[Paper Review] Robot's Inner Speech Effects on Trust and Anthropomorphic Cues in Human-Robot Cooperation

Arianna Pipitone, Alessandro Geraci+3 more|arXiv (Cornell University)|Sep 20, 2021

Social Robot Interaction and HRI4 citations

TL;DR

This study investigates whether a robot's overt self-talk—simulating human inner speech—enhances trust and anthropomorphism in human-robot collaboration. Participants interacting with a robot that verbalized its internal reasoning reported significantly increased trust and perceived anthropomorphic traits post-interaction, demonstrating that simulated inner speech improves human-robot cooperation through perceived transparency and human-likeness.

ABSTRACT

Inner Speech is an essential but also elusive human psychological process which refers to an everyday covert internal conversation with oneself. We argue that programming a robot with an overt self-talk system, which simulates human inner speech, might enhance human trust by improving robot transparency and anthropomorphism. For this reasons, this work aims to investigate if robot's inner speech, here intended as overt self-talk, affects human trust and anthropomorphism when human and robot cooperate. A group of participants was engaged in collaboration with the robot. During cooperation, the robot talks to itself. To evaluate if the robot's inner speech influences human trust, two questionnaires were administered to each participant before (pre-test) and after (post-test) the cooperative session with the robot. Preliminary results evidenced differences between the answers of participants in the pre-test and post-test assessment, suggesting that robot's inner speech influences human trust. Indeed, participant's levels of trust and perception of robot anthropomorphic features increase after the experimental interaction with the robot.

Motivation & Objective

• To examine whether a robot's overt self-talk, simulating human inner speech, enhances human trust in collaborative tasks.
• To assess the impact of robot self-talk on the perception of anthropomorphic traits in human-robot interaction.
• To evaluate changes in trust and anthropomorphism perception before and after human-robot cooperation sessions.
• To explore the role of transparency and perceived intentionality in human-robot collaboration through simulated internal monologues.

Proposed method

• A human-robot collaboration task was conducted with participants interacting with a robot that performed overt self-talk during the task.
• The robot verbalized its internal reasoning processes, such as planning and decision-making, to simulate human-like inner speech.
• Participants completed standardized questionnaires before (pre-test) and after (post-test) the interaction to measure trust and anthropomorphism.
• The study used a within-subjects design to compare pre- and post-interaction responses on trust and anthropomorphism metrics.
• Data were analyzed for significant differences between pre-test and post-test scores to assess the effect of robot self-talk.

Experimental results

Research questions

• RQ1 Does a robot's overt self-talk influence human trust in collaborative tasks?
• RQ2 How does robot self-talk affect the perception of anthropomorphic features in human-robot interaction?
• RQ3 Are there measurable changes in trust and anthropomorphism after a human interacts with a robot that performs self-talk?
• RQ4 To what extent does simulated inner speech enhance perceived transparency and intentionality in robots?

Key findings

• Participants reported a statistically significant increase in trust levels after interacting with a robot that performed overt self-talk.
• Perceptions of the robot's anthropomorphic features, such as intentionality and social awareness, increased post-interaction.
• The pre-test and post-test questionnaire results showed a clear shift in user perception, indicating that self-talk enhanced the robot's perceived transparency.
• The study provides empirical evidence that simulated inner speech improves human-robot cooperation by fostering trust and social attribution.