In-Vehicle Information Systems: Design Trends and Their Impact on Driver Behavior

Abstract

Introduction: Humans have a limited supply of resources to apply to a given task. Concurrent tasks that demand similar resources will cause interference and task performance will degrade. In-vehicle information system (IVIS) interactions are common while driving and can interfere with performing the dynamic driving task. Recent IVIS design trends have replaced physical analog controls with digital controls on touchscreens. Lack of feedback when interacting with digital controls may demand more of a driver as opposed to analog controls that may be used via tactile feedback alone. This dissertation investigated differences in driver distraction, driving performance, and duration of interactions between analog and digital IVIS use while driving. Method: Data from three naturalistic driving studies were used. Second Strategic Highway Research Program Naturalistic Driving Study (SHRP2 NDS) represented vehicles with analog IVIS (26 makes; model years [MY] 1996 – 2009), while Virginia Connected Corridor 50 NDS (VCC50 NDS; (Teslas; MY 2015 – 2017), and VTTI L2 NDS (Teslas; MY 2018 – 2022; Subarus; MY 2017 – 2021) represented digital IVIS. Driver eyeglances towards the IVIS were manually coded using video data to measure driver distraction. Standard deviation of lane position (SDLP), standard deviation of speed (SDS), and maximum X--Y accelerations were calculated for driving performance. Interaction duration was coded for each IVIS interaction and was used to represent duration. The distraction and driving performance studies used the same subset of data with representation across various driver ages (M = 39.30; SD = 16.35) and gender (Male = 616; Female = 494). Duration used a different subset of data with driver age (M = 43.05; SD = 13.66) and gender (Male = 219; Female = 163) similar to the other studies. Results: Driver distraction was higher for drivers interacting with a digital IVIS compared to analog IVIS. Driving performance, particularly SDLP, was worse for drivers interacting with digital IVIS compared to analog IVIS. Interaction duration did not differ between digital IVIS and analog IVIS. Conclusion: Interacting with digital IVIS had more impact on driver behavior than using analog IVIS. This difference is likely driven by using a touchscreen during interactions compared to physical controls with differences in interaction duration being ruled out as a confounding influence. Interpretation of these results should be limited to only the vehicles included in this study. Future research should include more digital IVIS models to better represent that population as only two manufacturers' approaches were in this dataset. Practical Implications: For the vehicles included in this study, digital IVIS interactions had greater impact on driver behavior than using analog IVIS. Previous research suggests digital IVIS can be designed to limit driver behavior impacts. Manufacturers considering an analog-to-digital IVIS transition should investigate the optimal design of digital IVIS to minimize impact on driver behavior.