Dynamic Modulation of Emotional Expressions in Social Robots: Effects on Liveliness and Naturalness
Haeun Park, Sun Jun Hwang, Hyojin Kim, Jiyeon Lee, Hui Sung Lee
AI summary
Problem
Social robots often lack dynamic modulation of expression intensity, leaving a gap in understanding how varying emotional energy affects human perception of liveliness and naturalness across different emotions.
Approach
The authors implemented a second-order dynamics model to systematically vary the damping ratio and overshoot of a robot’s facial, motion, and sound expressions across five intensity levels, then evaluated human perceptions in a controlled user study.
Key results
- Surprise liveliness ratings increased linearly with expression intensity
- Other emotions peaked at intermediate intensity levels before naturalness declined
- Excessive exaggeration reduces perceived naturalness for non-surprise emotions
- Emotion-specific and user-specific calibration of dynamic intensity is essential for optimal HRI
Why it matters
These findings guide HRI designers and researchers in calibrating robot expression dynamics to maximize engagement without sacrificing perceived naturalness.
Abstract
Humans naturally express emotions with subtle variations, and exaggerated expressions often appear as height- ened intensity in facial, bodily, or vocal cues. This paper intro- duces a method for exaggerating robotic emotional expressions by dynamically adjusting intensity within an emotion dynamics model. By systematically manipulating the damping ratio, we generated five distinct intensity levels for each emotion, thereby producing emotional expressions that exhibited different de- grees of overshoot. A user study revealed that liveliness ratings for surprise increased linearly with intensity, suggesting that exaggerated, high-energy dynamics are particularly effective for conveying surprise. In contrast, other emotions exhibited optimal points at intermediate levels, indicating that excessive exaggeration can reduce perceived naturalness. These findings highlight the need for emotion-specific and user-specific cali- bration of expression intensity, supporting more nuanced and engaging human-robot interactions.