Believing in BERT: Using expressive communication to enhance trust and counteract operational error in physical Human-robot interaction

The Perception of Emotion in Artificial Agents

“…From a design perspective, emotional reactions such as facial expressions often act as feedback to the human collaborating with the robot. For example, an artificial agent might react with an emotional facial expression in response to a mistake it made during a collaborative task [16], or in response to a human action such as touch for pet-like robots [17]. It has been reported that naïve users, for example children, can regard these responses as genuine emotions [18].…”

“…Specifically, participants reported being more comfortable and satisfied with the robot after the training when the emotion-based feedback was provided. Another intriguing example on the impact of emotion communication on HRI comes from a study featuring a physically present BERT2 robot [16]. During the collaborating with the human, the robot was programmed to work perfectly without expressing emotion, or to make a mistake and correct it without apologizing, or to make a mistake and correct it while apologizing and making a sad facial expression.…”

“…Emotions are largely expressed by a combination of facial, vocal, and bodily signals, and a well-developed literature documents how these different cues interact [19], [63], [130]. While studies on perception of artificial agents' emotion have mainly focussed on one channel, there is some indication that recognition accuracy and evaluation increases for a robot that uses multiple channels, for example, facial and bodily expressions [16], [21]. As human emotions are largely expressed by a combination of facial, vocal, and bodily signals, future studies should use a combination of these signals to study the expression and perception of emotion in artificial agents.…”

On the impact of different types of errors on trust in human-robot interaction

“…The notion of a right level of trust is discussed through existing literature. Hamacher, Bianchi-Berthouze, Pipe, and Eder (2016) state that some human-like behaviours lead to increased levels of trust, but might also have negative impacts when the "behaviour is deemed to cross a line". This is supported by Hancock et al (2011) who describe that there are lower rates of satisfaction when interacting with robots that instil disproportionate trust levels in their human partner.…”

The perception of a robot partner’s effort elicits a sense of commitment to human-robot interaction

“…Previous research has shown that humans' patience towards a robot that performs suboptimally can be increased if the robot employs a mitigation strategy such as seeking human assistance and/or adapting its approach (Lee et al, 2010;Brooks et al, 2016;Mirnig et al, 2017), or expresses a negative emotional reaction and attempts to rectify its mistake (Hamacher et al, 2016). The present study aimed to build upon this research by addressing the more general question of how to sustain human interactants' willingness to persist in interacting with a robot partner despite increasing boredom or frustration --irrespective of whether that boredom or frustration arises from errors on the part of the robot or from the nature of the interaction itself.…”