Game On: Simulation Tests Dynamic Message Signs

Posted along interstates and highways, electronic dynamic message signs serve several purposes. They can inform drivers about everything from traffic and construction to weather conditions. They also may be used to promote safe driving habits to help reduce crashes. The messages can be straightforward safety reminders, serious warnings or humorous pop culture references.

But what types of safety messages are most effective, and for which behaviors? That’s the question the Illinois Center for Transportation at the University of Illinois Urbana-Champaign and Illinois Department of Transportation sought to answer in a joint project.

U of I Civil and Environmental Engineering Professor Yanfeng Ouyang led a six-month special project with Sean Martschinke, IDOT’s bureau chief of Programs, Projects and Safety Outreach, to determine the effectiveness of these different types of messages on specific driver behaviors. The findings provided guidance on what types of messaging to use for targeted behaviors.

“The project was really trying to determine what type of message resonates or ‘sticks’ the most,” Martschinke said. “What type of messaging is effective and on what type of behaviors, and how do we adjust our messaging accordingly.”

To investigate these questions, Ouyang’s team developed an online simulation to gather experimental data on users between the ages of 18 and 30. The game, created by U of I Civil and Enviromental Engineering doctoral student Yuhui Zhai, simulates real-life driving conditions on a four-lane highway.

Players are positioned inside a vehicle and can control when to speed up, slow down, brake or change lanes, all while being exposed to various messaging on dynamic message signs. The participants also were given “distractions” they could engage with, such as taking phone calls or reading. The researchers instructed the players to drive for at least three minutes, during which they recorded the players’ actions.

“We can control everything behind the game – the type of messages, the frequency, the ambient traffic environment, and so on – and collect data,” Ouyang said. “We collect drivers’ subjective recollection of what happened via a post-game questionnaire. We also record their actual activities throughout the game, such as speeding up and slowing down behavior before and after the time a message was shown.”

After completing the simulation, players reported their opinions of the dynamic message signs as well as their self-reported driving behaviors and sociodemographic information. The research team could also see if any driving behavior changed during the simulation when a dynamic message sign was present.

In the exploratory experiments, Ouyang’s team found that humorous or pop culture messages were generally effective overall in altering the participants’ behavior and participants remembered them well. They found the most effective messaging tone for decreasing speed and discouraging honking (avoid humor), discouraging speeding (use emotionless), reducing calling and texting frequency (use humor) and encouraging seat belt use (use negative).

“For IDOT, the project will guide what messages we display around what behaviors we are trying to influence,” Martschinke said. “It’s adjusting our messaging on these signs to what this small-scale research shows is impactful. Our hope, based on this research, is that we will see better driving habits because of these adjustments.”