Exploring Persuasive Interactions with Generative Social Robots - An Experimental Framework

This work introduces an experimental framework to investigate the persuasive capabilities of generative social robots (GSRs). To test the applicability of this framework and to gain exploratory insights into persuasive interactions with GSRs, we conducted a pilot study that manipulated a GSR’s appearance and self-knowledge about its appearance. Using descriptive statistics and qualitative analysis, we investigated interaction quality, as well as GSR persuasiveness. Preliminary results suggest that while GSRs may effectively influence user decisions, their persuasiveness could differ depending on the persuasion topic, the robot’s appearance, and the robot’s self-knowledge. We propose refinements to the framework to further explore persuasive dynamics between GSRs and human users.