Roboticists are continually improving the quality of social robot behaviors and interactions with humans. This is a major goal of the field of social robotics, which seeks to create socially competent robots and improve their overall acceptance. Toward this effort, we propose the utilization of primal world beliefs (i.e., beliefs about the character of the world) to design behaviors that are relatable, intuitive, and based on an internal motivation. However, it is not yet clear whether humans can reliably discern these world beliefs in robots. In this work, we explore whether primals can serve as a novel framework to inform the design of robot personality and attempt to understand whether and how humans perceive primals within robots. Through two large online user studies ( \(n=300\) ; \(n=360\) ), we show that (1) participants are broadly able to discern intended primals in robots, (2) certain participant and robot primals predict participant perception of the robots, and (3) similarity between human and robot primals predicts improved perception of robots.

Siddharth Mangalik, Adithya V Ganesan, Abigail B. Wheeler, Nicholas Kerry, Jeremy D. W. Clifton, H. Schwartz, Ryan L. Boyd. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.