Social responsiveness and its effects on learning in human-human and human-robot interaction

Research in educational psychology describes social responsiveness as one of the central prerequisites for effective knowledge transfer (Aldrup et al., 2020). Understanding and responding appropriately to nonverbal social cues facilitates exchanges between interaction partners and thus helps reduce the complexity of communication. However, our understanding of social responsive behaviors in learning situations is mostly limited to some broad categories.

In this context, the project aims to develop socially responsive artificial agents that can be deployed in learning situations. A first subgoal is to analyze classroom videos using machine learning methods to identify key components of nonverbal social responsiveness in teachers' behaviors. A second subgoal is the transfer of the identified behaviors to artificially intelligent systems, thus enabling, for example, teaching robots to perceive, identify, and respond to social cues in teacher-student interactions.

The development of a synthetic equivalent of social responsiveness will not only help to improve the context sensitivity of synthetic agents, but will also provide insights into the principles of social interaction between humans.

Some of our research questions are:

• Going beyond our existing models of social responsive teaching behaviors, which unknown facets of social responsive behaviors can be identified in learning-related interactions between agents (human-human; human-robot)?
• Which specific social responsive behaviors are needed to ensure effective knowledge transfer between agents (human-human; human-robot)?

Funding and cooperation

The DFG-funded research project is part of the Cluster of Excellence “Science of Intelligence” (SCIoI: www.scienceofintelligence.de) at the Technische Universität Berlin and Humboldt-Universität zu Berlin in collaboration with the University of Potsdam, the Freie Universität Berlin, the Charité – Universtätsmedizin Berlin and the Max Planck Institute for Human Development.

The project is carried out in cooperation between the Department of Educational Sciences at the University of Potsdam (Prof. Dr. Rebecca Lazarides) and the Department of Computer Vision & Remote Sensing at the Technische Universität Berlin (Prof. Dr. Olaf Hellwich).

More information on the project can be found here:
https://www.scienceofintelligence.de/research/researchprojects/project_31/

Information for teachers

Please contact us: jonas.frenkel@uni-potsdam.de

Publications

Publications from our project will be posted here soon!

Contact

Please direct any questions about the content of the research project to
Jonas Frenkel: (jonas.frenkeluuni-potsdampde)