Reproduce “REACT: Two Datasets for Analyzing Both Human Reactions and Evaluative Feedback to Robots Over Time” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Kate Candon, Nicholas C. Georgiou, Helen Zhou, Sidney Richardson, Qiping Zhang, Brian Scassellati, Marynel Vázquez
Researchers from Yale University developed REACT, a database comprising two human-robot interaction datasets, REACT-Nao and REACT-Shutter, designed to analyze human reactions and evaluative feedback over time. Analysis of these datasets demonstrated that human expressiveness to robot stimuli significantly changes throughout an interaction, often becoming less pronounced over longer durations.
Key insights
- Human nonverbal expressiveness towards robots is not static but evolves over the course of an interaction.
- Initial strong reactions to a robot can diminish or become muted over longer interaction durations, a phenomenon akin to novelty effects.
- Models for interpreting implicit human feedback in HRI must account for the temporal context and full interaction history for accurate understanding.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.