Evaluating human perceptions of android robot facial expressions based on variations in instruction styles

Ayaka Fujii, Carlos Toshinori Ishi, Kurima Sakai, Tomo Funayama, Ritsuko Iwai, Yusuke Takahashi, Takatsune Kumada, Takashi Minato · 2025

Read Evaluating human perceptions of android robot facial expressions based on variations in instruction styles by Ayaka Fujii, Carlos Toshinori Ishi, Kurima Sakai, Tomo Funayama, Ritsuko Iwai, Yusuke Takahashi, Takatsune Kumada, Takashi Minato (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12747908.txt.

Description

An android with a human face can smile, frown, and blush, but how visitors judge it may depend on how they ask. Ayaka Fujii and seven co-authors staged that test at Expo 2025 in Japan, in this 2025 study. More than six hundred visitors described situations to the android Nikola, which used a large language model to pick from sixty-three built-in faces, from subtle feelings to physical states, at varied intensities.

Questionnaires showed a twist: when visitors phrased requests abstractly, leaving the emotion unnamed, they rated the robot's expressions as more fitting and more emotional than when they named the feeling outright. Vague instructions made the robot seem more like an agent with its own mind. Impressions also shifted with the faces shown and the visitors' own personalities, material for designers of social machines.

Catalog record

CREATORAyaka Fujii, Carlos Toshinori Ishi, Kurima Sakai, Tomo Funayama, Ritsuko Iwai, Yusuke Takahashi, Takatsune Kumada, Takashi Minato
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE46 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC12747908

Browse the full library · About the Tower of Babel