The Mind Brain and Behavior Research Center (CIMCYC) announces the return of its CIMCYC Sessions series for the 2025/2026 academic year. The first session of the "AI Mind and Brain" season will take place next Thursday, November 13, 2025.
These sessions aim to foster critical discussion and the exchange of knowledge on cutting-edge research. The methodology is based on the analysis and debate of a scientific article, encouraging participants to prepare by reflecting on questions and doubts for a collective discussion.
The inaugural session will focus on:
Paper: Zirui Chen, Michael F. Bonner, Universal dimensions of visual representation. Sci. Adv. 11, eadw7697 (2025). DOI: 10.1126/sciadv.adw7697
Podcast: bit.ly/3LITMVC
Kickstarting questions:
- The authors propose that a small number of high-level visual dimensions, like animacy, shape, or object function, are shared across human brains, deep neural networks, and behavioral similarity judgments. What kind of evidence supports the existence of such “universal dimensions,” and what might this imply about the structure of visual representations in the brain?
- The paper finds that models trained to align with behavioral similarity judgments are often more predictive of brain representations than raw DNN features. How should this affect the way we think about model–brain alignment? Should behavior become a primary benchmark in evaluating model similarity to human perception?
The series is open to the entire university and research community, as well as external professionals. No prior experience in Artificial Intelligence is required, only curiosity and a willingness to share ideas.
Date and Time: Thursday, November 13, 2025, at 12:30 PM.
Format: The session will be conducted entirely online and will be held in English.
Access: meet.google.com/sgz-quvr-rma
Contact: Please reach out to @email before the meeting in case you need confirmation for your attendance.