Mental Framework 6: Workload Sets the Speed
I’d love to see what a video game designer could do to a roadway design to keep drivers paying attention. Have you ever noticed how the best video games are just the right amount of difficulty? If they’re too hard, you quit. If they’re too easy, they’re not any fun. Since it’s a game, you can leave if you get bored.
Driving in real life isn’t that way. If it’s too hard, you slow down to deal with the information flow. If it’s too easy, your mind wanders to other things and your right foot gets heavy. Nature abhors a vacuum, and vacuous thinking. Researchers found that people (especially men) would prefer giving themselves a mild electric shock rather than remain with their thoughts; a shock they would have paid to avoid under different circumstances.
Driving researchers figured out the overload part could be a problem many years ago. Life comes at you fast when you’re going upwards of 40 mph. Add in conflicts and congestion and it’s easy to get overwhelmed. As a result, most of driving research has focused on reducing workload, assuming that crash rates would also be reduced.
Turns out that’s not entirely the case. There’s an entire psychological line of thought in driving research that explores the nuances of risk tradeoffs. Things do get a little safer as we reduce the difficulty of the driving task—at least for a bit—and then drivers adapt to the demands and an entirely new type of crash emerges, one that’s often much worse. It’s an easy mistake to make if you’re only looking at a specific problem. Reminds me of one of Solomon’s proverbs: “A man who remains stiff-necked after much reproof will suddenly be shattered beyond recovery.” We can plan around stubborn people, but then they get themselves into trouble we can’t fix. That’s especially true when the people that are at risk, like pedestrians and bicyclists, are not the ones causing the problem.
The new goal for our era is to get the workload right: just enough to keep people engaged but not so much that they’re overwhelmed.
Workload and Urban Settings
When you’re driving on a highway or arterial, most of the work comes in avoiding conflicts from other vehicles. You may be going really fast, but the dangers are all really big, so you can see them from a long way away. Flow interruptions happen, but they’re rare and we’ve made them all consistent and easy to see. Workload in an urban setting is a completely different matter.
To get drivers workload up and speeds down, you have to get their attention and keep it.
As we talked about in the first 4 frameworks, the potential for human interaction will get their attention if (and only if) it’s close enough to be a treat or threat. That’s the primary component in the driver’s workload. Then you have to keep them engaged over time, which means repeated novelty. In congested settings, the leaders of the pack set the pace, but the more often you give them something to do or see, the more they have to engage and the less they will mentally check out.
That means that workload in urban settings has three main components:
People
Nearby
Interruptions
Because workload manages speed, these are the primary levers for getting the speed you want.
Speed Prediction
We saw a few weeks ago that speed choice was very strongly impacted by the visual width of the corridor, but only if people are going to be there in the corridor. We saw last week that vehicle movement was directly impacted by how frequently you interrupt the driver, both physically and visually. It turns out that when I tested all the variables in our study for their ability to predict speed, the most important ones directly related to these three components. I generated hundreds of speed equations that fit the data from Tampa and Seattle and then tested them on Orange County speed counts.
Here’s my favorite:
FF 8th %ile Speed = -5.26 + 9.9 Ln (Eye Width) - 1.58 Doors/100’ + 0.0068 BL
where Eye Width is the visual width of the tunnel the driver sees and BL is block length.
This works remarkably well for most multimodal street-level corridors. It’s usually within about 3 mph and if it’s not, it’s usually easy to tell why. You can generate a better formula if you just use the data from a single region. Drivers pick up habits based on the context they regularly use—Rural Georgia drivers are not going to drive like folks from the Bronx.
Implications:
Speed prediction has tremendous power to transform our design systems and our community. It’s a much-needed dose of reality in the face of a bunch of wishful thinking. Knowing what it is that drivers are seeing and how that changes their speed selection gives us the power to change the environment to get the speed we want, or at least understand when our expectations are unrealistic.
For instance, the data shows that when the visual width gets over 60 feet, it’s going to be nearly impossible to get drivers to freely choose a speed below 30 mph, but just getting it that narrow doesn’t guarantee slow speeds. It’s probably unrealistic to believe that you’re going to get slow speeds in an area with a bunch of subdivision walls—there’s no doorways to indicate human presence that would pull the speed down. They’re usually quite long too, which further complicates matters.
Another implication is that access management can have disastrous impacts on speed. When you close off an access point, it doubles the block length. Going from 330 feet to 660 feet adds over 2 mph. Going from 660 feet to 1320 adds 4.5 mph. Increasing from 1/4 mile to 1/2 mile adds 9 mph. Access management has its place. It drops vehicular conflicts dramatically. If that’s all you care about, have at it. If you’re going to have people moving around, a strict access management strategy will cause more problems than it solves.
You can use a prediction speed to look at a corridor reconstruction plan and predict what the speed will be after you’re finished. Does your plan give you the change you want? It also gives you insight on what will change the outcome. If you know that a line of trees or more roadway connections will get you a lower speed, it may be worth it to redesign the project to get what you’re aiming for.
At the end of next week, I’ll talk about the caveats I found when I was testing the data, but on Monday, we’re going to look at the first of the BrickCity designs: residential traffic calming.