Understanding trust in an AV context: A mixed method approach
Conference contribution, 2018
Trust is a fundamental part of technology acceptance as well as an important factor for creating a positive user experience with Automated Vehicles (AVs). In order to fully understand users' trust in AVs it is important to consider the cognitive processes by which humans develop trust. We argue that a deeper understanding of these processes can be elicited by using a convergent mixed method design. The method design described in this paper was created during an experimental study investigating the effect of AV's driving behaviour on users' trust. The design consists of five data collection methods, three qualitative and two quantitative, used to collect data during and after test runs with an AV. The results show that the different methods elicited responses that may indicate different cognitive processes. The methods used during the test runs produced more affective and analogical responses while the methods used directly after each of the test runs generated more analytic responses. The last method, introduced after the completion of all test runs, produced a more mixed result. The participants elaborated on their earlier responses and sometimes turned their affective responses into analytic or analogical explanations. Hence, by combining and utilizing the strength of different data collection methods, more rich data was elicited on the trust formation process and thereby creating a more nuanced picture of users' trust in automated vehicles.