Biomedical Engineering Reference
In-Depth Information
Table 5.1
Examples of VE applications that, in spatial cognition terms, are model-, small-, or
large-scale
Application Environment scale
theme
Model
Small
Large
Design
Cockpit layout
Engine assembly Chemical plant
Control room layout Architecture
Retail shop layout
Training
Close-range naval weaponry
-
Search building
Helicopter crew collaboration
Learn evacuation route
Health
-
Motor rehabilitation Post-traumatic stress disorder
Science
Molecular docking
-
Data visualization
Other
-
Identity parade
Heritage and tourism
Computer games
determine where they wish to move, but it is often non-trivial to make the maneuvers
that are necessary for that movement.
The third category is
large-scale
applications where users travel through a large
environment (e.g., a building, city, forest or dataset) over an extended period of
time, integrating sensory information obtained during their movement to maintain
knowledge of their location in the environment and avoid getting lost. Sometimes,
and as with small-scale applications, maneuvering is non-trivial (e.g., when training
to search a building for the enemy [
7
,
8
]), but typically it is the acquisition of spatial
knowledge that is the greatest navigational challenge.
5.3 Ecological Validity
Experimental (and especially laboratory) studies of navigation use stimuli and tasks
that have been chosen to investigate specific hypotheses, and are sometimes simpli-
fied to an extreme. To assess the relevance of an experiment's findings, it is important
to balance the generality of those findings with the ecological validity of the stimuli
and tasks for a given type of VE application. In particular, attention should be paid
to the VE's scale, extent and visual scene, the paths users follow during navigation
and how frequently they follow them, and how users' knowledge is assessed.
AVE may be model-, small-, or large-scale in spatial cognition terms (see above).
The cognitive processes involved for navigation in each of these differ substantially,
as does the difficulty of and time required for users to acquire accurate spatial knowl-
edge. For example, in a few minutes users can learn an environment's layout from
a map (the map is effectively a model-scale representation of the environment), but
such knowledge takes orders of magnitude more time to learn by direct navigation
in the environment itself, which is large-scale [
9
], although knowledge gained in the
latter will be ultimately more detailed. Thus, particular caution should be taken when
