Humanoid facial expressions as a tool to study human behaviour.

Area of Science:

• Robotics
• Human-Robot Interaction
• Social Psychology

Background:

• Facial expressions are key social cues for inferring emotional states.
• Action vitality forms also convey social and affective information.
• Robots can serve as interactive agents to study human social perception.

Purpose of the Study:

• To investigate if robot facial expressions modulate human motor behavior when paired with different action vitality forms.
• To explore the interaction between visual social cues (facial expression) and action characteristics in human-robot interaction.
• To assess how a robot's affective display influences human responses in a physical task.

Main Methods:

• A kinematic experiment involving 18 healthy participants observing video clips of the iCub robot.
• Participants observed the iCub robot performing actions (rude/gentle request) with varying facial expressions (happy/angry).
• Participants then performed a grasping and object-passing task towards the robot, with their movements recorded.

Main Results:

• iCub's facial expressions significantly modulated participants' motor responses.
• A happy expression with a rude action decreased movement velocity, acceleration, and height.
• An angry expression with a gentle action increased these kinematic parameters.
• A behavioral study confirmed that facial expressions altered the perception of action vitality forms.

Conclusions:

• Robot facial expressions play a crucial role in modulating human motor behavior and perception during interaction.
• The combination of facial affect and action characteristics influences kinematic responses and social perception.
• Findings highlight the importance of affective displays in designing socially interactive robots.