Cognitive Aspects: How People Perceive, Memorize, Think and Talk About Landmarks - Landmarks: GIScience for Intelligent Services - page 93

in a mental spatial representation, especially in a problem solving process. In-situ
one may experience an object standing out in the environment, a figure on a ground.
Correspondingly, reading a map symbol or graphical structure on the ground of
the paper, or listening to an route description that emphasizes a particular object
for anchoring a left-turn, provides a similar figure-ground experience. Map objects
stand out for a purpose, i.e., by intention of the cartographer. The real-world object
represented by this medium is supposed to be relevant since it survived cartographic
generalization, and it anchors spatial configuration. The same can be said for the
object represented on a sketch, only that the sketch is even more focused on
the current communication context and may provide even more trusted relevant
references. The object referred to in a verbal route description, in the expectation
of the reader, must be standing out in the environment, must be recognizable, or can
even be matched with knowledge of the reader. The reader will expect landmarkness
of these objects in the environment that matches the landmarkness experience in the
reading process.
Research on understanding external spatial representations investigates their
impact on mental spatial representations. In early work Taylor and Tversky, for
example, asked participants in an experiment to read verbal descriptions of an
environment, provided in either a route or a survey perspective [ 212 ] . They did not
find differences in an ability to answer verbatim or inference questions, suggesting
that the participants formed the same spatial mental models capturing the relations
between landmarks from the descriptions. Also, when participants in another
experiment studied maps and then had to draw them and to verbally describe
them, and the order of doing so was studied, no difference in the organization
was discovered, again suggesting the same mental spatial representation—which
evidently also showed a hierarchical structure [ 211 ] . There are, of course, individual
differences of cognitive styles, as demonstrated by Pazzaglia [ 164 ] . More recently,
Lee and Tversky [ 116 ] have studied the comprehension times of the participants
being presented route and survey perspective descriptions of an environment,
enriched with references to landmark objects either of visual detail or of factual
detail. They observed that landmarks are neutral to the perspective: the landmark
descriptions did not increase the comprehension time, neither in route nor in survey
perspectives, but visual landmark descriptions seemed to support the comprehension
of descriptions with perspective switches.
It goes without saying that the way how people understand maps, sketches or
verbal descriptions is relevant for generating these messages. Hence, more recently
research moved into studying the reading of machine-generated spatial informa-
tion, where cognitive science meets human-computer interaction design. However,
cognitive research goes beyond usual methods of interaction research, such as
user satisfaction measures. Instead, the machine-generated information is tested
for comprehensibility and task success. Generally this research is only indirectly
about understanding landmarks, but rather on human spatial conceptualizations of
relationships (e.g., [ 22 , 102 , 117 ] ) and on relevance and communication efficiency
(e.g., [ 33 , 189 , 190 ] ). This type of research will be elaborated in later chapters, in
more appropriate contexts.