Academic journal article Human Factors

Factors Affecting the Usefulness of Impenetrable Interface Element Borders

Academic journal article Human Factors

Factors Affecting the Usefulness of Impenetrable Interface Element Borders

Article excerpt


Graphical user interfaces (GUIs; e.g., Windows, MacOS, Linux) allow users to interact directly with files and applications within a computer system. In order to select a particular system component (e.g., an application), a user typically would move a mouse to position a cursor over an icon that is associated with the desired system component. Once the cursor has been correctly positioned, the user would press a button on the mouse to select the desired item.

The proliferation of GUIs is at least partially attributable to the intuitive nature of this selection process. However, this process is not always the most efficient. Moving a mouse cursor to a target object takes a finite amount of time. If this movement must be repeated many times to complete a task, then this selection process can substantially slow the work that the user is trying to complete.

For example, consider the following scenario. Many users in the business world are required to interact with the Internet or intranets. During this interaction it is common for users to select the back button on their Web browser in order to go back to the previously viewed Web page. If each of those selections took approximately 500 ins, and a company had 100 employees (who each worked 250 days/year) who made such a selection 500 times a day, then making those selections would require approximately 1700 h/year. If one assumes that the employees are paid a reasonable wage (e.g., $15/h), then these selections alone would cost that company more than $25 000/year in employee salaries. However, if each selection time could be decreased by 250 ins, the overall amount of time to complete those selections would be cut in half, and so would the associated cost. Thus it is imperative that GUI objects be optimized for rapid selection.

This paper presents four studies that examine one way to address this issue. Specifically, these studies identified design guidelines for strategically placing targets on the edge of the display to substantially reduce target selection times.

Optimizing GUI Objects

There are many ways to optimize GUI objects for rapid selection (for a review of the diverse possibilities, see Shneiderman, 1998). Our discussion focuses on some of the manipulations that can be used to optimize the objects of a regularly used (i.e., a familiar) interface.

Once a user becomes knowledgeable about the location of interface objects, visual search times asymptote at a minimal level, and the time required to make the actual selection movement becomes the limiting factor in selection speed (Card, 1982). Thus for GUIs that are regularly used, one must decrease the time required to make selection movements in order to further optimize objects for rapid selection.

In general, two parameters of the GUI object will greatly affect selection speed: the target's placement and its size. Fitts (1954) recognized the effects of these parameters on movement time and formalized the relations in what is now known as Fitts' law. With regard to GUIs (see MacKenzie, 1992, for a review), Fitts' law can be used to predict the direction and magnitude of change in the time it takes to select a target, given variations in target placement and/or target size.

Target placement. One way to decrease selection time involves decreasing the target object's distance from the cursor's starting location. Numerous experiments evaluating computer input devices have demonstrated that selection time decreases as the target's distance from the cursor's starting location decreases (e.g., Card, English, & Burr, 1978; Epps, 1986; Radwin, 1990; Ware & Mikaelian, 1987). Thus, as predicted by Fitts' law, targets closer to the cursor's starting location will be acquired faster than will targets that are farther away.

In addition, task analysis techniques have been developed that facilitate this optimization. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.