In all species, parents transmit essential skills to their children through communication. Researchers from the University Hospital Bonn (UKB) and the University of Bonn have shown that effective communication depends on how the sender and receiver represent information. Their study reveals how this process underlies effective learning and task performance. Their results have been published in the journal Nature Communications.
Communication, whether through sounds, smells or movements, is essential for survival. Its social aspect is fundamental to cognition, because the descriptions of our tasks in the brain are shaped not only by sensory experiences, but also by the information communicated to us.
“We know from our everyday life that social communication is essential for our learning abilities in the real world, which is summed up by the saying ‘teaching is learning twice’,” says Professor Tatjana Tchumatchenko from the UKB Institute for Experimental Research in Epileptology and Cognition and member of the Transdisciplinary Research Area (TRA) ‘Modeling’ at the University of Bonn.
In a novel study, the Bonn researchers used artificial networks as agents that acted as teachers and students. The teacher network learned to solve a maze and then guided the student network through the task by transmitting a message. This setup allowed the researchers to study how linguistic-like communication between artificial agents improves learning and task performance.
The brain creates abstractions so that our real world can be shared
The results showed that both roles can develop language that allows the student to learn from the teacher. Interestingly, this language is influenced by both the task at hand and the learner’s performance.
“What we have discovered is consistent with what is known about language formation in animals,” says Carlos Wert-Carvajal, co-corresponding author and doctoral student at the University of Bonn in Professor Tchumatchenko’s research group at UKB. He emphasizes that the way our brain encodes our world is not only determined by our own experiences, but also creates abstractions that are understandable to others:
“For example, we don’t say ‘red or green, round, crisp, sweet fruit,’ but we do use the word ‘apple.’ That word exists because our language evolved to represent a shared experience that provides a pleasant reward,” Wert-Carvajal explains. In other words, every language must describe the world as effectively as possible.
This efficiency implied a concise message containing as much information as possible. A good language had to combine both the internal descriptions of the task by the teacher and the learner and the actual features of the real world.
“When we gave feedback on how the student performed the task, the teacher changed his language to convey more useful information,” says first author Tobias Wieczorek, who until recently was a master’s student at the University of Bonn in the Tchumatchenko group at UKB.
This process shows that effective communication is a two-way process. “The sender and the receiver must work together to ensure that the information exchanged is clear, accurate and truly useful,” says Professor Tchumatchenko, who led the study.
Language closes the circle of communication as a shared experience
Remarkably, by “closing the loop”—that is, by reflecting the learner’s language back to the learner—the Bonn researchers enabled learners to teach each other. Despite the lack of explicit teaching skills, the agents effectively communicated essential information and demonstrated the robustness of the language they had developed.
“Although they did not know how to ‘teach’, they were nevertheless able to use their language to convey important information,” says co-corresponding author Dr Maximilian Eggl, who until recently was a postdoctoral fellow at the University of Bonn in Professor Tchumatchenko’s research group at UKB.
This research highlights the fundamental role of language-like communication as a shared cognitive experience and demonstrates its critical importance for learning and generalization. The results provide valuable insights into the design of biological and artificial communication systems that optimize learning and task performance in different environments.
More information:
Tobias J. Wieczorek et al, A Framework for the Emergence and Analysis of Language in Social Learning Agents, Nature Communications (2024). DOI: 10.1038/s41467-024-51887-5
Provided by University Hospital Bonn
Quote:Language-like communication improves learning in artificial networks, study finds (2024, September 5) retrieved September 6, 2024 from
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without written permission. The content is provided for informational purposes only.