We tend to believe what we see with our own eyes, and we tend to extend that belief to what others saw with theirs. But there are several problems with this: one of the major ones is that we don’t always see what’s right in front of us.
That’s a pretty famous video by now, and you probably know what to look for. But many others have replicated Simons et al’s research. For example, a couple of high school students’ version of this test of perception and cognition is less polished than the Simons group’s, but they note that only 50% of the viewers catch the changes and discontinuities in the video.
Of course, brainteasers depending on this kind of thing: “What seven items are different in Picture 2?” — have long been a staple of activity books for children. But as adults we forget how hard it was to find every missing cufflink or changed shoelace in a sketch. (Or we assume that we could do it now that we’re grown-ups a lot easier than we could at Age 9. Exercise for the reader: next trip to the dentists’ office….)
Simons and other researchers have gone on to note that if the change is gradual, humans are remarkably blind to it:
We can’t go into all the details here, but in the 1930s the world’s militaries developed sophisticated tenets and tools of imagery analysis. From the 1960s, overhead imagery of even the most aggressively-denied areas became possible with artificial earth satellites. Beginning in the 1980s, the cascade of data emerging from these satellites was more than humans could process unaided, even in the wealthy nations that could afford to launch and maintain satellites on station over points of interest.
So the first, halting steps towards computerized analysis were begun under very great secrecy. And one development that shocked the analysts was that the computers, using change analysis algorithms, were finding changes on sites, the imagery of which had been thoroughly examined by human eyes.
What does this mean to you as an operational soldier, working cop, or armed citizen? It means you need to be humble about the quality of your perception, and cautious about not only the “known unknowns” but also the “unknown unknowns.”
A bunch of innumerate, credentialed but uneducated reporters beat up on Donald Rumsfeld 10 plus years ago for using those terms, but they’re venerable engineering concepts that need to be part of your constant self-assessment of your situational awareness. Known Unknowns are things that you don’t know, but at least know that exist, and so that you’re able to plan for. A Known Unknown might be: how hazardous a particular street in Philadelphia is if the Sixers are playing an important playoff game tonight, or how hazardous the registered sex offenders in your community are to your kids. And Unknown Unknown might be: whether there is an uncaught, unsuspected sex offender in your community (like the creepy general this week).
The thing is that, while Unknown Unknowns are obviously unknowable in detail, what Rumsfeld’s point was, was this: they can be anticipated as a group or as a concept. You need to know that they’re out there, even when you can’t tell what they are. Insurance companies are good at this. They have learned from long experience to include on their balance sheets, not only reserves for known claims, even when the amounts of the claims are unknowable; classic known unknowns. But they also include a reserve, the setting of which is more actuarial art than actuarial science, for claims that are “incurred but not reported” — classic unknown unknowns.
As good as the insurers, with in some cases centuries of experiential records, are at this, Beltway public policy intellectuals, particularly the young wunderkind sort of baby duck for whom every event is an unprecedented novelty, are hopelessly bad at it. Hence the excoriation of Rumsfeld. They didn’t see any gorilla, so there is no gorilla.
Unlike the Beltway intellectual set, we play real-world games for real-world stakes. A very high percentage of losers at the games of gunfight and ambush make that same mistake — and die complacent, or if they last a second or two longer, surprised. Don’t make that mistake. Broaden your palette of information sources. Question your own assumptions. Expect the gorilla.
He’s really there.