Geography Reference
In-Depth Information
generally fewer errors made with the MLD (7.5 % as opposed to 8.1 % with the
SLD), making it appear more effective.
There was one marked exception to this general pattern, Trial 24. Errors in this
trial alone accounted for over 27 % of the total number of errors in the MLD
condition). In this case, participants using the MLD made ten times more errors
than users of the standard display. As illustrated in Fig.
6
, the incident was located
mid-way between the Mt Wellington and Howick ambulance stations. Both sta-
tions were an equal straight-line distance away from the incident, but the inter-
vening harbour and the road system meant that the Howick station was much
further on the network, something not widely picked up by the participants. The
reported plausible explanation for this was ''attention tunnelling'' (see
Sect. 1.3
for
more discussion on this).
6 Discussion
6.1 Proximity Compatibility Principle Strategies
6.1.1 Spatial Proximity
Spatial proximity (Wickens
1992
) is perhaps the most obvious way of promoting
PCP and suggests that display proximity can be increased by simply increasing the
spatial proximity of information, i.e. relocating associated information together.
In a geographic context, to not uphold display proximity to represent spatial
proximity would be unthinkable (as with the original configuration of display in
Case Study 1). To do otherwise would be to reduce any cognitive benefits derived
from having the display arranged like the world it represents. It can be seen that
the world behaves in such a way that things situated closer together are more
related than those further apart—this is an expression of the First Law of Geog-
raphy (Tobler
1970
). Geographical displays, like the real world they represent,
should follow this rule, so much so that to come across a display that does not,
makes the mismatch with the real world immediately apparent. However, in
Geographical Information Science as a whole, the agreement of the more abstract
database table index arrangement with the real world it represents is a cutting-edge
issue (van Oosterom
1999
): ''closeness'' in the real world still does not mean
''closeness'' in the database. Therefore, the link between cognitive processing
proximity and display or spatial proximity is a pertinent issue from a geographical
viewpoint.
Caveats to this convention would include accounting for the acknowledged
distortions between the human's mental spatial model and the portion of the real
world that one inhabits, and its surrounding area (Bunge
1962
). The mental
geographical model is (in the main) faithful to the real world that the person
knows, therefore it is natural to use display proximity to attract the attention of the
