Item

Abstract

This study examined the working memory systems relevant for wayfinding. 24 participants learned two routes in a novel photorealistic virtual environment displayed on a 220 screen
while performing a verbal, a visual, a spatial or no secondary task. Performance in the secondary
task and in subsequent retracing of the previously presented routes with a joystick was
recorded. Participants without a secondary task performed better compared to participants with
a secondary task. On one route participants with the visual secondary task got lost less often
compared to participants with the verbal secondary task. Better performance in the visual secondary
task was found compared to the spatial secondary task. A trade-off between first and
secondary task could be ruled out. The results indicate that spatial and verbal memory were
used in wayfinding. Despite other results in pre-tests we measured a non-significant higher difficulty
of the spatial secondary task in a baseline condition, providing an alternative explanation
for the importance of spatial memory.
After the experiment we tested the participants’ memory for their local surroundings. In
a choice reaction task presented on a computer screen they had to discriminate pictures of
intersections they had encountered before from distracters. Intersections encountered before
were discriminated faster and more accurate from distracters when the perspective seen was
along the direction of travelling compared to 90 or the opposite direction. The participants
encoded their local environment view-dependent as they encountered it. At the level of large
scale spaces, this result contradicts with encoding spatial information relative to a single reference
direction [e.g. 1]. Pictures taken along the direction of travel were recognised better
and faster than pictures aligned with a reference direction defined by the initial direction or the
main orientation of the environment.
In agreement with studies in reorientation [e.g. 2], verbal memory was used for wayfinding.
These results are consistent with a dual coding approach, which states that spatial information
is also encoded verbally. Local intersections were stored view-dependent as experienced along
with the direction of travel rather than parallel to the initial orientation or the main orientation
of a route.