“I See What You Did There”

Abstract

Unpredictability in robot behaviour can cause difficulties in interacting with robots. However, for social interactions with robots, a degree of unpredictability in robot behaviour may be desirable for facilitating engagement and increasing the attribution of mental states to the robot. To generate a better conceptual understanding of predictability, we looked at two facets of predictability, namely, the ability to predict robot actions and the association of predictability as an attribute of the robot. We carried out a video human-robot interaction study where we manipulated whether participants could either see the cause of a robot’s responsive action or could not see this, because there was no cause, or because we obstructed the visual cues. Our results indicate that when the cause of the robot’s responsive actions was not visible, participants rated the robot as more unpredictable and less competent, compared to when it was visible. The relationship between seeing the cause of the responsive actions and the attribution of competence was partially mediated by the attribution of unpredictability to the robot. We argue that the effects of unpredictability may be mitigated when the robot identifies when a person may not be aware of what the robot wants to respond to and uses additional actions to make its response predictable.

Predictable Robots for Autistic Children—Variance in Robot Behaviour, Idiosyncrasies in Autistic Children’s Characteristics, and Child–Robot Engagement

“…While other studies did find relationships between IU and autistic features, such as the presence of sensory sensitivities [95], repetitive motor behaviours, and insistence on sameness [145], we found no results in our study that support a relationship between IU and predictability. Similarly, in an earlier study with (typically developing) adults where we manipulated predictability only in terms of topic variance, we found no relationship either between IU and the social perception of the robot in terms of warmth, competence, and discomfort [128]. While unpredictability and uncertainty are often used interchangeably, suggesting conceptual similarity [56], our results suggest that greater caution is warranted when doing so.…”

“…Thus, robot predictability could affect the engagement of autistic children more strongly when they become more sophisticated and humanlike. In our study, we looked at the ability to predict the robot's actions, which is not the same as the extent which one perceives the robot to be predictable (attributed predictability) [128]. Even though a robot's behaviour is more difficult to predict, it can still be considered to be more predictable [41,128].…”

“…In our study, we looked at the ability to predict the robot's actions, which is not the same as the extent which one perceives the robot to be predictable (attributed predictability) [128]. Even though a robot's behaviour is more difficult to predict, it can still be considered to be more predictable [41,128]. Current measures for attributed predictability are too complex to be used for autistic children in our study, as they rely heavily on language and intellectual ability.…”

“…In the context of social skill learning, experiencing discomfort due to dealing with unpredictability is problematic as it prevents children from being in a state where they are ready to learn. Incorporating a robot in social skill learning might be helpful in that it can provide a highly predictable manner of learning social skills, as we can systematically control the predictability of the robot's behaviour [128]. Indeed, the predictability of a robot is a commonly used argument for why robots may be promising tools for autism professionals working with autistic children [e.g., 34,35,43,67,120,141].…”

“…For more information on robot predictability, and different operationalisation thereof, see Schadenberg et al[128], or see Chapter 6 in Schadenberg[124] for a more extensive elaboration.…”