Citation

Kortum P, Scharff LVF. Proc. Hum. Factors Ergon. Soc. Annu. Meet. 2009; 53(18): 1244-1248.

Copyright

(Copyright © 2009, Human Factors and Ergonomics Society, Publisher SAGE Publishing)

DOI

10.1177/154193120905301817

PMID

unavailable

Abstract

While large scale changes to web pages have been shown to cause initial difficulties for users when they revisit sites, little work has been done to understand how small changes in web pages may affect user performance. Using eye movement data, this study investigated the impact of adding or removing a single shortcut link on the performance of users who visited a site twice. Participants visited a site and performed a search task, and then revisited the site either immediately or three weeks later and performed the same task. The short cut link was either consistently present, consistently absent, or was inconsistently present across the two sessions. Results indicate that in the consistent conditions, users exhibit roughly the same behaviors (page counts and eye scan patterns) in the second task as they did in the first, although they spent less time reading the body text. In the link-present-then-absent (YN) condition, users searching for the missing link looked at the navigation link area significantly more on their second visit than they did on their first visit and clicked to more pages. But they did not spend more time reading the content on the pages, indicating that they were expecting to recognize the target page when they found it. In the link-absent-then-present (NY) condition, almost half of the participants found the new link; there was an overall significant decrease in search times and a nonsignificant decrease in pages counts (due to those who didn't find the shortcut). Eye-tracking data showed a significant decrease in fixation counts in the content area but not the left navigation. As with the YN condition, this decrease in body text fixations suggests that participants spend less time reading the body text on a second visit due to memory of the target page. Overall these results suggest that users maintain a strong memory of link location and page structure, even after delays of up to three weeks after a single use.

Language: en