Information Technology Reference
In-Depth Information
be aware of. It is noted that driving is already a
complex task (largely using the visual sense) and
secondary systems (such as vehicle navigation)
will be associated with divided attention and ad-
ditional information processing (Srinivisan, 1999;
Moriarty and Honnery, 2003). Consequently, there
is potential for drivers to make fundamental errors
whilst engaging with a navigation system, such as
failing to observe in time a lead vehicle slowing
down or wavering out of lane.
Consequently, there is considerable literature
(especially from the 1980s and 1990s) focusing on
distraction effects (visual, cognitive and biome-
chanical) and several influential design guidelines
handbooks have been produced, informed by
research studies with this focus (e.g. Ross et al.,
1995; Campbell, Carney and Kantowitz, 1998).
These handbooks provide a wide range of guid-
ance for designers concerning issues as diverse as
the choice of modality for interfaces, the content
and timing of voice messages, display position,
colour combinations, font types/sizes, orienta-
tion of map displays, and so on. Obviously, such
handbooks can be important source documents for
HCI professionals in industry wishing to argue a
case for a specific user-interface design.
From my detailed knowledge of these hand-
books, and discussions with practitioners within
automobile companies, it is clear that many of
the current user-interfaces for vehicle navigation
systems have been influenced by their content.
In particular, many vehicle navigation systems
largely aim to make the workload associated with
the navigation task low. This is often achieved us-
ing simple turn-by-turn instructions given in the
auditory modality, combined with predominately
arrow-based graphics.
In some respects, this could be argued as a
success for HCI/Human Factors research. Studies
were conducted (often on public roads) to provide
the 'believable' empirical data for guidelines,
which, accordingly have informed best practice.
Unfortunately, however, as a result of the recent
mass uptake of the technology, it has become
evident that overload is not necessarily the most
important concern. Two key issues relating to
underload
have emerged, which can be consid-
ered broadly under the headings of reliability
and reliance.
Reliability
Surveys, in conjunction with considerable anec-
dotal evidence, have demonstrated the problems
associated with unreliable guidance information
from vehicle navigation systems. The resulting
problems have obvious safety implications (e.g.
when a driver turns the wrong way down a one-
way street) and can have a considerable impact
on congestion (e.g. when a lorry gets stuck under
a bridge).
Recent work by Forbes (2008) has examined
this phenomenon in detail and, in a survey of 872
navigation system owners, established that 85%
had received inaccurate guidance. When asked
about guidance that was considered dangerous/
illegal, 23% of respondents admitted to obeying the
instructions on at least one occasion. In addition,
there was a clear relationship with age, such that
older drivers were more likely to follow the un-
reliable guidance than their younger counterparts.
From an HCI perspective, it is most interest-
ing to consider here: a) why certain individuals
are prone to following such instructions, and b)
which characteristics of the user-interface can
contribute to the problem. This is an area around
which there has been very little research to date.
With respect to the former question, Forbes (2008)
conducted follow-up detailed diary studies with
30 navigation system users and used the data to
hypothesise that, for certain drivers in specific
situations, a
trust
explanation could be given.
Specifically, there was evidence for overtrust
(or complacency), that is, they saw the relevant
road sign/cue, but chose to ignore it and favour
the navigation instruction. In other contexts, there
was evidence that an
attention
-based explanation
could be put forward, since drivers did not believe
Search WWH ::
Custom Search