Altmann, Trafton, and Hambrick  created a procedure to test the effect of interruptions on tasks that require people to remember where they were before the interruption. The primary task involved engaging in a series of trials on a computer screen. In order to know which rule would apply for each upcoming trial, the person had to keep in mind where they were in the sequence.
This task can be considered to represent cognitive activities whose various steps depend on holding a train-of-thought. For example, in computer programming, each step depends on the previous steps. In writing anything of substance, a writer can’t succeed if she considers each sentence without reference to the sentence or sentences that came before. In most types of intellectual activity, it’s important to keep a series of previous steps in mind when moving forward to the next steps.
In the procedure, participants were trained on the rules and then performed a series of trials using these rules (the primary task). They were also trained and told that at various times between trials, a screen would come up displaying a simply code that they were to type into a box (the interruption). Once they had done this, the trials on the primary task resumed.
In the first experiment, the code to be copied into the box was four characters long; in the second, it consisted of only two characters. So, we’re talking about a very easy interruption task — just copying what they saw — and a very brief interruption, a mere 2 to 5 seconds!
So what happened right after these minor interruptions? People often forgot where they were: Their responses showed that they thought they were at the wrong step in the sequence. Compared to trials that didn’t follow an interruption, people who had to type in a 4-character code tripled their error rate, and even those responding to a 2-digit code made twice as many sequence errors as when there was no interruption.
Being interrupted did not affect participants’ errors in choosing the correct option within whatever step they thought they was next. So the interruption didn’t interfere with their ability to follow the rules; it just interfered with their place-keeping. Their working memory, which is so valuable in keeping relevant information in mind, was bollixed up with those trivial interruptions.
Think about this the next time you’re intently focused on a cognitive task and the phone rings or an alert appears up on your screen. — Or when someone pops into your office saying, “gotta second?” should you say, "OK, one, maybe; but definitely not two!"
Find ways to manage your interruptions by turning off your alerts when you need to concentrate. And when that interruption is unavoidable, try to jot down what you were just up to, so you can re-orient as quickly as possible.
For information on another problem with interruptions -- how much they slow you down, see my earlier post, Mindus Interruptus.
 Klingberg, T. (2009). The overflowing brain: Information overload and the limits of working memory. (Translated by Neil Betteridge). London: Oxford University Press.
 Altmann, E. M., Trafton, J. G., & Hambrick, D. Z. (2013). Momentary interruptions can derail the train of thought. Journal of Experimental Psychology: General, January 7, 2013. Advance online publication. DOI: 10.1037/a0030986.