Eryn Grant

Is it really good to talk? Testing the impact of providing concurrent verbal protocols on driving performance

Abstract Questions have been raised regarding the impact that providing concurrent verbal protocols has on task performance in various settings; however, there has been little empirical testing of this in road transport. The aim of this study was to examine the impact of providing concurrent verbal protocols on driving performance. Participants drove an instrumented vehicle around a set route, twice whilst providing a concurrent verbal protocol, and twice without. A comparison revealed no differences in behaviour related to speed, braking and steering wheel angle when driving mid-block, but a significant difference in aspects of braking and acceleration at roundabouts. When not providing a verbal protocol, participants were found to brake harder on approach to a roundabout and accelerate more heavily coming out of roundabouts. It is concluded that providing verbal protocols may have a positive effect on braking and accelerating. Practical implications related to driver training and future research are discussed. Practitioner Summary: Verbal protocol analysis is used by ergonomists to understand aspects of cognition and decision-making during complex tasks such as driving and control room operation. This study examines the impact that it has on driving performance, providing evidence to support its continued use in ergonomics applications.

How Do I Save It? Usability Evaluation of a Systems Theory-Based Incident Reporting Software Prototype by Novice End Users

The level of usability achieved by software tools is a key factor that determines their success and indeed uptake by end users. This paper describes a study that was undertaken to evaluate the usability of a prototype incident reporting software tool. The study involved novice end users completing a series of tasks using the software tool and then completing Ravden and Johnsons Human Computer Interaction HCI checklist. The findings identify aspects of the system that pose particular challenges for participants. Participants appeared to lack a clear understanding of the relationship between the information required from them, and the underpinning accident analysis method of the software tool. This is perhaps unsurprising, considering that most incident reporting systems do not include these functions. The findings indicate that the tool requires better levels of intuitiveness to assist users in complex tasks so the focus is on awareness of accident causation methods rather than task instructions. The implications for the design of incident reporting software tools are discussed.

Applying Ecological Interface Design principles to the design of rural highway-rail grade crossing infrastructure

Countries such as Australia and USA have many rail level crossings (a.k.a. highway-rail grade crossings) with limited protection (e.g., static signs only or flashing lights but no physical barriers). Lower cost design solutions are required as upgrading crossings using current infrastructure treatments is cost-prohibitive. Here we applied Ecological Interface Design (EID) principles to develop a novel rail level crossing design for high-speed rural roads, and then evaluated the design via two driving simulator studies. Experiment 1 provided an initial concept evaluation of the crossing, then Experiment 2 evaluated the crossing in safetycritical situations, including when a distractor was present and when crossing infrastructure was damaged through vandalism. Driving performance, subjective workload, and usability were compared against existing rural crossing designs. Findings suggest the EID crossing provides a feasible alternative to existing designs, with the potential to encourage safer decision-making by road users and thus reduce collisions.