Login

Cognitively-inspired direction giving

Show full item record




Title: Cognitively-inspired direction giving
Author: Look, Gary Wai Keung, 1978-
Other Contributors: Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science.
Advisor: Howard E. Shrobe.
Department: Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science.
Publisher: Massachusetts Institute of Technology
Issue Date: 2008
Abstract: Online mapping services and portable GPS units make it easy to get very detailed driving directions. While these directions are sufficient for an automaton to follow, they do not present a big picture description of the route. As a result, while people can follow these detailed turn-by-turn directions, it can be difficult for them to actually comprehend where they are going. Our goal is to make such directions more comprehensible. Our approach is to apply findings from human spatial cognition, the study of how people conceptualize and organize their knowledge of large-scale space, to create a system that generates written route overviews. Route overviews provide a big picture description of a route, and are intended to supplement the information in turn-by-turn directions. Our route overviews are based on cognitively-inspired design criteria such as: the use of spatial hierarchy, goal-directed descriptions, selective suppression of detail, and the use of the trunk segments and cognitive anchor points along the route. In our experiments, we show that we can make directions more comprehensible independent of the particular places a person knows - by using what we know about how people think about space to structure the way we present spatial information.
Description: Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2008.Includes bibliographical references (p. 133-140).
URI: http://hdl.handle.net/1721.1/44415
Keywords: Electrical Engineering and Computer Science.

Files in this item

Files Size Format View Description
Preview, non-printable (open to all) 41.11Mb PDF View/Open Preview, non-printable (open to all)
Full printable version (MIT only) 41.11Mb PDF View/Open Full printable version (MIT only)

This item appears in the following Collection(s)

Show full item record

Search DSpace@MIT


Advanced Search

Browse

My Account

Links