Existential Crisis: A Social Robot's Reason for Being
Dora Medgyesy, Joella Galas, Julian van Pol, Rustam Eynaliyev, Thijs Vollebregt
TL;DR
This study investigates how a robot's personality influences user perception in a medical questionnaire setting using a within-subject design with a NAO robot powered by a local Llama 3 LLM and Whisper for speech-to-text. Quantitative measures (PANAS and Godspeed) and qualitative feedback reveal that a personality-driven, humor-infused interaction increases positive affect and perceived anthropomorphism, intelligence, and likability, with participants reporting greater engagement and comfort. The findings support incorporating personality into socially intelligent healthcare robots while highlighting limitations related to sample size, cultural context, and model latency. The work demonstrates feasibility of local, LLM-driven robotic dialogue for real-time healthcare interactions and outlines concrete avenues for improving interaction quality through multimodal sensing and dataset-specific fine-tuning.
Abstract
As Robots become ever more important in our daily lives there's growing need for understanding how they're perceived by people. This study aims to investigate how the user perception of robots is influenced by displays of personality. Using LLMs and speech to text technology, we designed a within-subject study to compare two conditions: a personality-driven robot and a purely task-oriented, personality-neutral robot. Twelve participants, recruited from Socially Intelligent Robotics course at Vrije Universiteit Amsterdam, interacted with a robot Nao tasked with asking them a set of medical questions under both conditions. After completing both interactions, the participants completed a user experience questionnaire measuring their emotional states and robot perception using standardized questionnaires from the SRI and Psychology literature.