Navigation

The concept of navigation refers to the interpretation of user actions in hypermedia as a movement through virtual space. Navigation can thus be seen as selective exposure to hypermedia on a micro-level. Hypermedia commonly includes graphics and fragments of audio, video, and plain text (nodes), which are all knotted via hyperlinks. Navigation research is dedicated to the study of personal, situational, and media influences on how users move through hypermedia, select links, search for information, and learn from hypertext contents.

The standard situation while navigating in hypermedia can be characterized as follows: information about the destination of a link is mostly marginal, while structure is decentralized. Moreover, the paths of decision are rather long. At each step, there are a large number of available alternatives. All of these produce low transparency and high uncertainty. On the other hand, users can determine the way information will be processed and the speed at which it will do so, which leads to high situational control. Additionally, it does not require much effort to revise a wrong decision; users only need to click on the “back” button. Therefore, navigation can be described as a low-cost situation (time, money, physical effort). Finally, without the active selection between given alternatives (hyperlinks), the reception process will not proceed. Therefore, navigating hypermedia is associated with high selection pressure (contrary to other forms of media, such as television).

The majority of recent research is based on information-processing theories (mental and situational models and a limited capacity model). Some researchers deploy more specific approaches, like psychological decision theory, problem-solving theory, or text comprehension models. Typical research questions deal with typologizing search strategies and selection behavior, task differences, the effects of the structure of hypermedia on learning and understanding, individual differences, and problems of disorientation and cognitive overhead. Methodologically, one can roughly distinguish between the two types of studies. On the one hand, there are non-reactive web server log studies with large data samples but little insight into the mental states of users. On the other hand, there are experimental or quasiexperimental studies with smaller data samples but extensive information on user strategies, emotions, motivations, and cognitions. The latter studies use quantitative or qualitative questionnaires, observational and think-aloud methods, and, sometimes, physiological measurements. Several researchers try to combine both types of study.

There are several attempts to systemize search strategies and selection behaviors. By emphasizing the differences between single actions or action sequences, formal patterns, such as paths, rings, loops, spikes, and hubs-and-spokes, can be found rather frequently. For example, stressing the structure of a (well-advised) informationseeking process, Marchionini (1995) describes eight stages: (1) recognizing and (2) defining the problem; (3) choosing a search system; (4) formulating a query; (5) executing a search; (6) examining the results; (7) extracting information; and (8) reflecting on the results.

It is expedient to distinguish between different search tasks (Chen & Rada 1996). “Closed tasks” are information retrievals where a single type of information can be found (e.g., name, year, or date). “Open tasks” are retrievals where an undefined amount of information about an issue or a problem can be found and cognitively integrated (e.g., information about a country or historic event). Finally, scan browsing or “serendipitous browsing” is a more explorative navigation with no or weakly defined goal-orientation that enables the user to discover new and surprising information. Results suggest that closed, as opposed to open, tasks are conducted faster and more efficiently. However, it seems that at times types of tasks and task complexity are confounded.

Additionally, navigation depends on the structure of hypermedia, or how the information (e.g., the web pages) is linked. There are linear and hierarchical network structures, as well as several hybrid forms. Hierarchical structures provide some advantages regarding clarity and orientation. However, they hamper explorative and associative browsing and make serendipity effects less probable.

Studies focusing on individual differences in navigation behavior deal with the effects of involvement, motivation, age, gender, domain expertise, Internet (or search engine) expertise, and cognitive styles (Chen & Rada 1996) on navigation behavior. For example, experts in domain knowledge possess more comprehensive and better-structured mental representations of the concepts in the domain. Structures set by novices are less organized and more chaotic. Foremost, novices tend to achieve an overview using breadth-first strategies, while experts are able to locate detailed information by navigating through specific content links (depth-first strategy). In contrast, the differences between more or less experienced Internet users are of little consequence for information seeking or learning success, but Internet experience interacts with domain expertise. On the one hand, missing domain knowledge can hardly be compensated for; on the other hand, double experts (domain and Internet experts) are apparently most successful in searching and learning information on hypermedia (Hoelscher 2002).

Empirical studies identify problems of interaction with hypermedia, particularly disorientation and cognitive overhead. Both contradict assumptions that a structural isomorphism between the structures of hypermedia (nodes and links) and the structure of human memory as an associative network postulates advantages for information processing and learning in a hypermedia environment. Disorientation (often referred to as being “lost in hyperspace”) emerges as a result of navigation behavior in complex hypermedia environments. Usually novices report more disorientation than experts.

Cognitive overhead describes the idea that under the assumption of limited cognitive capacity, users spend cognitive effort navigating instead of comprehending tasks in order to orient themselves (Zumbach 2006). Both problems seem to be more severe for users with low prior domain knowledge and those navigating less structured environments. Disorientation may be reduced, especially for novices, by appropriately using tools like advisements, advance organizers, human support, graphical overviews, and structural cues (Chen et al. 2006). Alternatively, providing learners with adequate content structures will also lessen disorientation.

References:

1. Catledge, L. D., & Pitkow, J. E. (1995). Characterising browsing strategies in the world wide web. Computer Networks and ISDN Systems, 27, 1065 –1073.
2. Chen, C., & Rada, R. (1996). Interacting with hypertext: A meta-analysis of experimental studies. Human–Computer Interaction, 11(2), 125 –156.
3. Chen, S. Y., Fan, J.-P., & Macredie, R. D. (2006). Navigation in hypermedia learning systems: Experts vs. novices. Computers in Human Behavior, 22, 251–266.
4. Eveland, W. P., & Dunwoody, S. (2000). Examining information processing on the world wide web using think-aloud protocols. Media Psychology, 2, 219 – 244.
5. Eveland, W. P., Cortese, J., Park, H., & Dunwoody, S. (2004). How web site organization influences free recall, factual knowledge, and knowledge structure density. Human Communication Research, 30, 208 –233.
6. Hargittai, E. (2004). Classifying and coding online actions. Social Science Computer Review, 22(2), 210 –227.
7. Hoelscher, C. (2002). Die Rolle des Wissens im Internet: Gezielt suchen und kompetent auswählen. Stuttgart: Klett-Cotta.
8. Marchionini, G. (1995). Information seeking in electronic environments. Cambridge: Cambridge University Press.
9. Wirth, W., Böcking, T., Karnowski, V., & von Pape, T. (2007). Heuristic and systematic use of search engines. Journal of Computer Mediated Communication, 12(3), article 2. At http:// jcmc.indiana.edu/vol12/issue3/wirth.html, accessed August 6, 2007.
10. Zumbach, J. (2006). Cognitive overhead in hypertext learning reexamined: Overcoming the myths. Journal of Educational Multimedia and Hypermedia, 15(4), 411– 432.