SatNav or SatNag? - Human-Computer Interaction and Innovation in Handheld, Mobile and Wearable Technologies - page 318

they saw or processed the relevant road sign/cue.
In these cases, it is possible that characteristics
of the system user-interface disrupted drivers'
normal allocation of attention.
Issues concerning reliability will be a rich area
for future research and will be relevant to a wide
range of in-car computing scenarios (e.g. colli-
sion warning and avoidance systems). Whilst it is
likely that the reliability of these in-car computing
systems will increase with customer demand, it
is unlikely that they will ever be 100% reliable.
Importantly, research from other application do-
mains (e.g. process control) indicates that people
find it particularly difficult to calibrate objective
with subjective reliability when systems are close
to perfect (Wickens et al., 2004).

response, I would note the following advantages
for individuals who possess a well-formed cogni-
tive map of an environment:

• Enhanced navigational ability: such
people are able to accomplish navigation
tasks with few cognitive demands based
on their own internal knowledge. Indeed, it
should be possible in certain environments
(e.g. one's home town) to navigate using
automatic processing, that is, with no con-
scious attention.
• Increased flexibility in navigation be-
haviour: informed individuals have the
capacity to choose and then navigate nu-
merous alternative routes to suit particular
preferences (e.g. for a scenic versus ef-
ficient route), or in response to unantici-
pated situations (e.g. heavy traffic, poor
weather).
• Social responsibility: a well-formed cog-
nitive map provides a wider transport ef-
ficiency and social function, since it em-
powers a person to navigate for others, for
example, by providing verbal directions as
a passenger, pedestrian, or over the phone,
sketching maps to send in the post, and so
on (Hill, 1987).

Reliance

A further underload issue has emerged from on-
going research in this area and concerns drivers'
long-term dependency on navigation systems.
Specifically, I have argued with others that current
technology automates core aspects of the naviga-
tion task, including trip planning (where the user's
role is essentially to confirm computer-generated
routes) and route following (where users respond
to computer-generated filtered instructions)
(Adler, 2001; Burnett and Lee, 2005; Reagan and
Baldwin, 2006). As a result, drivers are largely
passive in the navigation task, and consequently,
fail to develop a strong mental representation of
the space in which they are travelling, commonly
referred to as a cognitive map. Several empirical
studies have demonstrated this effect for drivers
(Jackson, 1998; Burnett and Lee, 2005). Indeed,
there have also been recent studies raising this
concern for hand-held pedestrian navigation
systems (Young et al., 2008).
A cynical reader at this stage might remark:
why does this matter? Surely, it is positive that
drivers (and pedestrians) do not have to think too
hard about navigating and are able to make “best”
use of the technology they have purchased. In

This is essentially a complex trade-off problem
which has not so far been addressed in research
activities. Notably, there is a conflict between the
need to design user-interfaces which enable an
individual to acquire spatial knowledge ( active
navigation) and those which minimise the demands
(or workload) of navigating ( passive navigation).
Figures 2 and 3 are adapted from an original graph
by Burnett and Lee (2005) and hypothesise these
relationships for an active compared to a passive
navigation system.
Figures 2 and 3 make it apparent that, with
active navigation methods, task demands are
initially high, but as exposure to the environment
continues and spatial knowledge develops, de-