Conceptual Aspects: How Landmarks Can Be Described in Data Models - Landmarks: GIScience for Intelligent Services - page 121

happening on a coarse level of the hierarchy (major roads), an instruction or
description process involving all roads abstracted to linear objects, and the actual
driving process, which involves finer grained levels of a hierarchy, such as lanes and
traffic lights. By the way, such a conceptual model explains the correlation between
major roads and major traffic.
Tomko and Winter [ 58 ] provide a complementary approach to explore the
elements of the city formally, providing a means to computationally approximate
spatial mental representations of urban environments. Only their model is sensitive
to the context of mode of mobility. In their approach objects in the world are classi-
fied into Lynch's elements depending on this mode of mobility. Their model is also
hierarchic by spatial granularity. The instances of Lynch's elements at each level
of granularity are characterized by reference regions. Underlying these reference
regions is what Raper called geographic relevance [ 45 ] , a partition of space such
that each cell contains one relevant object. This relevant object would be used as a
direct relatum to locate a locatum. It is the object anchoring the location, or the local
landmark. The third component of their model are functional relationships .These
are the relationships between the spatial objects that stem from the actions they
afford to a person in some mobility context. For example, for a pedestrian a street
is classified as path, and an intersection as node. If they are topologically connected
their functional relationship is described by verbs of affordance such as “connect”,
“follow” or “cross”. As Tomko and Winter demonstrate, this model allows to switch
between contexts, and to model the acquisition of functional (context-specific)
spatial knowledge. This way they suggest a formal model of learning a mental
spatial representation by locomotion based on cognitive principles.

4.2.4 Formal Models for Place References

A formal model of landmarks needs linking to other developments of cognitively
motivated formal models of spatial objects. One of them, place , has been regarded
in detail in this topic (Sect. 1.2.1 ) . Compatible with our notion of place, one of the
earliest advances to capture the notion of place was by affordance [ 32 ] , recognizing
that place is an entity of assigned meaning, or a meaningful configuration, rather
than a physical object abstracted to an anchor point. Typical affordances of a place
are expressed by the image schemata of container , affording shelter, and surface ,
affording movement. The latter has been formalized recently [ 51 ] . Approaching
place is particularly interesting because it is another fundamental cognitive concept
that is missing in spatial information systems [ 64 ] .
An alternative approach to characterize places can be made by means of
contrast [ 63 ] . The contrast principle is rather cognitive and linguistic, compared
to the perceptual affordance mentioned above. Modelling place by contrast means
explaining the meaning of a (reference to a) place by a contrast set, which also
captures a communication context. For example, Alexanderplatz is a reference
to a place in Berlin. Talking to a tourist about a local attraction the reference