This preview shows page 1. Sign up to view the full content.
Unformatted text preview: panel lights up, and the car goes on high alert. It vibrates the steering wheel if needed to rouse her. Zelinsky started a company, Seeing Machines, to commercialize such gaze monitors, and Volvo and others have tested them in cars and trucks. c d E YES ON THE ROAD: Because Alice is looking away at a moment when the
demands on her attention are high, the system emits an alert sound through the car speakers and ﬂashes a sequence of lights on the dash that draws her attention back to the car ahead. Tests by Volvo have shown this method to be effective. e PETER HOE Y N OT NOW, PLEASE: T he onboard computer has a model of the driver’s abilities and is always monitoring the cognitive load, so it knows that now is a bad time for a phone call or a navigational alert. The computer turns off the ringer on the cell phone and instead lights an unobtrusive “incoming message” button on the steering wheel that Alice can press to take the call when it is safe to do so. The car likewise lights another button to let her know that the navigation system has an instruction for her. Chrysler has built such gentle notiﬁcation devices into some of its prototype cars. w w w. s c ia m . c o m COPYRIGHT 2004 SCIENTIFIC AMERICAN, INC. SCIENTIFIC A MERIC A N 57 A N AUTOMATED PERSONAL RECEPTIONIST
Notiﬁcation Platform, a prototype system developed by Horvitz and his co-workers at Microsoft, performs triage on incoming communications much as a receptionist would. The system, which runs on a central server, analyzes messages to decide whether, when and how to notify the recipient of their arrival. It would handle the same e-mail message to Alice from her boss differently if she were conducting a job interview (scenario 1), reading e-mail (2) or packing for a business trip (3). One part of the system estimates the value of the message and how that value decays with time. A second part uses sensors and a statistical model to guess the focus and intensity of Alice’s attention, what devices she is using, how likely she is to see the message without an alert, and how these variables will change in the future. A third part decides what kind of alert to issue. ERIC HORVITZ (left) pioneered the use of sensors and statistical models (above) to build attention-aware communications systems. Redmond, Wash., Bestcom silently handles one call after another. First it checks whether the caller is listed in his address book, the company directory, or its log of people he has called recently. Triangulating these sources, it tries to deduce their relationship. Family members, supervisors and people he called earlier today ring through. Others see a message on their computer that he is in a meeting and won’t be available until 3 P.M. The system scans Horvitz’s and the caller’s calendar and offers to reschedule the call at a time that is open for both. Some callers choose that option; others leave voice mail. E-mail messages get a similar screening. When Horvitz is out of the ofﬁce, Bestcom automatically offers to forward selected callers...
View Full Document