The perception of benefits and risks as external variables of situational trust in autonomous vehicles

Boff Ferronato, Priscilla

Permalink

https://hdl.handle.net/2142/116147 Copy

Description

Title

The perception of benefits and risks as external variables of situational trust in autonomous vehicles

Author(s)

Boff Ferronato, Priscilla

Issue Date

2022-06-06

Director of Research (if dissertation) or Advisor (if thesis)

Bashir, Masooda

Doctoral Committee Chair(s)

Bashir, Masooda

Committee Member(s)

• Twidale, Michael
• Jones, Patricia
• Rogers, Wendy

Department of Study

Illinois Informatics Institute

Discipline

Informatics

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• trust
• automation
• autonomous vehicles
• perception of risks
• human-computer interaction

Abstract

Driving is a widespread quotidian activity that is at the core of human transportation. At the same time, it is the human driver who is the primary cause of most of traffic accidents in the US. The development and implementation of autonomous vehicles (AVs) propose to dramatically reduce the number of traffic accidents by removing the human from the driving performance. However, the interaction of humans and AVs is currently one of the biggest challenges in AV development. Regarding accidents involving semi AVs from 2014 to 2018, the California Transit Department indicates humans are still the primary cause while only one out of every 38 accidents was due to autonomous system failure. Trust has been identified as one of the main determinants of successful use of different autonomous systems (AS) when calibrated to avoid situations of overtrust, distrust, and mistrust. Trust is a complex process in which individual differences, and external and internal variables have shown to impact the formation and calibration of trust in automation. Drawing from the theoretical model of trust in automation proposed by Hoff and Bashir (2015), in which trust is a dynamic and multidimensional process composed of dispositional, situational, and learned trust, this dissertation aims to empirically examine the situational trust dimension and how external variables—especially the perception of benefits and risks of AVs—are capable of influencing the formation of situational trust in AVs. To achieve these research goals, a survey and interview were conducted which focused on both the perceptions of benefits and risks of AVs as well as the perceptions of an AV’s analogous system. This study found that the perception of benefits is more in relation to attributes of AVs for society (accessibility) or in relation to the purpose of the system, while the perception of risks is more in relation to the individual level and performance of the system. This shows that participants perceive risks of AVs according to those attributes responsible for the formation of reliance on the system: purpose, performance and process. These attributes form the foundation for the development of trust in automation. This study also makes contributions on how to investigate the variables that influence situational trust in AVs and finds especially external ones are related and they have the capacity to influence another. Finally, the participants’ level of trust in AVs related to the adoption of an analogous system (cruise control). While further investigation is needed, this relationship can facilitate the understanding of situational trust in AVs, bringing contextual data from real life events, to complement the current investigation with prototypes and simulations.

Graduation Semester

2022-08

Type of Resource

Thesis

Copyright and License Information

Copyright 2022 Priscilla Boff Ferronato

Owning Collections