In the same boat

Abstract

In this paper, we analyze what effects indicators of a shared situation have on a speaker’s persuasiveness by investigating how a robot’s advice is received when it indicates that it is sharing the situational context with its user. In our experiment, 80 participants interacted with a robot that referred to aspects of the shared context: Face tracking indicated that the robot saw the participant, incremental feedback suggested that the robot was following their actions, and comments about, and gestures towards, the shared physical situation and linguistic references to the dialog history indicated to participants that the robot had learned from the interaction and perceived its surroundings. The results show that especially the linguistic and gestural references to the shared context have a significant influence on participants’ compliance with the robot’s suggestions. Thus, indicating that it is ‘in the same boat’ with the user, i.e. that it is sharing the situational context, increases a robot’s persuasiveness during advice giving.

Defining Interaction as Coordination Benefits both HRI Research and Robot Development: Entering Service Interactions*

“…As for the what, human interaction is characterized by creating and evoking common ground (Clark 1996), and we know already for certain that grounding and common ground have a considerable effect on interaction quality (e.g. Fischer et al 2021;Ligthart 2022). Moreover, interactants coordinate their focus of attention, for instance, by means of gaze.…”

Social Robots for Social Institutions: Scaling up and Cutting Back on Cognition

“…Thus, research in human-machine interaction and related areas indicates that physical embodiment is a game-changer in human-robot interaction: In comparison to embodied conversational agents, computer simulations, voice chatbots, etc., interaction with locomoting social robots yield human behaviour patterns that resemble patterns in human-human interaction to a significantly greater extent than do patterns caused by virtual agents (embodied conversational agents, agent-based simulations, voice chatbots, tele-present robots), including video replay of the same robot as in the physical condition. Humans take physically embodied robots more seriously (Fischer et al 2019), follow their suggestions more often (Bainbridge et al 2011), and find them more convincing (Fischer et al 2021). Physical robot interaction results in better learning outcomes (Leyzberg et al 2012), and a more positive perception of the robot and better user performance (Li 2015).…”