A simple demonstration of the inefficiency of task switching. Time yourself writing the letters of the alphabet in order from a to z followed by the numbers 1-26. Then try the same task only instead switching from numbers to letters, like this a,1,b,2,c,3, etc. Compare the two times. They should be the same, but the second version takes much more time for most people. It seems that whether you sit next to a giant bookshelf or put your phone in your pocket and occasionally think of the information either contains the effect on the efficiency of whatever you were trying to process would be the same.
On the flip side my smart phone often allows me to look up needed information about something I'm currently thinking about or working on. So I can stay in the same task instead of noting it and having to come back to it later after I've looked it up from a laptop or in a library prior to the modern internet.
I think the effects of a smart phone on someone varies greatly depending on how exactly the individual uses it.
How could one design a test for the last statement above (which resembles a hypothesis)? In practice it seems the statement "varies greatly depending on how the individual uses it" equates too easily with "varies greatly depending on the individual" (fundamental attribution error). Also, it may be that we believe we are very good at using something to our advantage when in fact we should not be using it at all (casinos exploit this problem). The reality is that it is impossible to split test oneself to discover whether in fact there was a benefit, although I wish this were possible. Maybe try something very difficult in two scenarios: play a game of skill which may benefit from looking something up (like chess openers) several times without a phone and several times with access to the phone to see whether there is a difference in performance against a continuously adaptive algorithmic opponent.
