Framework 3: The limits of seeing each other
What's the difference between a person who is driving and a driver?
When pedestrians describe their interactions, their language is telling: they don’t fear drivers. They fear cars. This is the key difference between an urban space and a suburban one:
In an urban space, conflicts are negotiated eye to eye: people are driving.
In a suburban space, conflicts are not negotiated, so pedestrians and cyclists beware! The cars have no eyes to see you.
In theory, that makes no sense. As a driver, you remain a human being, even when you’re operating heavy machinery. Your brain and heart haven’t evaporated. You should care whether someone is there, no matter how remote the chance is. In reality, it’s obvious there’s a huge difference and anyone who naively believes otherwise is taking their lives in their own hands.
That is because of Mental Framework 1. The automaticity we use to drive makes us reflexively oblivious to anything that isn’t a personal threat or treat. We might respond cognitively (if we see the person at all) but that’s the slow route and doesn’t get the same biophysiological response. Mental framework 2 clues us in to the ways that can change when we are interacting directly with another human being. Human faces and bodies can elicit a biophysiological response because they are the best ways we get treats and the most dangerous threats we can face. The real possibility of a face to face interaction shifts our psyche—but only under the right circumstances.
Mental Framework 3: the People Perception Panorama
Here’s where Framework 3 comes in to play: the person we’re seeing has to be a realistic threat or treat, which means there are spatial and temporal limits involved. Remember, System 1 is only looking for those two things and if what it’s seeing is too far away to be of value, it’s going to disregard it. The person has to be close enough to interact with you. A person that’s too far away can’t hurt you or give you anything you would enjoy. They have to have been in your field of view long enough for them to become a part of your mental model of the space. If they’re out of sight, they’re out of mind. You also need time to process the sensory information you’re receiving. If you’re going by too fast, you may not even be able to process that you saw them, much less interact with them.
So what are the limits?
We can get hints from a few directions, but let’s build this up from what we know:
How far and wide can you see a person?
Dr. Paul Ekman and his student, Dr. Joseph Hagar have spent a lifetime studying the mechanics of facial expressions and the ways that others respond to them. They were the ones who discovered that facial expressions are hardwired into our nervous system and are therefore, universal. By 1979, they had established that all facial expressions can be decoded within 90 feet and extreme expressions get decoded out to about 135 feet.
From an optical standpoint, that’s astounding. To do that, you need to have a visual resolution that can distinguish differences of less than 0.08 inches at 90 feet—which works out to about half a degree within your visual field. Your eyes are actually much better than that in your focal point—about 0.02 degrees—but it drops off quickly:
By the time you’re out to 16-20 degrees, your resolution drops off to that 1/2 degree mark. You can reliably distinguish a clock face from a human face out to 16 degrees and you’ll give human faces priority in your peripheral vision out to that angle.
Body language hasn’t been as well researched on those scores, but the bike researchers have found it takes a very high level of contrast on the moving joints to get much past 150 feet—it’s possible out to about 1,000 feet, but it takes a lot of time to identify the person within that larger area. Point light walker research shows similar dimensions as well. Remember, the brain wave for body language shows up a little later than for faces. Other studies indicated it can take as much as 3.5 seconds to identify a body in a cluttered scene.
So, what can you see from the back of a galloping horse?
Let’s put this in plan view.
Assuming 90-135 feet of linear distance for facial expressions. (you might be able to see farther for body language, but it’s going to take you much longer to even find them in your field of view.) You could also be generous and assume 20 degrees out from center. Researchers call that your Useful Field of View and it’s a metric they use for testing aging in vision. Your resolution drops off below what you need to see when you get out farther than that. This is what I get:
A similar combination of 16 degrees and 135 feet puts you around 75 feet wide.
In essence, you can passively surveil everything within that orange area with just your System 1 (subconscious) mental resources. Let’s call that the people perception panorama (PPP). You may not consciously recognize you’ve seen something there, but your brain will so you can passively surveil it without having to use System 2, conscious thinking. If you’re walking or driving through a corridor that is no more than 60 feet wide, you don’t have to lay your focal point on the people that are there for your body to respond to them if it needs to. If the corridor is any bigger and you’ll have to keep your head on a swivel to feel safe. You can monitor extreme threats in the larger yellow area, but it’s not going to leave you feeling as relaxed.
Is it possible to grow that area? Probably. Trauma victims or those with incomplete attachment have very strong monitoring systems that keep them on high alert at all times, but they also tend to dart their eyes everywhere to keep that flashlight beam covering everything. In areas where drivers expect to see people there, like at frequently used crosswalks, they are likely doing the same thing—but that requires the next one of the mental frameworks.
The Mystery of 60 feet.
What’s wild about this is that we found a bunch of ways that this 60 foot wide corridor hides in other research studies and impacts the way people drive.
Safety researchers love to cart out the idea of perceptual narrowing—in essence, the faster you go, the less you see, or so we thought. When I looked at the original study for perceptual narrowing, I noticed something odd. What they were measuring was the size of the driver’s visual field—in essence, the angle they were looking at. That angle dropped pretty dramatically between 18 mph and 68 mph, but they weren’t taking into account how far ahead the driver was looking at each speed.
Remember the two second rule in defensive driving? Researchers have found that drivers tend to fixate between 1 and 2 seconds ahead of them. So, I put this in plan view as well, creating a circle on the ground centered at 1.5 seconds of travel time in front of the vehicle.
It turns out you’re always monitoring an area about 60 feet wide, no matter how fast you’re going. At slow speeds, you’re monitoring out to that 60’ width anyway. Your fixations are not overshooting what you’re capable of seeing and interacting with. Up to about 30 mph, your fixations stay well within that relaxed monitoring area. By 40 mph, over half of that oval is more than 90 feet away. Landscape architects have long known that something dramatically changes in the driver’s behavior between 30 and 40 mph.
That 60 foot corridor also popped up in the speed prediction formula data. When the visual tunnel you’re seeing as you drive gets wider than 60 feet, the free flow 85th percentile speed no longer drops below 30 mph. When it gets over 90 feet, it’s hard to get a free flow 85th percentile speed less than 45 mph.
The PPP as a historical design default
There’s one other way we found that 60’ dimension manifests. If you go to the oldest parts of the oldest cities—the parts that predate autos and carriages—what you’ll find is that the main arterials are usually about 60 feet wide from building face to building face. The center of Paris is this way—if it gets wider, they throw a linear park in the middle to break it up. London, Rome, Florence, Portland, Seattle—the oldest parts are all 60-90 feet wide. Barcelona was designed with much wider arterials to accommodate carriages, but the typical streets are 65’ wide.
Takeaway:
The physical scale of an urban space is generated by our human perceptual limitations. If you want drivers to slow down and engage in the environment, you need to have it tight enough that they’re interacting with human beings within the PPP or very near to it. This should radically transform how we plan and engineer our streets. Early next week, I’ll describe a few of the implications on our design and operations.