When Calculators Were Not in the Classroom
In Korean schools from the 1980s through the early 2000s, calculators barely showed up in math class. Pulling one out during a lesson was close to taboo, and exams — let alone everyday arithmetic — were expected to be done in your head. The multiplication table was drilled starting in second grade, and the demand that "you should answer any multiplication fact instantly, in random order" is a shared memory for anyone who went through that era.
This isn't just nostalgia. It matters because an entire generation went through roughly the same training. In an environment where calculators were not allowed, handling numbers mentally every day builds up repetitive automation of small operations. What gets sharpened is number sense — an intuitive grasp of the size and relationships of numbers, trained as a daily habit rather than a special skill.
This is an average tendency, not a rule about individuals. Even within the same era, someone who liked math and someone who avoided it ended up very different. But the shared environmental fact — "a whole generation trained to handle numbers without a calculator" — can be a meaningful difference when compared with peers of the same age in other countries.
Why Automation Frees Up Working Memory
The key concept here is automation. When the multiplication table and single-digit addition and subtraction are fully automated, recalling that 7×8=56 is closer to a reflex than a calculation. Why does that matter?
Working memory has a limited capacity for information held at once. As we saw in the working memory discussion, pure capacity sits around 4±1 chunks. If the computation process itself occupies that capacity, there's less room left to understand the actual problem. When basic operations are automated, the computation stops consuming capacity, and those resources go to grasping the problem's structure instead.
Take a word problem like "a page has 24 lines, the book has 17 pages, and each line averages 8 words." If computing 24×17 eats your working memory, the intermediate result fades by the time you multiply by 8. But if 24×17 is at an automated level, you can hold the result and move to the next step. The training the pre-calculator generation received was, in effect, a generational push toward this "computation that doesn't consume capacity."
The Suneat and Time Pressure
This automation training wasn't just educational philosophy — it was a practical pressure, and the reason was the Suneat (수능, the national college entrance exam). The math section, in particular, gave very little time per question, and every answer had to be accurate. With no calculator allowed, fast and accurate computation wasn't optional.
Suneat math problems leaned more on reasoning than raw calculation, but each reasoning step was laced with small operations — fractions, factoring, side-length calculations. Stumble once and time bled away. So students, whether at school or at hagwon (cram schools), naturally drilled themselves on fast, accurate basic computation. "If your calculation is slow, you run out of time no matter how well you think" was a felt reality for the Suneat generation, not a theory.
This time-pressure structure connects directly to what the Arithmetic game measures. That game also tests how accurately you can compute within a time limit. For the Suneat generation, it touches a familiar sensation — the pressure to "calculate correctly, fast, without mistakes" was trained decades ago.
The Calculator Generation Is Different
From the late 2000s onward, especially Gen Z and after, the situation shifted. Schools gradually permitted calculators and digital tools, and everyday occasions to compute numbers by hand decreased. The multiplication table is still memorized, but the automation beyond it doesn't get repeated to the same degree as in earlier generations.
This does not mean "this generation is bad at math." The degree of automation differs, but problem comprehension and reasoning ability don't divide cleanly by generation. The structural point is the same, though: when basic operations aren't automated, working memory gets pulled more into computation during complex problems. So even on the same Arithmetic game, the sense of familiarity can differ by generation.
Again, this is an average tendency, not an individual verdict. A calculator-generation person who enjoys mental math or studied the abacus can have excellent automation, and a Suneat-generation person who avoided calculation may not. Mental arithmetic speed is ultimately governed by amount of practice — especially automation of the multiplication table and single-digit operations. Generation is just the environmental backdrop for that practice.
The Abacus and Mental-Abacus Hagwons — A Spatial Memory Tradition
When talking about number automation in Korea, you can't skip the abacus (주산, jusan) and the mental-arithmetic hagwons that taught it. The abacus is a traditional tool where you move beads to compute, and skilled practitioners reach a stage called mental abacus (심산, simsang) where they compute without a physical abacus by visualizing one in their mind.
Studies of mental-abacus experts show they rely on visuospatial memory during calculation. Rather than verbally rehearsing numbers, they manipulate a mental image of the abacus. This is a different route from typical mental arithmetic, which leans on verbal rehearsal. Research has also found that children trained in the abacus show different patterns on digit-memory tasks.
Mental-abacus experts are a specialized group, and generalizing them to "all Koreans" would be wrong. But the cultural fact that abacus and mental-arithmetic hagwons were widespread in Korea shows that number automation was trained outside school too. The number sense of the pre-calculator generation was shaped by overlapping forces: school math, hagwons, and the abacus tradition.
How to Read Your Game Score
If you score well on the Arithmetic game, it's more likely because your basic operations are automated than because your brain is special. For the Suneat generation, that automation may be a residue of old training; for the calculator generation, it may be the result of deliberate practice. Either way, the score measures your current level of automation, not your innate ability.
Conversely, a low score doesn't mean weak mathematical reasoning. Computation automation and reasoning are different areas, and in calculator-permitted environments, reasoning can matter even more. Treat the score as a reference for observing your current state.
Note: results from this site are not a medical diagnosis or a substitute for a cognitive assessment. They're a reference point for observing your own state.