A small restaurant I pass every day on the way to my lab has a dog. It’s a medium-sized stocky yellowish fuzzy dog, with a face and build approximating something between a dingo and an Akita. He lives outside in a small dwelling made of wood and decorated with the words “진돌이 집 (Jindori’s House)” scrawled in black marker across the back panel.
Yeah, I know what you’re thinking; that’s the whole point of this post. A dog outside a Korean restaurant? That won’t last long, har har har. Except it’s not a joke because some small part of me did think that, every day, when I passed the dog tied up with a fraying rope to its little dirty house behind the dumpster, chewing on some newly discovered truc du jour: Crumpled newspaper, a tin can, his own tail.
I scrutinized daily the restaurant’s short list of menu items pasted into the window. Potato pancakes, doenjang jjigae, shaved noodles. Soup a la Jindori was, however, absent. In any case, the dog didn’t look that much like the typically eaten 똥개 who spent empty days dining on their own poop in grassless backyards until sold to be butchered at the market. Too much lean muscle beneath his explosion of fur. Or maybe too much love in his eyes.
But today the dog was gone, his lonely wooden house pushed back against the restaurant facade behind some old boxes. They couldn’t have… could they? The cherry blossoms had just opened two weeks ago, and it hadn’t even broken fifteen degrees yet…
Those heartless, thoughtless, soulless dog-eating Koreans. They’d eat their own children if it meant kick-starting their stagnant qi, said my unbidden brain.
I suddenly thought back to a Korean child I had seen outside the Ministry of National Defense the day before. He was belting out some tuneless song that seemed to have but one line: “대한민국, ba-bum BUM!”
He must have gotten it wrong because his mother stepped in to correct him: “대한민국, ba-bum ba-dum BUM!”
The kid, who was maybe about four, broke into a huge smile. “대한민국, ba-bum BUM!”
His mother hesitated. “No… listen. 대한민국, ba-bum ba-dum BUM!”
The hamster wheel squeaked for a while. “대한민국, ba-bum.. ba-dum BUM!”
I could hear a relieved smile in his mother’s voice. “That’s right! Ba-bum ba-dum BUM!” she repeated, reinforcing the lesson.
“Ba-bum ba-dum BUM!” He responded again.
His elated mother sang the line with him again. And then again. And again…. and again… and again… and again. She must have made him run through the stupid ba-dum bums The Right Way seven more times before I had finally escaped from earshot, and perhaps more afterward as well.
There goes another tiger mom sucking all the fun out of learning, my brain had sneered before I could stop it. Going to turn the kid into a test-eating automaton. All in the name of blindly sought and unconsidered perfection and narrow definitions of—
Now, wait a moment, I had countered myself. Maybe he’s practicing for a performance with real stakes involved. Maybe he has to take a test for some music hagwon. Maybe he has a developmental disorder, and his mom’s extra attention is going to make the difference between integration into society and lifelong handicap. A multitude of factors beyond blind adherence to Asian Tiger Momness might be sufficient for explaining this phenomenon; why are you gravitating toward the easiest and most judgmental one?
Why indeed. Because the human brain is stupid? Far from it, in fact. The title of this post is woefully inaccurate (thus acting as its own evidence?). Our brains don’t barrage us with tiresome generalizations because they’re stupid but because they’re smart. Information processing becomes easier, in the sense that we can do it at higher volumes of data, when we “chunk” similar factors together into larger units. The average human has a working memory capacity that allows her to store the equivalent of 5-12* digits for short-term recall (a number that varies with a variety of conditions, including, of course, presentation modality), meaning that the only way one might quickly process and store a list of random numbers like this:
might be to recognize patterns and group them together into fewer sets of more meaningful values:
         
The bottom set has your age (24), a mediocre test score (82), two extremely recognizable dates (1776, 2012), your professor’s age (38), a well-known airplane model (747), a palindromic area code (202), your kid cousin’s age (8), another palindromic area code (232), and your professor’s boss’s age (46). These ten little pieces of comparatively familiar information are a lot easier to swallow than the vomit-inducing number soup with which we were originally confronted. And if you wanted to consolidate this information even further, you need only connect like with like to build the series into an image of yourself carrying a B-grade test on time-traveling 747 hurtling from 1776 to 2012 with your professor, while dialing your kid cousin at area code 202 and your professor’s boss at 232. And wow your friends with the unbelievable smartness of your brain. Let me reemphasize that all this can be done despite the fact that I typed in the above string at random.
But wait! You say. That’s so unfair for the 7, whose appearance between the 1 and 76 had nothing to do with its involvement in the Declaration of Independence. And those 2’s, they’re so sick of being typecast as ages and area codes. And who’s to say that the 4 didn’t intend to be the kid cousin’s younger sister?
Too bad, retorts the experience of individuals who cultivate seemingly superhuman powers of both short- and long-term memory by generalizing otherwise chaotic data into larger and more predictable groupings: participants in the U.S. memory championship , world champions in pi recitation , and my personal favorite, the 16th-century Jesuit missionary Matteo Ricci, who could allegedly glance at a page of Chinese text for like a minute and then recite all the characters backwards .
This “chunking”–or “stereotyping,” if you will–in the working memory doesn’t just confer benefits to simple data storage. As the working memory is comparable to a computer’s RAM rather than its ROM, the amount of information we are able to hold within it may act as a limit to the complexity of our brain’s output at any moment. And the number of different information cakes we might produce certainly increases with the number of different ingredients we’re allowed to use. As one pair of authors put it, “individual differences in working memory capacity will have implications for any task that requires controlled effortful processing” .
In this vein, a superior ability to stereotype certain kinds of data as being related to other kinds in the absence of any objective connection might underlie the extreme cognitive powers of some savants, like musical savant Derek Amato  and math genius Daniel Tammett . Many of these people have some form of synesthesia, a condition that associates sensory input usually experienced by the rest of us as comparatively dry and naked (numbers, notes, words) with a rich panoply of very concrete and literal colors, shapes, and emotions, affording affected individuals with more proverbial strings with which to tie together and chunk otherwise unrelated information.
So our tendency to connect large amounts of chaos into smaller amounts of easily digestible stereotypes might make us smart. But isn’t being able to intelligently select the situations in which we might use or block this skill a way to make us even smarter? How can we train ourselves to leave the chunking for geography exams and cocktail party name games and better appreciate the world in all its beautiful, messy, and deliciously varied and unpredictable complexity?
One way to escape from inappropriate stereotyping is to exert constant conscious effort pinpointing and deconstructing the brain’s natural smart but lazy generalizations before they form the basis of any larger conclusions. No, that guy who said he majored in French literature can actually have a decent conversation about current directions in physics. No, that gentleman could be overweight because he has endocrine issues, not because he eats too much. No, silly; there’s nothing at all unusual about that Korean woman on the subway speaking perfect American English into her cell phone. In fact, how do you even know she’s Korean at all? And what was that about plastic surgery?
Another way is to build a rich neural networks so that the brain doesn’t have just one or two easy go-to categories for chunking a certain type of stimulus. In other words, meet a lot of people and read a lot. It’s sort of hard for the brain to say “Aha! senior citizen —> retired!” when the concept “senior citizen” calls to mind to one’s 76-year-old unretired grandmother and the word “retired” is connected to an acquaintance in his thirties, while “Americans eat nothing but bread and hamburgers” is apparently quite easy when you’ve never left Beijing.
A sudden movement ahead of me. A woman walking her dog. Medium-sized stocky yellowish fuzzy thing, with a face and build approximating something between a dingo and an Akita.
The dog wagged his tail and grinned at me as he cantered past. As if to tell my stupid brain, I told you so.
* The old adage is “seven plus or minus two,”  but the Harvard students I tested from 2006 to 2008 tended to demonstrate digit spans around 9 to 12. Age-biased sample, Flynn effect, differential effort, or something else?