Welcome to the SIM lab. We study social intelligence and behaviour in humans, story characters, and machines. Curious how these elements fit together? See our Projects and Publications pages, or get in touch!
The lab kicked off on May 6, 2025, and is a in some respects a continuation of the Creative Intelligence Lab and the Telling Story Project. Some relevant history was transferred to the current website.
Why are we called ‘SIM’ lab? We study social intelligence in human development and evolution, but also in story characters and non-human intelligent systems, including bots and AI-models.
SIM stands for social intelligence modelling, which is reflected in our research using agent-based and classification models.
But it also stands for stories in the mind, reflecting what social cognition in essence is: a form storytelling in which we (silently or out loud) connect visible elements of the social world to invisible ones, such as beliefs, intentions, desires, and emotions. We work with children aged 4-12 to study the link between storytelling and forms of social cognition known as Theory of Mind or Intentional Reasoning. And we train language/AI models using narrative data to study the link between stories and intelligence in a completely novel way.
We also study language/AI models that collaborate with humans on various tasks. Here, SIM refers to simulation: in order to coordinate actions successfully, both sides have to simulate the others’ perspective.
Theory of Mind (ToM) is a skill that is used in both human and AI cognition to infer the mental states of others, and to use these mental states to predict their behaviour in the world. Experts from all over the world joined an AAAI conference workshop in Philadelphia to discuss the benefits and challenges of using ToM in AI systems. The diversity in the field was represented by 4 keynote speakers, 7 spotlight speakers, and 33 poster presentations during a day full of exchanged ideas.
Ramira van der Meulen joined this event as a spotlight speaker, to highlight the misconceptions that exist around implementing ToM in AI. In addition, she presented a poster on how different models of intelligence influence the learning process and knowledge retention of both individual and collectives of AI agents. Overall, it was an inspiring day with many insights on how ToM could be practically useful in AI applications, and we look forward to potential future sessions.
The two presented works can be found as included extended abstracts in the Workshop Proceedings. A longer version of the work on AI-ToM Misconceptions can be found on ArXiv.
Bron: Volkskrant
“Taal- en cognitiespecialist Max van Duin is als wetenschapper altijd op zoek naar meerdere perspectieven, naar logische verhalen. Dat gevoel het soms toch wint van ratio ontdekte hij toen hij botkanker kreeg: ‘Ik maakte een heel onwetenschappelijke keuze.’
Mensen onderscheiden zich van dieren door hun empathisch vermogen. Maar hoe dat inlevingsvermogen zich precies ontwikkelt is nog altijd niet duidelijk. Vanaf jonge leeftijd kunnen kinderen met anderen meevoelen. ‘Maar het duurt nog een hele tijd voordat ze zich echt in anderen kunnen verplaatsen’, weet Max van Duijn (40). Hij doet aan de Universiteit Leiden onderzoek naar inlevingsvermogen, en hoe dat samenhangt met taal en verhalen. Want via verhalen leren we emoties van anderen ‘lezen’. ‘Een van de dingen die ik met mijn studenten doe is naar goede films kijken. Met de vraag: wat maakt een verhaal krachtig, hoe komt het dat je je erin verplaatst?’ Verhalen blijken een belangrijk instrument om ons inlevingsvermogen te ontwikkelen. ‘Wij hebben als mensen niet alleen onze eigen levenservaring als materiaal. Dankzij verhalen hebben wij een hele bibliotheek aan ervaringen tot onze beschikking.’”
Lees het volledige interview op de website van de Volkskrant hier.
How can AI be used in the professional field of the provinces? Developments in artificial intelligence are moving very quickly. It offers opportunities but also challenges. Experts from all twelve provinces shared their knowledge and experiences today at the first Interprovincial Generative AI day in Zwolle. An initiative from the Province of Limburg, the Province of Overijssel and the Interprovincial Consultation.
Max van Duijn was a keynote speaker together with Eric Postma and Anne Fleur van Veenstra. Max talked about the capabilities of systems such as ChatGPT and Google’s Gemini. Through recent research examples, he discussed whether these systems merely reproduce language structures or actually possess understanding, thinking skills and creativity. Implications for current and future applications of generative AI were discussed interactively with the audience.
Assistant professor Max van Duijn combines cognitive science with Artificial Intelligence (AI). His research is on children’s stories and the question if they have a so-called Theory of Mind. This is the ability to understand the mental state of somebody else and use this information to explain and predict human behavior.
But here’s the twist: Max’s research not only applies to the exploration of storytelling but also serves as a test for AI. With the data collection, he and his team investigate uncharted territory: Does AI possess a mind? Can entities like ChatGPT truly empathize?
Computers don’t byte is a series by the Leiden Institute of Advanced Computer Science (LIACS). Leading computer scientists from a variety of fields share their expertise and insights. Dive into the minds of these researchers and learn about real-world applications, the future of AI and related technologies and cutting-edge research. From chatbots to cybersecurity, from quantum to children’s stories, each episode offers its own perspective on the changing landscape of computer science. Whether you’re a seasoned professional or an aspiring enthusiast, this podcast offers knowledge and inspiration.
AI can do more every day than the day before, but no one can explain exactly how AI does what it does. In the new television program ‘We gaan er allemaal aan’ Valerio Zeno examines various disaster scenarios that could bring down humanity.
In the first episode, he examines the dangers of AI. University lecturers Maarten Lamers and Max van Duijn, both working at the Leiden Institute of Advanced Computer Science (LIACS), both shine their light on AI and its possible dangers. Can AI empathize with humans? Max van Duijn explains it using the Sally-Anne test.
Do self-improving machines pose a threat to humans? The answers to these and other questions are covered in ‘We gaan er allemaal aan’.
Watch the episode back on NPO Start.
Researcher Max van Duijn and PhD student Bram van Dijk apply language models to stories told by children to investigate empathy. For this research, they received the Best Paper Award at the Computational Natural Language Learning Conference in Singapore.
Read the full interview in Dutch here, or in English, here.
Humbled to receive the best paper award @CoNLL@EMNLP in Singapore from programm chairs Jing Jiang and David Reitter for the Children’s Story Corpus we have been working on since 2019 in my Veni project at LIACS. It’s a pity Bram van Dijk wasn’t there as he did the majority of the work. Many, many thanks to the 442 children and their parents and teachers for sharing their stories with us and to all project assistants!! And to NWO (Nederlandse Organisatie voor Wetenschappelijk Onderzoek). Find the paper and the corpus here: https://aclanthology.org/2023.conll-1.23/
sim-lab