I do free diving recreationally and I think the opposite is true. Air in the room is not the bottleneck, ever. Finding the right goal with the right people is. Even with very very little air a lot can be done, in fact constraints, even as fundamental as air, can help.

After digging around, it looks like this area has some replication trouble. And as other commentors have pointed out, submarines operate well beyond these levels and the results failed to replicate in those contexts. Doesn't rule out CO2 as proxy for other components of air though, and most of the studies that failed to replicate added pure CO2.

The real bottleneck is the CO2 while you sleep. More CO2 means worse sleep means worse daytime performance.

At some point I worked with a team of ~10 people, and we did sprint plannings in 20sq meters room from 10am to 5pm. It was like everyone was high

People don't understand ppm like they do temperature units. These rebreathed stickers can help drive home the point. Especially when the next respiratory virus season rolls around. https://naltic.com/product/aranet4-rebreathed-percentage-sti...

You will take ~ 1000 breaths per meeting hour. really gross!

> Then, somewhere in the second hour, the room quietly gets worse at making them.

Maybe it's not just the air but also the multi-hour meetings that drive people to a sense of "oh god let this finally end now", which leads do decisions that fall short.

"Opening a window or a door costs nothing."

Central ventilation system at my job can handle open windows. So that is prohibited :-(

Does it work the other way around? Does breathing air with 0% CO2 improve human cognitive performance? I haven't been able to find any research on the effects of lower CO2 concentrations.

For the DIYers, it's simple to get an SCD4x sensor and hook it to a pi, arduino, ESP32 etc (then use CC to create a live web interface). I did this after trying an Inkbird CO2 monitor, which gets reasonable scores in reviews and wanting a live web report in the office.

Interestingly the Inkbird and the SCD4X quite often diverge by anything up to hundreds of PPMs; I kind of back the SCD4x (on a Pi in my case) for accuracy after lots of experimentation, reading the datasheet and ensuring the correct calibration procedurs are followed (basically expose the sensors to outside air once a week).

It's also interesting how much it varies day to day in my one-person office - possibly down to how windy it is outside, even with windows closed one day it never goes about 800ppm, other days it'll hit 1500ppm by lunchtime if I don't open a window.

N.B. Quite possible the Inkbird uses an SCD4x internally, seems reasonable kit so I have no explanation for the differences in readings.