Cheating with robots: how at ease do they make us feel?

Abstract

People are not perfect, and if given the chance, some will be dishonest with no regrets. Some people will cheat just a little to gain some advantage, and others will not do it at all. With the prospect of more human-robot interactions in the future, it will become very important to understand which kind of roles a robot can have in the regulation of cheating behavior. We investigated whether people will cheat while in the presence of a robot and to what extent this depends on the role the robot plays. We ran a study to test cheating behavior with a die task, and allocated people to one of the following conditions: 1) participants were alone in the room while doing the task; 2) with a robot with a vigilant role or 3) with a robot that had a supporting role in the task, accompanying and giving instructions. Our results showed that participants cheated significantly more than chance when they were alone or with the robot giving instructions. In contrast, cheating could not be proven when the robot presented a vigilant role. This study has implications for human-robot interaction and for the deployment of autonomous robots in sensitive roles in which people may be prone to dishonest behavior.

The effect of a robotic agent on dishonest behavior

“…In retrospect, we acknowledge that the reward given was a very low value, at the time we thought this was the standard payment value. We think this fact did not affect cheating behavior because in another study with university students cheating happened in similar levels in an alone condition and being rewarded with approximately 5.8$ USD [26].…”

“…Following this literature, in human-robot interactions it has been seen that in public spaces people steal more in the presence of a watchful robot than a person [14] but in this case being a public space could have shielded people from the watching eyes effect due, for example, to the effect of others in changing the norms of conduct. In more controlled conditions, studies show that just the presence of random [16] or direct gaze behavior [26] of a physically present robot was enough to decrease cheating. Still, this previous study also showed that having a robot showing simple verbal interventions while accompanying a task created the opposite effect, dis-inhibiting cheating.…”

“…However, for this paper we used a video of a robot to see if we could replicate the same effect found with its physical presence (e.g. [16,26]). By trying to mimic the "eyes effect" could a video of a robot looking, influence cheating behavior?…”

“…Some studies have already explored the effects of the presence of a physical robot on dishonesty, showing that a simple watchful behavior can affect cheating (e.g. [16,26]). However, when physical presence is not possible, for example to lower maintenance costs, would that remain the case?…”

“…Here however, we use a video of an anthropomorphic robot blinking its eyes and looking ahead during the task. We opted for the robot video in a loop, instead of a static image, in order to give a greater sense of being with someone else (in this case the EMYS robotic head [19]) and to see if the effect is similar to the one found with the robot physical presence on cheating [16,26]. Overall, due to the effect that eyes can have on making people feel watched [28] and the presence that a watchful avatar can have on cheating [23], we expected that people would cheat more when alone than with the "virtual robot", so we hypothesized: Hypothesis 1: Cheating will be higher when alone than with the robot looking.…”