Investigating Gender Differences in Human Interactions with a Transparent Swarm

Abstract

Research suggests that different operationalisations of automation transparency can influence team collaboration and performance. Yet, little is known about how gender differences can affect humans’ ability to process the information provided by their automation teammate when the automation is transparent. The significance of this research question stems from the fact that the broader areas of judgment and decision making show that females and males deploy different information processing strategies. The aim of this paper is to explore whether gender differences exist in the way people interact with a transparent swarm. We conducted a user study in which 33 subjects (15 males and 18 females) interacted with a simulated swarm under two conditions: with and without transparency. While no significant differences were detected under the control condition, results indicate that when transparency was added, males were able to utilise transparency early on, while females needed significantly more time to benefit from it. Besides, under the transparency condition, females exhibited significantly longer response times than males. However, as time progresses and towards the last time window of the experiment, females could achieve slightly higher decision accuracy than males. These findings stress the need to consider gender differences when designing transparent human-machine interaction and its training protocols.