It’s a typical afternoon in the Emergency Room. The attending physician orders an MRI for the new patient who came in with a head injury. The pulmonologist who’d been previously called on a consult has just returned the call and is waiting on hold, a staff member needs a release signed, and the blood results for an unconscious patient with spiking fever have just arrived. Four additional patients in various rooms around the unit are awaiting further care.

Many ER patients cannot be treated all at once; a test result must come back, or treatment must be given before a patient can be evaluated further. Solutions are probabilistic and thus require monitoring, and priorities shift on a moment’s notice. The time-course of availability of information and resources varies. An obvious and intuitively appealing assumption is that interruptions under circumstances like these are bad, and you would think (as the professional and research communities typically have) that the aim should be to eliminate them.

Indeed, the evidence is overwhelming that interruptions come at a cost – in time, errors, and sometimes lives. Previous research on this issue by cognitive psychologists and human factors researchers has concerned irrelevant interruptions and their consequences for remembering to perform a planned task. This type of goal, known as prospective memory, involves activating an intention at the appropriate time, and such activation is easy to disrupt^{i, ii, iii}. Costs have been noted in a wide range of settings, both for college students performing novel tasks and for experts performing aeronautical, medical, and other well-practiced actions.

Certainly, irrelevant interruptions that pull us away from an intended goal are annoying. But, despite the fact that prior research has focused exclusively on them, they aren’t the only type of interruption that commonly occurs. In an environment that requires multi-tasking, like the ER, interruptions to one’s current actions might refer to one’s other, temporarily suspended goals. In this case, the interruptions may serve as critical and well-timed reminders to return to an unfinished task.

The work reported on this site investigates the impact of relevant interruptions on performance. We consider the possibility that the optimal goal in the ER and similar environments may be not to eliminate interruptions, but to expertly manage them using well-trained personnel who understand how to quickly prioritize incoming tasks. Interruptions are certainly not always helpful. But in the ER, where they often signal the appropriate time to return to a temporarily inactive task, limited working memory and the inherent uncertainty of the work environment may render them necessary.