|
Research Review Of Potential Safety Effects Of Electronic Billboards On Driver Attention And Distraction
The following annotated bibliography summarizes three
research papers on the possible distraction effects of systems that are located
inside the vehicle. The references
cited and summarized herein represent a sample of the extensive research
literature on possible driver distraction from in-vehicle systems. This section briefly describes several
studies that were evaluated in the present research effort, but were not
included in the literature review itself.
- Gellatly, A.W., and Kleiss, J.A.
(2000). "Visual Attention Demand Evaluation of Conventional and Multifunction
In-vehicle Information Systems." In: Proceedings of the IEA 2000/HFES
2000 Congress. Joint International Ergonomics Association 14th
Triennial Congress and Human Factors and Ergonomics Society 44th
Annual Meeting July 29, 2000-August 4, 2000.
This paper presents the results of a study of driver
responses to executing conventional in-vehicle tasks and to operating a novel
re-configurable, multifunction information system inside the vehicle. Six older and six younger drivers operated
an instrumented test vehicle along a two-lane divided highway. Visual scanning behavior and driving task
performance were measured while the research participants executed conventional
tasks (e.g., climate control, audio control, and cellular telephone use) and
advanced /future tasks (e.g., e-mail, navigation, audio, voice and video
communications). The results showed
that drivers completed all the tasks by means of a series of 1-1.5 sec glances
to the in-vehicle display/control systems.
Mean glance frequency increased linearly with mean task completion time,
but mean glance duration remained constant over the range of mean task times
observed. Mean task completion times
were about twice as long for the re-configurable, multifunction information
system as for the conventional in-vehicle devices. Certain driving task variables were correlated with mean task
completion time. Both speed variability
and lane variability (peak lateral acceleration) increased linearly as a
function of mean task completion time.
Although variability within the driving lane increased with task time,
lane departure did not. Thus the
research participants tended to stay in the lane, but exhibited more aggressive
lateral position adjustments as the in-vehicle tasks became more complex and
took longer. Overall, the data revealed
a decrement in driving performance as the in-vehicle tasks took more time to
complete. Therefore one design goal for
any in-vehicle display/control device should be to minimize overall task
completion time.
- Rockwell, T.H. (1988). "Spare Visual
Capacity in Driving - Revisited." In: A.G. Gale, M.H. Freeman, C.M. Hasleman, P
Smith and S.P. Taylor (Eds.), Vision in Vehicles II. North Holland:
Elsevier Science Publishers.
This paper summarizes a series of studies on in-vehicle
visual sampling. The data were
collected from 106 research participants in over 200 highway trips ranging in
duration from 45 minutes to 1 hour. The
studies measured off-road glances made by the drivers while completing a
variety of in-vehicle tasks. Over 6,000
off-road glances were recorded. The
in-vehicle tasks consisted of checking the speedometer, adjusting all three
mirrors (left, right and rear-view), adjusting the stereo system (volume and
tuning controls) and using a touch-screen CRT display. While the mean number of glances varied
considerably from 1 glance to 40 glances, depending on the task, the average
glance durations were extremely consistent, ranging from 1.27 to 1.42 sec for
stereos and mirrors. Older drivers
tended to require 20 percent more glances to execute a given command than
younger drivers. Traffic density and
highway geometrics had a substantial effect on average glance duration. While driving in dense traffic at high
speeds on curves with short headway distances, participants exhibited average
glance durations that were 20 percent shorter for both stereo and mirror
tasks. Thus glance durations are
affected more by the requirements of the driving task than by the requirements
of the in-vehicle task. As concerns the
design of in-vehicle display/control devices, good design will be reflected
more by a smaller number of glances than by shorter glances at the device.
- Wierwille, W.G. (1993). "Visual and
Manual Demands of In-car Controls and Displays." In: Smith and Solame (Eds.), Automotive
Ergonomics. New York: Taylor and Francis.
This chapter synthesizes a series of different studies by
different researchers on the visual demands of in-vehicle tasks. Five types of tasks are defined, ranging
from almost automatic manual only tasks, such as setting the directional signal
lever, to complex visual-manual tasks, such as interacting with a navigational
map display. A time-sharing model was
developed to describe how drivers gather in-vehicle information. The model results in a driving strategy
where short glances are made away from the driving task until the necessary
in-vehicle information has been gathered.
These in-vehicle glances tend to be between 1.0 and 1.6 sec. After each
glance, the driver returns to the visual driving task. Compensatory visual sampling strategies have
been demonstrated in studies where an apparatus periodically blocked the
driver's forward view of the road for brief periods. In addition, forward view of the road increases with increasing
traffic, a more difficult roadway or strong crosswinds. As concerns the older driving population,
for a given in-vehicle task, single glance times and the number of glances into
the vehicle both increase with age. The
transition time between the forward view and the in-vehicle view also increases
with age. Several remedies are
suggested to minimize the visual load of in-vehicle tasks. These include the use of heads-up displays,
virtual image displays and displays located near the top of the vehicle
instrument panel. Other remedies
include the use of auditory displays and training on improved visual sampling
techniques. Designers of in-vehicle
displays and controls are urged to minimize the demands of in-vehicle
tasks. Some suggestions include
avoiding clutter and improving labels and legends. Two areas are identified where future research is needed: 1)
expansion and refinement of models of in-vehicle task performance and 2) better
guidelines for in-vehicle task communications.
|
