Levels of What?Investigating Drivers’ Understanding of Different Levels of Automation in Vehicles

Abstract

Extant levels of automation (LoAs) taxonomies describe variations in function allocations between the driver and the driving automation system (DAS) from a technical perspective. However, these taxonomies miss important human factors issues and when design decisions are based on them, the resulting interaction design leaves users confused. Therefore, the aim of this paper is to describe how users perceive different DASs by eliciting insights from an empirical driving study facilitating a Wizard-of-Oz approach, where 20 participants were interviewed after experiencing systems on two different LoAs under real driving conditions. The findings show that participants talked about the DAS by describing different relationships and dependencies between three different elements: the context (traffic conditions, road types), the vehicle (abilities, limitations, vehicle operations), and the driver (control, attentional demand, interaction with displays and controls, operation of vehicle), each with associated aspects that indicate what users identify as relevant when describing a vehicle with automated systems. Based on these findings, a conceptual model is proposed by which designers can differentiate LoAs from a human-centric perspective and that can aid in the development of design guidelines for driving automation.

Do Researchers Use the Levels of Automation Appropriately? How Automation is Discussed in Research

“…Although SAE levels provided a structure to classify and describe automation, many studies have found that a humancentric description of automation may be more valuable (Jamieson & Skraaning, 2018; Kaber, 2018; Novakazi et al, 2021; J. H. Yang et al, 2017).…”

“…A more general issue, affecting human subject research is the "psychologist's fallacy", a cognitive bias that describes how experimenters mistakenly assume their perception of a situation matches the participants (James, 1890). This is particularly critical in vehicle automation studies because drivers' mental models are unlikely to be based on SAE levels and so may diverge from those of the SAE (Novakazi et al, 2021;Lee et al, In Press). Mental model quizzes can assess participants' understanding of the automation and mitigate the psychologist's fallacy.…”

“…Specifically, they were designed as a reference for engineers and are often difficult for drivers to understand. This difference in understanding can be seen in the results of studies in which driver mental models do not align with the SAE levels (Novakazi et al, 2021). Likewise, researchers miscite the SAE levels when they state nonexistent levels of automation capabilities like “Level 2.5” (Ataya et al, 2021).…”

“…Although this taxonomy seems to unambiguously describe vehicle automation, in practice both researchers and drivers often struggle to appreciate the nuances within and between levels (Novakazi et al, 2021). For example, within Level 2, a vehicle may include features like intelligent speed control and automatic lane changes.…”

Design for Perception - A Systematic Approach for the Design of Driving Automation Systems based on the Users’ Perception

“…As a result, the similarities in behaviors between functions and their limitations, makes it hard for drivers to build a correct mental model of the driving modes in the vehicle and build appropriate usage strategies [17] [11]. Therefore, designers and developers need to consider the driver's understanding as more than just a task-allocation issue, as often suggested in prior taxonomies [4] [8] [12], but that the user's perception of the system is a key to designing a desirable interaction. Hence, to design DAS that adequately support the driver in building appropriate trust and usage strategies, it is necessary to understand both the process by which drivers understand a driving automation system and the factors that influence their perception of the DAS [10].…”