JuniorAkademie Berlin visits Science of Intelligence: Young minds explore intelligence research

The day began with an introduction to SCIoI’s core question: What is intelligence, and how can we understand it across individuals, groups, and machines? This was followed by an exploration of the science of collective behavior and swarming, where Pawel Romanczuk explained how individuals in a group, whether fish, birds, or robots, coordinate their actions. He showed that in fish schools, faster-moving individuals tend to lead, while others follow, highlighting how simple interaction rules can lead to sophisticated group decisions.

In the labs

Students then explored a series of interactive demos showcasing different types of intelligent behavior:

Individual Behavior: In the Čapek Lab, Oussama Zenkri demonstrated how a robot arm solves a puzzle-like lockbox task inspired by escape rooms. His work, based on design principles from biology, addresses one of the biggest challenges in robotics: robustness. The students learned how researchers build robotic systems that maintain performance even under unexpected changes, by actively connecting perception, planning, and control in ways that mirror biological systems.

Social Behavior: In the Asimov Lab, Anna Lange introduced the students to Pepper, a social robot designed to interact with people in a friendly and engaging way. She explained that in one of her recent studies at SCIoI, Pepper took the role of a tutor, teaching a vocabulary game to participants. Drawing on the results of this study, Anna explained how adaptive guidance behavior in a robot teacher can support self-regulated learning, while simultaneously affecting the emotional experience of learners and the social perception of the robot.

Collective Behavior: SCIoI researcher David Mezey presented the CoBe Arena, an interactive system where fish schools projected on the floor respond in real time to human movement, allowing visitors to “become the predator” and observe how the swarm reorganizes under threat. Students could actively engage with the system, experiencing firsthand how real-world testing complements biological models and deepens our understanding of swarm intelligence.

Interactive session

After lunch, Pawel led an interactive modeling session, where students explored the dynamics of flocking behavior using Complexity Explorables, browser-based simulation tools. Adjusting parameters and observing emergent group motion, the students got hands-on experience with some of the computational tools used in scientific modeling of collective intelligence.

The visit ended with a Q&A session, where the students were able to ask questions about university life, career paths and what it means to do science at the intersection of disciplines.

By opening its doors to the next generation of thinkers, SCIoI continues to invest in education, science communication, and early talent development. “We’re excited to inspire curiosity in young minds,” said Pawel. “These are exactly the kinds of interdisciplinary questions and open-ended explorations that define our work at Science of Intelligence.”

The JuniorAkademie students will present the results of their project, including insights from their SCIoI visit, during a public presentation on August 2nd. We’re proud to have contributed to their journey.