AI Meets Psychology and Systems
This project investigates how advanced AI systems can augment human competence in creativity, learning, and scientific reasoning, combining psychological research on human–AI interaction with algorithmic modeling and multi-agent system design.
IOL Project Members
External Project Members
Project Description
A collaboration between the Weizenbaum Institute, HU Berlin, and ZIB, this project develops and evaluates coordinated, agentic AI architectures that support complex cognitive and educational processes under meaningful human control.
The research spans two pillars: foundational understanding of LLM capabilities for simulating human behavior and social interaction, including frameworks of Human–AI Co-Creativity; and applied system development, producing the Synthetic Divergent Association Task (S-DAT) for automated creativity assessment and FACET, a multi-agent architecture for teacher-centered personalized worksheet generation.
Talks and posters
Conference and workshop talks
- Aug 2026
- FACET: Multi-Agent AI Supporting Teachers in Scaling Differentiated Learning for Diverse Students by Nicolas Leins
IJCAI Conference, Bremen
Poster presentations
- Aug 2026
- FACET: Multi-Agent AI Supporting Teachers in Scaling Differentiated Learning for Diverse Students by Nicolas Leins
IJCAI Conference, Bremen - Apr 2026
- Stable Personas: Dual-Assessment of Temporal Stability in LLM-Based Human Simulation by Jana Gonnermann-Müller and Nicolas Leins
CHI Conference, Barcelona - Apr 2026
- Within-Model Vs Between-Prompt Variability in Large Language Models for Creative Tasks by Jana Gonnermann-Müller and Nicolas Leins
CHI Conference, Barcelona
Publications and preprints
Preprints and reports
- Gonnermann-Müller, J., Haase, J., Leins, N., Kosch, T., and Pokutta, S. (2026). LLM-Based Educational Simulation: Evaluating Temporal Student Persona Stability Across ADHD Profiles.
[arXiv]
[BibTeX]
@misc{2026_GonnermannmuellerEtAl_Llmstudentsimulation_2605-06307, archiveprefix = {arXiv}, eprint = {2605.06307}, arxiv = {arXiv:2605.06307}, primaryclass = {cs.HC}, author = {Gonnermann-Müller, Jana and Haase, Jennifer and Leins, Nicolas and Kosch, Thomas and Pokutta, Sebastian}, title = {LLM-Based Educational Simulation: Evaluating Temporal Student Persona Stability Across ADHD Profiles}, date = {2026-05-07} }
Conference contributions
- Haase, J., Gonnermann-Müller, J., Hanel, P. H. P., Leins, N., Kosch, T., Mendling, J., and Pokutta, S. Within-Model vs Between-Prompt Variability in Large Language Models for Creative Tasks. Advances in Neural Information Processing Systems, 39.
To appear.
Associated with NeurIPS 2026.
[arXiv]
[BibTeX]
@inproceedings{2026_HaaseEtAl_Promptvariability_2601-21339:1, booktitle = {Advances in Neural Information Processing Systems}, series = {Advances in Neural Information Processing Systems}, volume = {39}, publisher = {Curran Associates}, note = {To appear}, archiveprefix = {arXiv}, eprint = {2601.21339}, arxiv = {arXiv:2601.21339}, primaryclass = {cs.AI}, author = {Haase, Jennifer and Gonnermann-Müller, Jana and Hanel, Paul H. P. and Leins, Nicolas and Kosch, Thomas and Mendling, Jan and Pokutta, Sebastian}, title = {Within-Model vs Between-Prompt Variability in Large Language Models for Creative Tasks} } - Gonnermann-Müller, J., Haase, J., Leins, N., Igel, M., Fackeldey, K., and Pokutta, S. (2026). FACET: Multi-Agent AI Supporting Teachers in Scaling Differentiated Learning for Diverse Students. Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, 7165–7173.
Associated with IJCAI 2026.
DOI: 10.24963/ijcai.2026/797
[arXiv]
[BibTeX]
@inproceedings{2026_GonnermannmuellerEtAl_Facetmultiagent_2601-22788, year = {2026}, booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence}, publisher = {International Joint Conferences on Artificial Intelligence Organization}, pages = {7165--7173}, doi = {10.24963/ijcai.2026/797}, archiveprefix = {arXiv}, eprint = {2601.22788}, arxiv = {arXiv:2601.22788}, primaryclass = {cs.HC}, author = {Gonnermann-Müller, Jana and Haase, Jennifer and Leins, Nicolas and Igel, Moritz and Fackeldey, Konstantin and Pokutta, Sebastian}, title = {FACET: Multi-Agent AI Supporting Teachers in Scaling Differentiated Learning for Diverse Students} } - Gonnermann-Müller, J., Haase, J., Leins, N., Kosch, T., and Pokutta, S. (2026). Maintaining Stable Personas? Examining Temporal Stability in LLM-Based Human Simulation. Proceedings of the Extended Abstracts of the 2026 CHI Conference on Human Factors in Computing Systems, 1–6.
Associated with CHI EA 2026.
DOI: 10.1145/3772363.3799334
[arXiv]
[BibTeX]
@inproceedings{2026_GonnermannmuellerEtAl_Stablepersonas_2601-22812, year = {2026}, booktitle = {Proceedings of the Extended Abstracts of the 2026 CHI Conference on Human Factors in Computing Systems}, publisher = {Association for Computing Machinery}, pages = {1--6}, doi = {10.1145/3772363.3799334}, archiveprefix = {arXiv}, eprint = {2601.22812}, arxiv = {arXiv:2601.22812}, primaryclass = {cs.HC}, author = {Gonnermann-Müller, Jana and Haase, Jennifer and Leins, Nicolas and Kosch, Thomas and Pokutta, Sebastian}, title = {Maintaining Stable Personas? Examining Temporal Stability in LLM-Based Human Simulation} } - Haase, J., Gonnermann-Müller, J., Hanel, P. H. P., Leins, N., Kosch, T., Mendling, J., and Pokutta, S. (2026). It’s Not Just the Prompt: Model Choice Dominates LLM Creative Output. Proceedings of the Extended Abstracts of the 2026 CHI Conference on Human Factors in Computing Systems, 1–5.
Associated with CHI EA 2026.
DOI: 10.1145/3772363.3799284
[arXiv]
[BibTeX]
@inproceedings{2026_HaaseEtAl_Promptvariability_2601-21339, year = {2026}, booktitle = {Proceedings of the Extended Abstracts of the 2026 CHI Conference on Human Factors in Computing Systems}, publisher = {Association for Computing Machinery}, pages = {1--5}, doi = {10.1145/3772363.3799284}, archiveprefix = {arXiv}, eprint = {2601.21339}, arxiv = {arXiv:2601.21339}, primaryclass = {cs.AI}, author = {Haase, Jennifer and Gonnermann-Müller, Jana and Hanel, Paul H. P. and Leins, Nicolas and Kosch, Thomas and Mendling, Jan and Pokutta, Sebastian}, title = {It's Not Just the Prompt: Model Choice Dominates LLM Creative Output} } - Haase, J., and Pokutta, S. (2026). Structured Creativity Methods for Multi-Agent LLMs: Brainwriting Outperforms Disney and Double Diamond. Accepted at ICML 2026 Workshop on Human-AI Co-Creativity (GenAICreativity).
[BibTeX]
@misc{2026_HaasePokutta_BrainwritingCreativity, howpublished = {Accepted at ICML 2026 Workshop on Human-AI Co-Creativity (GenAICreativity)}, year = {2026}, author = {Haase, Jennifer and Pokutta, Sebastian}, title = {Structured Creativity Methods for Multi-Agent LLMs: Brainwriting Outperforms Disney and Double Diamond} } - Haase, J., and Pokutta, S. (2026). The Missing Drive: Functional Analogs of Intrinsic Motivation in Large Language Models for Creative Tasks. Accepted at ICML 2026 Workshop on Human-AI Co-Creativity (GenAICreativity).
[URL]
[BibTeX]
@misc{2026_HaasePokutta_MissingDrive, howpublished = {Accepted at ICML 2026 Workshop on Human-AI Co-Creativity (GenAICreativity)}, year = {2026}, url = {https://genaicreativity.org/icml2026/files/9/9_paper.pdf}, author = {Haase, Jennifer and Pokutta, Sebastian}, title = {The Missing Drive: Functional Analogs of Intrinsic Motivation in Large Language Models for Creative Tasks} } - Haase, J., Hanel, P. H. P., and Pokutta, S. (2025). S-DAT: A Multilingual, GenAI-Driven Framework for Automated Divergent Thinking Assessment. Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society, 8(2), 1194–1205.
Associated with AIES 2025.
[URL]
[arXiv]
[BibTeX]
@article{2025_HaasePPokutta_SdatDivergentthinking:1, year = {2025}, journal = {Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society}, volume = {8}, number = {2}, month = jul, pages = {1194--1205}, url = {https://arxiv.org/pdf/2505.09068}, archiveprefix = {arXiv}, eprint = {2505.09068}, arxiv = {arXiv:2505.09068}, primaryclass = {cs.CL}, author = {Haase, Jennifer and Hanel, Paul H. P. and Pokutta, Sebastian}, title = {S-DAT: A Multilingual, GenAI-Driven Framework for Automated Divergent Thinking Assessment} }
Articles, chapters and books
- Haase, J., and Pokutta, S. (2026). Human–AI Cocreativity: Exploring synergies across levels of creative collaboration. In Generative Artificial Intelligence and Creativity (pp. 205–221). Elsevier.
DOI: 10.1016/B978-0-443-34073-4.00009-5
[arXiv]
[BibTeX]
@incollection{2024_Pokutta_HumanAICOcreative, year = {2026}, booktitle = {Generative Artificial Intelligence and Creativity}, publisher = {Elsevier}, chapter = {16}, pages = {205--221}, doi = {10.1016/B978-0-443-34073-4.00009-5}, archiveprefix = {arXiv}, eprint = {2411.12527}, arxiv = {arXiv:2411.12527}, primaryclass = {cs.HC}, author = {Haase, Jennifer and Pokutta, Sebastian}, title = {Human–AI Cocreativity: Exploring synergies across levels of creative collaboration} } - Haase, J., and Pokutta, S. (2026). Beyond Static Responses: Multi-Agent LLM Systems as a New Paradigm for Social Science Research. Humanities and Social Sciences Communications, 13, 1490.
DOI: 10.1057/s41599-026-08832-2
[arXiv]
[BibTeX]
@article{2025_HaasePokutta_Multiagentllm_2506-01839, year = {2026}, journal = {Humanities and Social Sciences Communications}, volume = {13}, pages = {1490}, doi = {10.1057/s41599-026-08832-2}, archiveprefix = {arXiv}, eprint = {2506.01839}, arxiv = {arXiv:2506.01839}, primaryclass = {cs.MA}, author = {Haase, Jennifer and Pokutta, Sebastian}, title = {Beyond Static Responses: Multi-Agent LLM Systems as a New Paradigm for Social Science Research} } - Gonnermann-Müller, J., Haase, J., Fackeldey, K., and Pokutta, S. (2026). Teacher-centered LLM-based multi-agent systems–towards differentiated educational worksheets. Innovation: The European Journal of Social Science Research.
DOI: 10.1080/13511610.2026.2727110
[arXiv]
[BibTeX]
@article{2025_JanaHaaseFackeldeyPokutta_Facetllm_2508-11401, year = {2026}, journal = {Innovation: The European Journal of Social Science Research}, doi = {10.1080/13511610.2026.2727110}, archiveprefix = {arXiv}, eprint = {2508.11401}, arxiv = {arXiv:2508.11401}, primaryclass = {cs.HC}, author = {Gonnermann-Müller, Jana and Haase, Jennifer and Fackeldey, Konstantin and Pokutta, Sebastian}, title = {Teacher-centered LLM-based multi-agent systems–towards differentiated educational worksheets} } - Haase, J., Hanel, P. H. P., and Pokutta, S. (2025). Has the Creativity of Large-Language Models peaked? An analysis of inter- and intra-LLM variability. Journal of Creativity, 35(3), 100113.
DOI: 10.1016/j.yjoc.2025.100113
[URL]
[arXiv]
[BibTeX]
@article{2025_HaasePPokutta_Llmvariability, year = {2025}, journal = {Journal of Creativity}, month = dec, volume = {35}, number = {3}, pages = {100113}, doi = {10.1016/j.yjoc.2025.100113}, url = {https://sciencedirect.com/science/article/pii/S2713374525000202}, archiveprefix = {arXiv}, eprint = {2504.12320}, arxiv = {arXiv:2504.12320}, primaryclass = {cs.CL}, author = {Haase, Jennifer and Hanel, Paul H. P. and Pokutta, Sebastian}, title = {Has the Creativity of Large-Language Models peaked? An analysis of inter- and intra-LLM variability} } - Haase, J., Hanel, P. H. P., and Pokutta, S. (2025). S-DAT: A Multilingual, GenAI-Driven Framework for Automated Divergent Thinking Assessment. Proceedings of the AAAI/ACM Conference on AI Ethics and Society, 8(2), 1194–1205.
DOI: 10.1609/aies.v8i2.36622
[arXiv]
[BibTeX]
@article{2025_HaasePPokutta_SdatDivergentthinking, year = {2025}, journal = {Proceedings of the AAAI/ACM Conference on AI Ethics and Society}, volume = {8}, number = {2}, pages = {1194--1205}, doi = {10.1609/aies.v8i2.36622}, archiveprefix = {arXiv}, eprint = {2505.09068}, arxiv = {arXiv:2505.09068}, primaryclass = {cs.CL}, author = {Haase, Jennifer and Hanel, Paul H. P. and Pokutta, Sebastian}, title = {S-DAT: A Multilingual, GenAI-Driven Framework for Automated Divergent Thinking Assessment} }








