We see this in information technology, where these issues have been worked on for many years now. The traditional distinctions, in increasing order of technical complexity (and corresponding user capability) are: single tasking, task switching, multitasking, and parallel processing.

In single tasking, only one thing is being done. In order to do another, the first task must be “put away”, likely to it’s permanent home (storage). This is almost always a manual step too.

In task switching, only the forground task gets any processor time. Background tasks exist (versus single tasking, where there’s no such thing), but a background task is completely inert. The background is a temporary place to put a task, from which it can more quickly be revived by bringing it to the foreground. Again, in every case I can think of, this is a manual step.

In multitasking, both foreground and background tasks get processor time and are therefore “active”. Scheduling algorithms typically chose to give the foreground task more time than background tasks, but both can accomplish work. However, and this is important, there’s no need for true simultaneous activity. Multitasking computers have been around a long time and have no absolute need for parallel work. The computer accomplishes this bit of magic by extremely rapid, entirely automated task switching. The appearance of several things happening simultaneously is just that–appearance. Human perceptual limits mean that we cannot perceive that everything is in fact happening serially (I’m deliberately ignoring I/O buffering, which does allow a very limited amount of concurrent activity).

Then you get to parallel tasking. In this architecture the compute system actually has multiple processing resources and can truly do 2 things at exactly the same time. The latest CPU’s from Intel and AMD have multiple cores that not only provide this capability, but do so at an extremely reasonable price, cheap enough for the majority of PC-type computers to come with this as standard equipment.

The brain is an interesting case. My understanding is that at the micro level, it is massively parallel, and there’s no way (not that you’d want to) turn that off. However at the macro level, the level of consciousness and awareness, it’s single tasking (oops, task switching).

This leads to the phenomenon of people being less safe while driving and talking on the cell phone–it takes a person a finite amount of time to switch tasks. Their attention can be on the driving or the cell conversation but not both simultaneously. Faster switch times certainly help but do not entirely resolve the competition for the person’s attention.

I’m thinking about such things as driving or playing a particular song on the piano — that perhaps the brain does still push some functions down to areas of the brain or central nervous system (or that those parts of the brain/CNS become efficient enough to not need much involvement by the cortex). When I first learned to drive, it took everything I had to make sure I was getting it all right: scanning where I’m going and the traffic situation, shifting, traffic rules, etc. Now, the physical act of driving seems automatic and I can talk on the phone (hands-free, of course :) ) and only need to focus on driving when traffic gets interesting. Likewise, when I’m learning a new song on the piano, it’s the only thing I can do. One that I’ve been playing for years I can play while chatting, thinking about other things, etc.

What about something even more fundamental like walking. I’m sure a baby is burning up the upper level neurons to balance and put one foot in front of the other. I like to think the walking is pretty automatic (autonomic?) as I chew gum and maintain a conversation while walking.

And that whole difference in brains between the sexes comes from sexist culture.