Fig. 6
From: The power of justifications to repair human-robot trust, even under moral disagreement
Experiment 3 reveals the dynamics of people’s trust from baseline (T1 = time 1) to after they learn about the agent’s decision (T2 = time 2) to after the agent clarifies the decision with a justification (blue lines) or explanation (red lines) (T3 = time 3). The dynamics vary substantially depending on whether the participant’s normative recommendation and the agent’s decision in the dilemma agreed or disagreed and depending on whether the agent clarified the decision by using a justification or explanation. The figure shows the robot agent in the toxic gas dilemma, and results are similar in the other conditions (see SM). Panel (a) displays the total trust scores, panel (b) displays the moral trust scores, and panel (c) displays the performance trust scores.
