1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35...

Counting isn’t just a human habit—it’s a window into how the mind maps numbers and proportions. The record-chasing stories at the start set up a bigger point: people can count in wildly different ways, and the “default” number line in the brain often isn’t additive at all.

Canadian Mike Smith holds the world record for the largest number counted to in one breath—125. But the largest number ever counted to belongs to Jeremy Harper of Birmingham, Alabama. Harper never left his apartment for the effort, sleeping normally but spending roughly three months counting nonstop from wake-up to bedtime. He streamed the process online, raised money for charity, and eventually reached one million. The scale is hard to grasp: one million seconds is more than 11 days, while a billion seconds stretches beyond 31 years. A full video of Harper reaching the entire million isn’t available, but John Harchick has streamed counting up to 100,000, and his other channels include hundreds of videos of himself eating carrots and thousands drinking water—some with virtually no views. A separate website by Jon van der Kruisen helps surface such overlooked videos by autoplaying clips that have yet to be watched.

From there, the discussion shifts from endurance to cognition. The familiar counting method—additive counting—moves by adding one each step. Logarithmic counting instead multiplies by a number each step, creating a scale where equal “distances” correspond to equal ratios. That matters because human perception often behaves multiplicatively. Loudness is the clearest example: two boomboxes at the same volume don’t sound twice as loud. To be perceived as about twice as loud, the sound intensity needs to be about ten times greater; doubling again requires about a hundred, and another doubling pushes toward a thousand.

Logarithmic intuition may also help explain survival decisions where proportions matter more than raw counts—such as whether there is one lion in shadows versus two, a difference that’s far more consequential than whether there are 96 lions versus 97. It may even connect to why life feels like it speeds up with age: each new year becomes a smaller fraction of the total time lived.

Yet logarithmic thinking can mislead in everyday behavior, especially when people overreact to proportional differences that don’t actually change the underlying value. The “price paradox” research finds people will work hard to save $5 from a $10 purchase, but show far less effort saving $5 from a $2,000 purchase—despite the savings being identical.

The mind’s number sense isn’t uniform either. People can “subitize,” instantly recognizing up to about four objects in a scene without counting. Beyond that, an approximate number system kicks in, accurate only around 15%—enough to distinguish quantities that differ by at least that margin (like 100 vs. 115, or 1,000 vs. 1,150).

Children illustrate how this changes with development. Under age three, kids can compare small sets of coins without counting even when the physical spacing changes. Around 3.5 years old, that ability appears to falter, likely because physical size cues become more dominant; then, as linear counting is learned, the correct comparisons return.

Finally, the transcript uses physics to stress-test intuition. If 1 and 9 sit on a linear number line, the halfway point is 5. But without linear training, many children answer 3—because on a logarithmic scale, 3 is proportionally centered between 1 and 9. When 1 is treated as the Planck length and 9 as the observable universe, the “middle” becomes a number of Planck lengths that matches the number of brain cells needed to span the observable universe—an intentionally mind-bending way to connect perception, scaling, and the limits of observation.