Hi! I am a postdoctoral scholar at MIT's Brain and Cognitive Sciences Department where I work in the Computational Cognitive Science Lab supervized by Josh Tenenbaum. I obtained my PhD from the same department working in the Computational Psycholinguistics Lab with Roger Levy.

Before coming to MIT, I obtained a MSc in Cognitive Science from the University of Vienna, worked in the VisionLab at Central European University in Budapest, and at the Center for Adaptive Behavior and Cognition (ABC) at the Max Planck Institute for Human Development in Berlin.

My primary research focus is human communication. I am particularly fascinated by how communication systems, such as language, develop and evolve over time, and how they interact with the diverse cognitive, social, and cultural aspects that define the human experience. To explore these questions, I employ a range of tools from experimental psychology, linguistics, as well as machine learning and probabilistic modeling.

During my Phd I co-founded and co-orgnized the MIT BCS Philosopy Circle. Besides research, I am passionate about language learning, travel and photography, and you can occasionally find me in the ceramics studio or making electronic music.



  • July 2024: I will be attending CogSci 2024 in Rotterdam.
  • May 2024: I will be presenting PhD-related work at EVOLANG XV in Madison.
  • July 2023: I presented PhD-related work at CogSci 2023 in Sydney.
  • February 2023: I started my postdoctoral research at MIT.
  • December 2022: I defended my PhD dissertation, 'The Cultural Emergence of Combinatorial Structure'


(A more up-to-date list can be found on Google Scholar.)
  • Finding structure in logographic writing with library learning. G Jiang, M Hofer, J Mao, L Wong, JB Tenenbaum, RP Levy (2024) arXiv preprint arXiv:2405.06906
  • Simplicity and Informativeness in the Evolution of Combinatorial Structure. M Hofer, S Kirby, RP Levy (2023) Proceedings of the Annual Meeting of the Cognitive Science Society 45
  • Learning evolved combinatorial symbols with a neuro-symbolic generative model. M Hofer, TA Le, RP Levy, JB Tenenbaum (2021) arXiv preprint arXiv:2104.08274
  • Hierarchical Inferences Support Systematicity in the Lexicon. M Hofer, T Verhoef, RP Levy (2020) Proceedings of the 42nd Annual Meeting of the Cognitive Science Society
  • Linguistic overhypotheses in category learning: explaining the label advantage effect. A Ivanova, M Hofer (2020) Proceedings of the 42nd Annual Meeting of the Cognitive Science Society
  • Emotion, entropy evaluations and subjective uncertainty. L Bertram, E Schulz, M Hofer, JD Nelson (2020) Proceedings of the 42nd Annual Meeting of the Cognitive Science Society
  • The Paradox of Help Seeking in the Entropy Mastermind Game. N Taylor, M Hofer, JD Nelson (2020) Frontiers in Education, pp. 533998
  • Exploring the space of human exploration using Entropy Mastermind. E Schulz, L Bertram, M Hofer, JD Nelson (2019) Proceedings of the 41st Annual Meeting of the Cognitive Science Society
  • Iconicity and Structure in the Emergence of Combinatoriality. M Hofer, RP Levy (2019) Proceedings of the 41st Annual Meeting of the Cognitive Science Society
  • Comparing Models of Associative Meaning: An Empirical Investigation of Reference in Simple Language Games. JH Shen, M Hofer, B Felbo, RP Levy (2018) Proceedings of the 22nd Conference on Computational Natural Language Learning
  • Modeling Sources of Uncertainty in Spoken Word Learning. M Hofer, RP Levy (2017) Proceedings of the 39th Annual Meeting of the Cognitive Science Society


The emergence of combinatorial structure in cultural symbol systems

The ability to learn and culturally transmit symbol systems is central to human intelligence. From phonemes in spoken words, to notes in a melody or the lines of a diagram, these systems rely on combinatorial structure (a finite set of building blocks that can be rearranged to compose new concepts). Some of the aspects enabling combinatorial behavior are shared with other animal species but humans put these abilities to unique uses, going far beyond the kinds of systems we see in non-human animals.

How do combinatorial symbol systems emerge? To answer this question, I am developing computational models that enable us to characterize combinatorial structure, assess whether these systems are structured efficiently, and understand their evolutionary dynamics.

Active learning and psychological models of uncertainty and information acquisition

Another theme of earlier, but still ongoing, work is is trying to understand how humans ask questions and gather information using formal methods from optimal experimental design and Bayesian statistics. One of the tasks we use to study this is a probabilistic version of the popular code breaking game Mastermind, where the subject must guess codes that are generated from distributions over a set of items.