The influence of nodes sequence and extraneous load induced by graphical overviews on hypertext learning

Eniko Bezdan, Liesbeth Kester, Paul A. Kirschner

    Research output: Contribution to journalArticleAcademicpeer-review

    2 Downloads (Pure)


    The effects of four hypertext learning environments with a hierarchical graphical overview were studied on the coherence of the node sequence, extraneous load and comprehension. Navigation patterns were influenced by the type of overview provided (i.e., dynamic, static) and whether navigation was restricted (i.e., restricted, non-restricted). It was hypothesised that redundant use of the overview for inducing a high-coherence reading sequence would result in high extraneous load and low comprehension. Coherence was higher in the dynamic than in the static conditions. Coherence was also higher in the restricted than in the non-restricted conditions. Mental effort as a measure of extraneous load was higher at the end than at the beginning of the learning phase, especially in the dynamic restricted and the static non-restricted conditions, although there was no significant interaction. Comprehension was lowest in the dynamic restricted condition and highest in the dynamic non-restricted and static restricted conditions. Low comprehension in the dynamic restricted condition indicates that overviews can become redundant for reading sequence coherence, negatively impacting comprehension. The evidence suggests that severe restriction of navigation paths should be avoided and that continuous use of overviews such as in dynamic overviews may be detrimental to learning.
    Original languageEnglish
    Pages (from-to)870-880
    Number of pages11
    JournalComputers in Human Behavior
    Issue number3
    Early online date17 Jan 2013
    Publication statusPublished - May 2013


    • hypertext
    • graphical overview
    • coherence
    • node sequence
    • extraneous load


    Dive into the research topics of 'The influence of nodes sequence and extraneous load induced by graphical overviews on hypertext learning'. Together they form a unique fingerprint.

    Cite this