U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
|This report is an archived publication and may contain dated technical, contact, and link information|
Publication Number: FHWA-RD-95-197
Date: December 1996
Development of Human Factors Guidelines for Advanced Traveler Information Systems and Commercial Vehicle Operations: Comparable Systems Analysis
Human Factors Design Guidelines
The TravTek design process was conducted under the same circumstances as many human factors designs for advanced systems. That is, some design support data were available from the human factors handbooks or guideline documents, but much of the data had to be generated from the review of empirical research, conduct of selected laboratory study, and review of selected on–road testing. The sources of information used to collect or generate the guidelines and information used to design TravTek are described below.
Application of Existing Principles and Guidelines to TravTek
Some of what is known from human/computer interaction research can be applied to drivers' interactions with automotive systems; however, driving is a complex combination of tasks that may be executed in potentially dangerous situations. In this sense, automotive human factors shares much with avionic human factors (Marshak, Kuperman, Ramsey and Wilson, 1987; Roscoe, 1980). However, the automobile user is quite different from the specially selected, highly trained population of pilots. So, it is generally necessary to develop laboratory and simulator research, along with field studies, to better understand driving with advanced displays, as well as cognitively complex tasks such as way finding (Wetherell, 1979; Zwahlen and DeBald, 1986; Streeter, Vitello, and Wonsiewicz, 1985). A primary reason for this research requirement is that many differences among individual drivers and many different driving situations exist, making modeling and generalization difficult (Noy, 1990; Avolio, Kroeck, and Panek, 1985; Pauzie, Marin–Lamellet, and Trauchessec, 1989).
Although the task of driving must be carefully and accurately accounted for in the design of an advanced automotive system, one of the major challenges of the TravTek interface was the pre–drive function design. In essence, the design of the pre–drive interface constitutes nothing more than a human–computer interaction (HCI) design. For this process, two guideline documents were heavily utilized. These were Smith and Mosierð=s HCI guideline document and MIL–STD 1472D. These documents (Smith and Mosier, 1986) provided individual guidelines ranging from the use of color to menu design.
Some of the most important examples of principles applied from these documents included guidelines dictating high legibility and large character size because of the age range of the drivers and, in the pre–drive case, the need for them to quickly glance at and extract information from the screen. For example, the font size is specified as 19.7 arc–minutes. Information is presented on menu and pick screens, map and route guidance screens, and by auditory messages. Color is used conservatively, with low–saturation levels of colors selected from the middle of the spectrum. The principle of consistency was followed in the placement of information, the grouping of information, the use of nomenclature, and coding by shape and color. There is auditory feedback for all key presses so that the user can be sure that the computer has entered each selection.
A major human factors concern for navigation systems in general and for TravTek specifically (Rockwell, 1972; Dingus, Antin, Hulse, and Wierwille, 1988) is that the addition of visually displayed information, especially moving maps, may divert visual attention from primary visual tasks such as lane tracking and obstacle detection. Although drivers may often have spare attentional resources (Dingus et al., 1988), there may be situations in which processing and response demands exceed capacity. This may result in increased mental workload and errors.
In general, principles utilized in addressing these concerns for TravTek included limiting the information presented to the driver to that which was required for a given circumstance (as determined from task analyses) and use of the auditory mode to limit visual attention requirements. Specifically, this resulted in restricting the use of displays and controls that were accessible during driving, reducing the information density of visually presented information, and using auditory tones and messages to augment the visual displays.
Utilizing Previous Research: The ETAK Navigator Studies
The ETAK Navigator (Dingus, Antin, Hulse, and Wierwille, 1986; Antin, Dingus, Hulse, and Wierwille, 1990; Wierwille, Hulse, Fisher, and Dingus, 1989) was comprehensively evaluated in several human factors studies that addressed the attentional demands imposed by the ETAK upon the driver, the effectiveness and efficiency of the navigation system, and driver adaptation behavior to the navigation system. Each of these studies was performed on–road, utilizing an instrumented camera car. These studies revealed that several of the ETAK functions required a high degree of attentional demand compared to other automotive tasks. Despite this fact, the ETAK was found to be a usable and somewhat useful device that could potentially be improved by conceptual and design changes. On the basis of the ETAK studies' results, the investigators recommended several modifications to the driver interface for future systems. The recommended modifications included: automated route selection, simplified information displays, and a path feature for route planning. Each of these recommendations was included as part of the initial TravTek concept.
Preliminary TravTek Research Used to Generate and Validate Guidelines
Recommended modifications of the ETAK Navigator and additional navigation system designs and studies (e.g., Catling and Belcher, 1989) were considered in a preliminary conceptual design of TravTek. This early conceptual design was subjected to several laboratory usability tests (Dingus and Hulse, 1990).
Testing of the "pre–drive" functions revealed that manual route selection and browse map/area traffic scanning were difficult to use. It was the opinion of the investigators that because these functions were of marginal value to the overall system, the functions could be effectively eliminated. Pre–drive testing also revealed that system labeling, nomenclature, and messaging should be allotted substantial testing in order to minimize errors of confusion and/or required instruction. Furthermore, a larger control/display screen was recommended. Or, in cases where a small screen is the only option, a system with somewhat limited function was recommended. For the TravTek system, an automatic route selection feature was included for driver convenience. In addition, the TravTek nomenclature and labeling were subjected to comprehensive usability testing.
Evaluation of the "drive" functions included testing of the visual attention demand requirements. The inclusion of a turn–by–turn guidance map was strongly recommended for future systems as a lower attention demand option. Additional results included: (1) directly providing critical information (e.g., next–turn distance), (2) increasing the salience of critical roadway names through highlighting, and (3) the elimination of street names of secondary importance and/or increasing the lettering size. Each of the above recommendations was included as part of the TravTek driver interface.
Following refinement of the TravTek interface prototype, a series of design studies using naive subjects drawn from the general population were performed to arrive at display screens that were as self–teaching as possible (Carpenter et al., 1991). The display screen designs resulting from these comprehension tests were intended to facilitate user understanding of what to do and what would happen when soft keys were activated. To enhance driver understanding of how to operate TravTek, a Help key was provided on almost all of the pre–drive screens. In addition to using the Help key, the driver could also request an on–screen tutorial or connection to the Help Line at the TravTek Information and Services Center (TISC) via the cellular phone. The operator at the TISC had a system similar to the one in the car and could guide the driver through the process.
The design study process described above was also extensively used in the development of the on–screen tutorial. This tutorial was accessed by selecting "Instructions" from the main menu and was divided into lessons covering all major functions. The driver had the capability of proceeding through the entire tutorial or viewing lessons selectively. The extensive help screens, tutorial instructions, and the TISC Help Line were intended to obviate the need for extensive training sessions and bulky manuals.