An interesting brain-based explanation of why we can only multitask two things at a time, with some fascinating connections to motivation, goals and rewards. The medial prefrontal cortext in each hemisphere deals with calculating rewards, with each lobe pursuing its own goal. If there are more goals, there's no space left so, if we want to multi-task further, we have to time-slice. This carries a big cognitive hit, of course. Some interesting parallels with multi-tasking in computers, in which dual-core systems do much the same thing.

It also gives some insight into how we choose between many options. As it turns out, we don't: we simply reduce them to two. I suspect there are some potential applications of this principle in the design of online learning.