Traffic models
(I don't think I've posted this geeky take on driving before.)
I grew up driving in Nebraska, where "limited access highway" more or less implies "sparsely traveled" (except for a few places in Omaha), so learning to drive well on Chicago and LA freeways was quite an experience. After I'd more or less figured it out, and more importantly after a bunch of conversations with Kim, I hit upon a possibly useful analogy to explain the difference.
In Nebraska, I naturally used a "particle" model of interstate driving. I was able to track all the other cars on the road individually (maybe even for a mile or so in each direction), and I had detailed expectations about their motion. But on big city freeways, that mental model was overwhelming: there were just too many cars to track in such detail, and their constant interactions meant that predicting individual behavior was futile.
Instead, my approach to city freeway driving seems to use a "fluid" model. I don't really track individual cars: I track "the surrounding traffic" as a whole. I see when it's speeding up or slowing down, when it shifts from smooth to chaotic, and the like. I also track significant disruptions in that flow like aggressive drivers or stopped cars. Just like a real fluid, it's the collective motion that allows me to grasp what's going on amidst the underlying complexity. (And just like a real fluid, scientists and engineers have studied phenomena like wave propagation and turbulence in traffic: I should have known to make that mental leap much sooner than I did.)
One day, I'd like to be able to match Kim's LA-native comfort level with traffic. She seems to use an improved fluid model that also manages to keep careful track of the nearby cars (or perhaps the nearby gaps... hmm) to make lane changes a breeze. Meanwhile, I'm just entertained to see again how naturally our brains do complex physical modeling to adapt to new situations.
I grew up driving in Nebraska, where "limited access highway" more or less implies "sparsely traveled" (except for a few places in Omaha), so learning to drive well on Chicago and LA freeways was quite an experience. After I'd more or less figured it out, and more importantly after a bunch of conversations with Kim, I hit upon a possibly useful analogy to explain the difference.
In Nebraska, I naturally used a "particle" model of interstate driving. I was able to track all the other cars on the road individually (maybe even for a mile or so in each direction), and I had detailed expectations about their motion. But on big city freeways, that mental model was overwhelming: there were just too many cars to track in such detail, and their constant interactions meant that predicting individual behavior was futile.
Instead, my approach to city freeway driving seems to use a "fluid" model. I don't really track individual cars: I track "the surrounding traffic" as a whole. I see when it's speeding up or slowing down, when it shifts from smooth to chaotic, and the like. I also track significant disruptions in that flow like aggressive drivers or stopped cars. Just like a real fluid, it's the collective motion that allows me to grasp what's going on amidst the underlying complexity. (And just like a real fluid, scientists and engineers have studied phenomena like wave propagation and turbulence in traffic: I should have known to make that mental leap much sooner than I did.)
One day, I'd like to be able to match Kim's LA-native comfort level with traffic. She seems to use an improved fluid model that also manages to keep careful track of the nearby cars (or perhaps the nearby gaps... hmm) to make lane changes a breeze. Meanwhile, I'm just entertained to see again how naturally our brains do complex physical modeling to adapt to new situations.