Personality Perception of Robot Avatar Teleoperators in Solo and Dyadic Tasks

Abstract

Humanoid robot avatars are a potential new telecommunication tool, whereby a user is remotely represented by a robot that replicates their arm, head, and possible face movements. They have been shown to have a number of benefits over more traditional media such as phones or video calls. However, using a teleoperated humanoid as a communication medium inherently changes the appearance of the operator, and appearance-based stereotypes are used in interpersonal judgments (whether consciously or unconsciously). One such judgment that plays a key role in how people interact is personality. Hence, we have been motivated to investigate if and how using a robot avatar alters the perceived personality of teleoperators. To do so, we carried out two studies where participants performed 3 communication tasks, solo in study one and dyadic in study two, and were recorded on video both with and without robot mediation. Judges recruited using online crowdsourcing services then made personality judgments of the participants in the video clips. We observed that judges were able to make internally consistent trait judgments in both communication conditions. However, judge agreement was affected by robot mediation, although which traits were affected was highly task dependent. Our most important finding was that in dyadic tasks personality trait perception was shifted to incorporate cues relating to the robot's appearance when it was used to communicate. Our findings have important implications for telepresence robot design and personality expression in autonomous robots.

Identifying Personality Dimensions for Engineering Robot Personalities in Significant Quantities with Small User Groups

“…Conscientiousness was identified in many tests as well but significantly less frequent than extraversion. The identification of extraversion agrees with several studies about the head and eye movements or gaze of humanoid robots or virtual agents [20,33,[44][45][46], as does the identification of conscientiousness [33,45]. However, the two dimensions were by no means the only possibilities.…”

“…It is almost certain that, due to the complexity of personality and how it is perceived, a group of users is unlikely to perceive the robot personalities under construction as similar to the archetypes without optimisation, even when the robots exhibit the most 'archetypal' behaviour. This 'self-other' discrepancy has been observed in perceiving human personalities for a long time [32], and in the perception of robot personalities as well [33]. The discrepancy should be minimised for the sake of consistent user experience.…”

Webcam controlled robotic arm for persons with SSMI

“…Nvidia and ITU Copenhagen [2014] published articles on gaze controlled drones achieving primitive movements by following eye gaze of an operator although the system is not yet aimed for people with different range of abilities. Bremner [2016] and colleagues investigated effect of personality cues of a robotic tele-avatar like gesture, speech and appearance on its perception by the operator. They also reported that the perception and individual behavior of operator is subjective and vary for individuals and concluded that any human robotic system should be designed considering the user background, perception and behavior.…”

Multimodal Human-Human-Robot Interactions (MHHRI) Dataset for Studying Personality and Engagement

“…This might have affected the flow of the human-robot interactions, namely, the human participants might have adapted to the questions during the human-human interactions. In our follow-up studies related to personality, we always randomised the order of interactions [47]. • Subject-level engagement annotations.…”