Persuasive Robotics

Persuasive Robotics is the study of persuasion as it applies to human-robot interaction (HRI). Persuasion can be generally defined as an attempt to change another’s beliefs or behavior. The act of influencing others is fundamental to nearly every type of social interaction. Any agent desiring to seamlessly operate in a social manner will need to incorporate this type of core human behavior. As in human interaction, myriad aspects of a humanoid robot’s appearance and behavior can significantly alter its persuasiveness.

DSC_9156

MDS_head_composite_680x682

In order to explore the space of Persuasive Robotics we designed a human subject study, run at the Museum of Science in Boston at Cahners ComputerPlace using Nexi, the MDS robot.  The study examined the effect of gender (robot and human gender), touch (between robot and human), perceived autonomy (whether or not the user perceived the robot to be autonomous), and interpersonal distance (how close the robot stood to the human).  Two primary measures were used to quantify the effect of these conditions: donation, and a questionnaire.  The donation was in the form of five $1 bills which the subject received before participating, and had the option of donating in any increment to the robot.  The questionnaire was completed after the interaction and used a number of standard psychological scales to measure the subject’s engagement, perceived credibility of the robot, trust, and happiness.

CCP_subjects_680x498

CCP_space_680x459

Over 300 hundred subjects participated, spread across the many conditions.  The results of this study indicate that factors such as robot-gender, subject-gender, touch, interpersonal distance, and the perceived autonomy of the robot, have a significant impact on the interaction between human and robot, and must be taken into consideration when designing sociable robots. This work applies the term persuasive robotics to define and test the theoretical and practical implications for robot-triggered changes in human attitude and behavior. Its results provide for a vast array of speculations with regard to what practical applications may become available using this framework.  The results are described in the paper, and thesis linked below.

This work was the subject of my Master’s Thesis at the MIT Media Lab, Personal Robots Group.  My thesis advisors included Dr. Cynthia Breazeal (MIT Media Lab, head of Personal Robots Group), Dr. Jeremy Bailenson (Stanford University, director of Virtual Human Interaction Lab), and Dr. Michael Norton (Harvard Business School).

This was a truly multi-disciplinary adventure and without question the academic pinnacle of my time at MIT.  When the work began, Nexi was a robotic head sitting on my desk, with a body that arrived a few weeks later.  The first 6 months consisted of working with a talented group at the Personal Robots Group to develop the basic functionality of the robot.  In parallel I was studying up on the relevant work in the field of Social Psychology as it pertained to persuasion and behavior change.  I was lucky enough to persuade Dr. Michael Norton and Dr. Jeremy Bailenson to act as thesis advisors – their guidance, along with that of Dr. Cynthia Breazeal, proved invaluable.

Mikey Siegel, Cynthia Breazeal, and Michael I. Norton. 2009. Persuasive robotics: the influence of robot gender on human behavior. In Proceedings of the 2009 IEEE/RSJ international conference on Intelligent robots and systems (IROS’09). IEEE Press, Piscataway, NJ, USA, 2563-2568. [pdf]

Thesis: Persuasive Rootics: How Robots Change our Minds [pdf]

This video is just for fun.  Jesse Gray and I at a conference in Laval, France with Nexi, the MDS Robot.