REACT: Two Datasets for Analyzing Both Human Reactions and Evaluative Feedback to Robots Over Time

TL;DR

This paper introduces the REACT datasets, capturing human reactions and feedback to robots over time in different scenarios, highlighting the importance of interaction history in understanding human responses.

Contribution

The paper provides two new datasets of human-robot interactions and analyzes the impact of interaction history on human reactions, facilitating future research in implicit feedback interpretation.

Findings

Interaction history influences human reactions to robots.

The datasets include natural reactions during collaborative and photography scenarios.

Future models should incorporate interaction history for better understanding.

Abstract

Recent work in Human-Robot Interaction (HRI) has shown that robots can leverage implicit communicative signals from users to understand how they are being perceived during interactions. For example, these signals can be gaze patterns, facial expressions, or body motions that reflect internal human states. To facilitate future research in this direction, we contribute the REACT database, a collection of two datasets of human-robot interactions that display users' natural reactions to robots during a collaborative game and a photography scenario. Further, we analyze the datasets to show that interaction history is an important factor that can influence human reactions to robots. As a result, we believe that future models for interpreting implicit feedback in HRI should explicitly account for this history. REACT opens up doors to this possibility in the future.