Blustein, J., I. Ahmed, & K. Instone. An evaluation of look-ahead breadcrumbs for the WWW. Proceedings of the sixteenth ACM conference on Hypertext and Hypermedia, pages 202-204 (2005).
This study evaluated look-ahead breadcrumbs, a new type of navigation that presents at each breadcrumb item node a list of links to webpages that are reachable from that node. The study showed that users actually preferred the look-ahead navigation, and that this type of breadcrumb had no detrimental effect on navigation.
Bowler, D., Ng, W., and Schwartz, P. (2001). Navigation bars for hierarchical websites. Retrieved 4/23/2006 from University of Maryland, Student HCI Online Research http://www.otal.umd.edu/SHORE2001/navBar/index.html
Bowler et. al. proposed a modified version of breadcrumb navigation called "selection list navigation bars". Instead of only linking to a page one level up in a hierarchy, or to a page previously visited, pull-down menus are presented at each node, separated by the standard > delimiter. Each pull-down list links to a visited point in the site hierarchy AND that page's siblings. In a study conducted by the authors, participants were required to complete ten tasks using one of three systems: one with no breadcrumb, one with a standard breadcrumb nav bar, and one with the selection list nav bar. During the first five questions, participants performed poorly, but by the last five questions, they were completing tasks with greater efficiency and less page loads. Participants rated the selection list navigation lowest in subjective comfort, but noted it would be better with more practice.
Colter, A., K. Summers, & C. Smith. Exploring User Mental Models of Breadcrumbs in Web Navigation. University of Baltimore School of Information Arts & Technologies independent research conducted 2002. Retrieved 4/22/06 from http://www.angelacolter.com/site/breadcrumbs/index.html
Researchers surveyed 3,453 commercial sites to discover trends in breadcrumb usage, and subsequently explored user mental models of breadcrumbs. 17% of the sites used breadcrumbs. 47% of those sites used the > character for element separation, 11% used a colon, nearly 9% used forward slashes, and the rest used other conventions. 14 subjects were selected to complete a set of scripted tasks at several sites. 13 of 14 participants used breadcrumbs at least once, and overall, breadcrumbs were used in 22% of the 175 tasks assigned. Users' understanding of the purpose of breadcrumbs was high, although it is worth noting that 5 of 14 users confused the utility of location breadcrumbs with path breadcrumbs, stating they thought they indicated where else in the site they had been.
Hudson, W. Breadcrumb navigation: there's more to hansel and gretel than meets the eye. Interactions 11(5): 79-80 (2004).
Hudson outlines reasons why breadcrumbs are good for navigation, namely: they enable users to stay in a general area of interest without backing out to the homepage, it is otherwise hard to navigate backwards, and they help users understand a site's hierarchy.
Hull, S. Influence of Training and Exposure on the Usage of Breadcrumb Navigation. Usability News (6.1), 2004. http://psychology.wichita.edu/surl/usabilitynews/61/breadcrumb.htm.
In this study, 29 participants were tested in three controlled groups to determine the impact minimal training and exposure had on influencing usage of breadcrumb navigation, and whether breadcrumb usage improved efficiency in finding items to purchase on the Wal*Mart website. Group 1 received both exposure and instruction about the breadcrumbs, group 2 received only exposure, and group 3 was not given exposure nor instruction. Group 1 completed the same set of tasks in significantly less time with fewer page views than groups 2 and 3.
Instone, K., 2002. Location, path and attribute breadcrumbs. Retrieved 4/22/06 from http://user-experience.org/uefiles/breadcrumbs/KEI-Breadcrumbs-IAS.pdf.
In his poster presentation, Instone defines three types of breadcrumbs: location, path, and attribute. He sets a research agenda for more fully understanding the roles of breadcrumbs on the Web.
Krug, S. Don't Make Me Think: A Common Sense Approach to Web Usability. New Riders Publishing (2000).
Krug (77) asserts that good breadcrumbs are "self-explanatory, they don't take up much room, and they provide a convenient, consistent way to do two of the things you need to do most often: back up a level or go Home." He recommends using breadcrumbs as an accessory to complement other solid site navigation, particularly when a site is large and sub-sites need to be tied together.
He also indicates best practices (78): put them at the top, use small type, use the words "You are here," boldface the last item, and use > as the delimiter character. The greater than symbol "seems to be the most satisfying and self-evident--probably because it visually suggests forward motion down through the levels."
Lazar, N. and Eisenbrey, M. Website Structural Navigation. Retrieved from University of Maryland, Student HCI Online Research. Spring 2000. http://www.otal.umd.edu/SHORE2000/webnav/index.html.
This study required 67 participants to complete information-seeking tasks on simulations of three different websites (C|Net, Yahoo!, and AltaVista). The independent variable was the inclusion or exclusion of a breadcrumb nav bar to aid navigation. Researchers found that those who were provided with the breadcrumb navigation were able to complete tasks in 60 seconds less time, on average, for each task.
Lida, B., Hull, S. and Pilcher, K. Breadcrumb Navigation: An Exploratory Study of Usage. Usability News (5.1), 2003. http://psychology.wichita.edu/surl/usabilitynews/51/breadcrumb.htm.
In this study, researchers evaluated participants' navigational strategies, satisfaction, and preferences when seeking information from two sites that feature breadcrumb navigation: Google Directory (http://directory.google.com) and Office Max (http://www.officemax.com). Out of 72 participants, 57 recalled seeing the breadcrumb path on Google, but only 29 remembered seeing it on the Office Max site. However, participants used the breadcrumb tool more frequently on the Office Max site than they did on Google. When comparing those who used the breadcrumbs and those who only used other navigational strategies, there was no significant difference in efficiency to complete information-seeking tasks.
Nielsen, J. Is Navigation Useful? Alertbox. January 9, 2000. Retrieved 4/22/06 from http://www.useit.com/alertbox/20000109.html.
Nielsen says breadcrumbs help users interpret a product, seeing which family to which it belongs, and also allows them to navigate easily to a broader page higher in a hierarchy. Users often arrive directly at pages through searches or other means by which they do not simply start at a homepage and drill down. Nielsen claims that breadcrumbs help to contextualize pages for users.
Nielsen, J. Why This Site Has Almost No Graphics. Useit.com. Retrieved 4/22/06 from http://www.useit.com/about/nographics.html.
Nielsen describes the breadcrumb navigation he utilizes in his own Useit.com website. He originally used a colon as a separator, though changed it to an arrow, as his users were confused by the colon. According to Nielsen, "The arrows seem to be slightly easier to understand as an indication of moving deeper and deeper into the site."
Rogers, B.L. and Barbara Chaparro. Breadcrumb Navigation: Further Investigation of Usage. Usability News (5.2), 2003.http://psychology.wichita.edu/surl/usabilitynews/52/breadcrumb.htm.
Rogers and Chaparro attempted to discover whether users choose to use breadcrumbs, whether they improve navigational efficiency, whether their location on a page matters, and to what extent the breadcrumb aids users' mental model of a site. The study took 45 participants through a fictional gardening website and each had to complete 21 information retrieval tasks. Three separate versions of the site were used, the difference being the presence and location of breadcrumb navigation. User mental models were gauged by a multiple-choice question with sample diagrams. Satisfaction was measured by a series of questions on a Likert scale (1-5).
14 of the 30 users (40%) who had breadcrumbs as an option clicked a breadcrumb link 5 or more times in their tasks. However, breadcrumb links accounted for only 6% of total navigation, with users preferring the back bar, the primary nav bar, and embedded links relatively equally. There were "no significant differences between the groups for total pages visited, embedded link clicks, navigation bar clicks, or time to complete the tasks." There was significantly more breadcrumb use when the navigation was located underneath the page title rather than above it. Participants whose systems had breadcrumbs had a better mental model of the site, as almost all in this group correctly identified the site as hierarchical, whereas many of those who did not have breadcrumbs thought the site was non-hierarchical.