Pro cyclists asked to draw a bike in 15 seconds: https://www.youtube.com/watch?v=6hXwbgio5cU

Wonder how their mechanics would have done.

I think the only reason I could draw a bike with any accuracy is because my dad taught me how to fix it. Otherwise it would be a complete black box.

Results visible at 0:55, 1:16, 1:52, 2:15 (correct), 2:41 (badly visible), 3:04 (proportions wrong in a way that doesn't work anymore, but somewhat correct), 3:29, 3:59, 4:21 (correct), and 5:03 (mostly correct, might have been correct if he had taken the time to properly connect the line to the point instead of slightly next to it).

This is exactly what was done in the paper[0] the drawings came from. And, actually, it was the chain of the bike, not the frame, that had most errors. The results are also broken down by gender, and are interesting. For male non-cyclists the error rate was about half that of female non-cyclists. For cyclists, the males had almost zero errors but female error rates were pretty much unchanged (the chain position errors dropped, but were still higher than male non-cyclists).

This (probably?) says a lot about the way men and women think...

0. https://www.liverpool.ac.uk/~rlawson/cycleweb.html

I'll bet many of said cars will end up not really having the right proportions for the internals or correct door placement. It will be recognizable as a car, sure.

Getting into the weeds a bit we see this more obviously with drawings of tanks, ships and planes that aren't drawn with a reference to a real one - guns too big to fit inside the turret, the turrets themselves being way too big, no consideration for where the engine must be etc.