Early Retirement Extreme Forums

I know there is research or a book showing that interruptions not only have the direct cost of the time consumed but there is also time wasted when the worker has to get back up to speed in the work the were doing. Does anyone have a source for this? I can't think of the name of the book.

Cognitive switching costs?

This? -> https://www.apa.org/research/action/mul ... ive%20time.

"Meyer has said that even brief mental blocks created by shifting between tasks can cost as much as 40 percent of someone's productive time."

This reminds me of a discussion of interrupt service routines in embedded processing. Not quite what you’re asking for but there is a relationship.

When a processor goes off to service an interrupt routine it may only use a few cycles. But there is something called context saving and context restoration where the programmer needs to basically record where things were before the interrupt routine and then put things back the way they were when done. These are things like interrupt flags, program count and I/o semaphores et al. You basically have to restore things before leaving interrupts or else you can get erratic behavior like looping interrupts. This stuff can actually take more time than the routine itself and is referred to as interrupt overhead, I.e. additional time wasted when interrupted.

Better explanations in this slide set.

https://courses.cs.washington.edu/cours ... rrupts.pdf

I think the same thing happens with humans. If I recall right another programmer used this analogy when he was pissed off about everyone stopping his flow with an occasional question.

Not research, but I keep linking to Paul Graham's essay for this issue
http://www.paulgraham.com/makersschedule.html

i never put this together but if it's true, the constant switching between apps etc. is actually making us stupider in addition to distracted.

Loner wrote: ↑

Mon Jun 29, 2020 10:27 pm

Cognitive switching...

That's it! Thanks everyone, lots of useful ideas here!

I looked into this after reading Newport's Deep Work a couple of years ago. I read several books and papers on the subject.* I found it a little odd that all the authors seemed to base their ideas on Meyer's work. I'm not questioning the validity of the work, but it makes me uncomfortable that it's all based on the same research.

I also found that most new research on anything related to executive function is done on spectrum kids or adults struggling with dementia. While useful, I'm not sure how well that applies to high-functioning adults during their peak performance years. Advice and training for world-class athletes can be vastly different from what's recommended to novices or even hobbyists.

Personally, I find a combination of Graham's maker/manager model with 'me' time built in that includes ART works best.

During the lockdown, I discovered that having too few distractions and 'things on my plate' is just as challenging to my overall production and happiness as having too many. It's only in the last few weeks that I've found my groove wrt maker/manager/me time. It's funny, I've been working on something related to this over the last year but the pandemic/lockdown helped me see the missing piece of it because it kind of flipped the problem on its head. (I was able to finish writing a long piece on how to implement it but it's Graham's idea so it doesn't feel write to publish it, but it helped me clarify my thoughts.)


*search for terms like task switching, cognitive switching, cognitive flexibility, and attention restoration and control