Abstract Title

Building a Trustworthy Robot- Robot Color and Perceived Trust

Presentation Type

Poster

Abstract

Trust in robots is crucial in medical, military, and everyday environments because we rely so heavily on the aid of robots in many situations; this trust becomes particularly vital in high risk circumstances when we are forced to depend on robots. In order to rely on robots, we must trust them first. Previous research has shown robot color to be a significant predictor of perceived approachability, playfulness, aggressiveness, professionalism, and masculinity, but there is currently insufficient research on color’s relation to trust. We aim to assess attitudes about robots and investigate how robot appearance, specifically color, relates to trust. As part of a larger study, participants’ perception of robot color will be examined. We are in the process of collecting data which will be used to examine what similar robot characteristics participants employ when asked to build a “trustworthy” robot using simulated robot software; we propose that similarities in color across participant built robots will be found. In order to account for individual differences, participants (200 college undergraduate students) complete an array of surveys measuring their attitudes toward robots prior to creating a robot on a computer simulation called V-Rep. V-Rep is a fully-customizable robot simulator in which each object can be individually controlled. Participants first watch a fifteen minute tutorial which explains in depth how to use the simulation. After completing the surveys and training, participants are asked “What do you think a trustworthy robot looks like?” and are prompted to build a robot that resembles this in a fixed amount of time. Participants are then given two post surveys regarding robot color and other characteristics they find trustworthy in a robot. Upon completion of data collection, analysis of the aforementioned variables will be performed. Implications for our findings include providing a basis for robot design improvement in an array of domains – not limited to those previously stated.

This document is currently not available here.

Share

COinS
 

Building a Trustworthy Robot- Robot Color and Perceived Trust

Trust in robots is crucial in medical, military, and everyday environments because we rely so heavily on the aid of robots in many situations; this trust becomes particularly vital in high risk circumstances when we are forced to depend on robots. In order to rely on robots, we must trust them first. Previous research has shown robot color to be a significant predictor of perceived approachability, playfulness, aggressiveness, professionalism, and masculinity, but there is currently insufficient research on color’s relation to trust. We aim to assess attitudes about robots and investigate how robot appearance, specifically color, relates to trust. As part of a larger study, participants’ perception of robot color will be examined. We are in the process of collecting data which will be used to examine what similar robot characteristics participants employ when asked to build a “trustworthy” robot using simulated robot software; we propose that similarities in color across participant built robots will be found. In order to account for individual differences, participants (200 college undergraduate students) complete an array of surveys measuring their attitudes toward robots prior to creating a robot on a computer simulation called V-Rep. V-Rep is a fully-customizable robot simulator in which each object can be individually controlled. Participants first watch a fifteen minute tutorial which explains in depth how to use the simulation. After completing the surveys and training, participants are asked “What do you think a trustworthy robot looks like?” and are prompted to build a robot that resembles this in a fixed amount of time. Participants are then given two post surveys regarding robot color and other characteristics they find trustworthy in a robot. Upon completion of data collection, analysis of the aforementioned variables will be performed. Implications for our findings include providing a basis for robot design improvement in an array of domains – not limited to those previously stated.