the well-known kiki-bouba effect in psycholinguistics. Our work provides a novel method for demonstrating sound symbolism and understanding its nature using computational tools.

Recently, Large Language Models (LLMs) and Vision Language Models (VLMs) have demonstrated aptitude as potential substitutes for human participants in experiments testing psycholinguistic phenomena. However, an understudied question is to what extent models that only have access to vision and text modalities are able to implicitly understand sound-based phenomena via abstract reasoning from orthography and imagery alone. To investigate this, we analyse the ability of VLMs and LLMs to demonstrate sound symbolism (i.e., to recognise a non-arbitrary link between sounds and concepts) as well as their ability to "hear" via the interplay of the language and vision modules of open and closed-source multimodal models. We perform multiple experiments, including replicating the classic Kiki-Bouba and Mil-Mal shape and magnitude symbolism tasks and comparing human judgements of linguistic iconicity with that of LLMs. Our results show that VLMs demonstrate varying levels of agreement with human labels, Figure 1: Illustration of the 3 main experiments we and more task information may be required perform. Firstly, Shape Symbolism is a binary choice for VLMs versus their human counterparts for between two pseudowords to best describe an object that is in silico experimentation. We additionally see spiky or rounded. Magnitude Symbolism involves a binary through higher maximum agreement levels that choice between two pseudowords to best describe an object Magnitude Symbolism is an easier pattern for that is small or large. Finally, Iconicity involves rating VLMs to identify than Shape Symbolism, and the perceived iconicity of words, or how much their written/phonetic that an understanding of linguistic iconicity is form is representative of what they describe.

Audio-visual pre-trained models have gained substantial attention recently and demonstrated superior performance on various audio-visual tasks. This study investigates whether pre-trained audio-visual models demonstrate non-arbitrary associations between sounds and visual representations$\unicode{x2013}$known as sound symbolism$\unicode{x2013}$which is also observed in humans. We developed a specialized dataset with synthesized images and audio samples and assessed these models using a non-parametric approach in a zero-shot setting. Our findings reveal a significant correlation between the models' outputs and established patterns of sound symbolism, particularly in models trained on speech data. These results suggest that such models can capture sound-meaning connections akin to human language processing, providing insights into both cognitive architectures and machine learning strategies.

Iconography and iconology are fundamental domains when it comes to understanding artifacts of cultural heritage. Iconography deals with the study and interpretation of visual elements depicted in artifacts and their symbolism, while iconology delves deeper, exploring the underlying cultural and historical meanings. Despite the advances in representing cultural heritage with Linked Open Data (LOD), recent studies show persistent gaps in the representation of iconographic and iconological statements in current knowledge graphs (KGs). To address them, this paper presents IICONGRAPH, a KG that was created by refining and extending the iconographic and iconological statements of ArCo (the Italian KG of cultural heritage) and Wikidata. The development of IICONGRAPH was also driven by a series of requirements emerging from research case studies that were unattainable in the non-reengineered versions of the KGs. The evaluation results demonstrate that IICONGRAPH not only outperforms ArCo and Wikidata through domain-specific assessments from the literature but also serves as a robust platform for addressing the formulated research questions. IICONGRAPH is released and documented in accordance with the FAIR principles to guarantee the resource's reusability. The algorithms used to create it and assess the research questions have also been made available to ensure transparency and reproducibility. While future work focuses on ingesting more data into the KG, and on implementing it as a backbone of LLM-based question answering systems, the current version of IICONGRAPH still emerges as a valuable asset, contributing to the evolving landscape of cultural heritage representation within Knowledge Graphs, the Semantic Web, and beyond.