Using Artificial Landmarks to Improve Revisitation Performance and Spatial Learning in Linear Control Widgets

Abstract

Linear interface controllers such as sliders and scrollbars are primary tools for navigating through linear content such as videos or text documents. Linear control widgets provide an abstract representation of the entire document in the body of the widget, in that they map each document location to a different position of the slider knob or scroll thumb. In most cases, however, these linear mappings are visually undifferentiated - all locations in the widget look the same - and so it can be difficult to build up spatial knowledge of the document, and difficult to navigate back to locations that the user has already visited. In this paper, we examine a technique that can address this problem: artificial landmarks that are added to a linear control widget in order to improve spatial understanding and revisitation. We carried out a study with two types of content (a video, and a PDF document) to test the effects of adding artificial landmarks. We compared standard widgets (with no landmarks) to two augmented designs: one that placed arbitrary abstract icons in the body of the widget, and one that added thumbnails extracted from the document. We found that for both kinds of content, adding artificial landmarks significantly improved revisitation performance and user preference, with the thumbnail landmarks fastest and most accurate in both cases. Our study demonstrates that augmenting linear control widgets with artificial landmarks can provide substantial benefits for document navigation.

Publication
Proceedings of the 5th Symposium on Spatial User Interaction, SUI 2017. 48-57.
Md. Sami Uddin
PhD Candidate

The Interaction Lab, University of Saskatchewan, Canada.

Dr. Carl Gutwin
Professor

The Interaction Lab, University of Saskatchewan, Canada.

Dr. Alix Goguey
Associate Professor

Université Grenoble Alpes, France.