Abstract
As teams of people increasingly incorporate robot members, it is essential to consider how a robot's actions may influence the team's social dynamics and interactions. In this work, we investigated the effects of verbal support from a robot (e.g., “good idea Salim,” “yeah”) on human team members' interactions related to psychological safety and inclusion. We conducted a between-subjects experiment (N = 39 groups, 117 participants) where the robot team member either (A) gave verbal support or (B) did not give verbal support to the human team members of a human-robot team comprised of 2 human ingroup members, 1 human outgroup member, and 1 robot. We found that targeted support from the robot (e.g., “good idea George”) had a positive effect on outgroup members, who increased their verbal participation after receiving targeted support from the robot. When comparing groups that did and did not have verbal support from the robot, we found that outgroup members received fewer verbal backchannels from ingroup members if their group had robot verbal support. These results suggest that verbal support from a robot may have some direct benefits to outgroup members but may also reduce the obligation ingroup members feel to support the verbal contributions of outgroup members.
Highlights
Over the past several decades, researchers have consistently demonstrated that a team’s social dynamics are powerful predictors of both the satisfaction of team members and the team’s overall performance (Jones and George, 1998; Edmondson, 1999; Woolley et al, 2010; Shore et al, 2011)
Team Social Dynamics: Psychological Safety and Inclusion Psychological safety is a term coined by Amy Edmondson and is defined as a “shared belief held by members of a team that the team is safe for interpersonal risk taking” (Edmondson, 1999)
Using post-hoc comparisons using Tukey-adjusted estimated marginal means, we found that outgroup members with no robot verbal support received significantly more verbal backchannels (M = 27.96s, SD = 14.71s) than both ingroup members with robot verbal support (M = 15.26, SD = 8.28, c = −10.72, standard error (SE) = 3.44, p = 0.013) and ingroup members with no robot verbal support (M = 14.92, SD = 8.93, c = −11.29, SE = 3.78, p = 0.019)
Summary
Over the past several decades, researchers have consistently demonstrated that a team’s social dynamics are powerful predictors of both the satisfaction of team members and the team’s overall performance (Jones and George, 1998; Edmondson, 1999; Woolley et al, 2010; Shore et al, 2011). As robots increasingly join human teams and collaborate with people on a variety of tasks, it seems reasonable to program these robots with the ability to positively contribute to important team social dynamics, like inclusion and psychological safety in order to maximize team performance In line with this idea, recent work has discovered robot behaviors that can positively shape specific social dynamics in groups and teams of people, including cohesion (Short and Mataric, 2017), conflict resolution (Shen et al, 2018), conversation dynamics (Traeger et al, 2020), and verbal participation (Tennent et al, 2019). A comprehensive survey at Google, involving over 200 interviews and examining hundreds of attributes of more than 180 Google teams, concluded that psychological safety was the most influential factor in the success of Google teams (Rozovsky, 2015)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
