Core Concepts
Enhancing audience engagement through dynamic personas in interactive robots.
Abstract
This paper introduces the Masquerading Animated Social Kinematics (MASK) system, which utilizes anthropomorphic robots to interact with guests using non-verbal cues. The system is designed to enhance audience engagement by providing a more immersive and interactive experience. By leveraging persona-driven dialog agents, the MASK framework integrates perception and behavior selection engines to enable real-time interactions with minimal human intervention. The study focuses on whether users can recognize intended characters in film-character-based persona conditions. The research aims to explore the role of personas in interactive agents and factors for creating engaging user experiences.
Stats
Throughout the user subject studies, we examined whether the users could recognize the intended character in film-character-based persona conditions.
We define 67% accuracy as successful in classifying the correct persona, which is four times higher than a random guess 16.7%.
We recruited 108 participants (gender: M=47, F=63, ages: 20-53) to participate in this between-subjects study where the personas were randomly assigned to participants with equal numbers.
Quotes
"We hypothesize that by embedding a character template into robotic behavior, interactive agents can convincingly embody distinct personas for users to engage with."
"Our main question is whether our system enabled users to distinguish robot’s personas while considering the effects of persona-based interactions on user experience and engagement."
"The confusion matrix for classification is shown in Figure 3, where we have adjusted the user’s fitting score obtained from the survey question S6."