| | When moving through an environment, people unconsciously
build up a mental image, or cognitive map, of that environment. When later
planning a trip or giving directions to someone else, they can mentally walk
through the environment, remembering features relevant to their current task.
This process of building up a cognitive map of a place and using that map
later is called cognitive mapping. This dissertation presents the novel idea of using the cognitive mapping process to teach relationships between data items, called the spatial learning method. By creating a VE where the buildings or rooms represent data items, and the paths between the buildings or rooms indicate the relationships between the data items, visitors exploring the VE would not only be building up a cognitive map of the environment, but also learning the relationships implied by the layout. |  |
 | To investigate the feasibility of the spatial learning method, such a VE
was created. Three studies using this VE were run concurrently on a single
set of 26 participants. The first study investigated whether visitors to
a VE can in fact build up an accurate cognitive map of it, and studied the
effect of the VR display system on the cognitive mapping process as well
as the effect of having provided participants with a map of the VE.
The second study investigated whether data relationships can be inferred
from the cognitive map, looked at the effect of display type and having a
map on learning, and compared learning via the spatial learning method with
that via a conventional lecture (presented to a separate group of 7 participants).
The third study examined the relationship between various psychological factors
(emotions such as enjoyment, interest, and distress, as well as the sense
of presence) and cognitive mapping and learning. | |
|
Study 1 found that while most participants did build up a cognitive map of
the VE used in the study, the maps (as measured by wayfinding scores and
sketch maps drawn by participants) were generally of low quality.
Study 2 showed that the learning of the underlying data set varied greatly
between participants, with some remembering almost all of the data points
and the relationships between them while others could barely answer the most
rudimentary questions about the data set. Study 2 also showed that
participants who attended the conventional lecture performed significantly
better at the learning test than participants who were taught via the spatial
learning method. Study 3 found that the participants who attended the VR
sessions did not experience any more positive emotions that those who attended
the lecture, and also showed that emotions and the sense of presence were
unrelated to both cognitive mapping and learning. Studies 1 and 2 also showed
that using an immersive, head-tracked VR system as opposed to a desktop system
did not affect either cognitive mapping or learning. |  | |
 |
With these findings it is difficult to recommend the use of the spatial learning
method as a teaching tool. However, some of the results are encouraging,
and it may be possible to improve the method at least to the point where
it could be used a teaching aid to supplement conventional methods rather
than replacing them. Download the full dissertation (zipped pdf 1.17 Mb) |