Categorisation, conceptualisation, language and affordances of space
Tina Thomson
Bowerman, M. and Choi, S. (2003). Space under construction: Language
specific spatial categorization in first language acquisition. In D.
Gentner and S. Goldin-Meadow (Eds) Language in Mind: Advances in the
study of Language and Cognition (pp. 387-428). Cambridge: MIT Press.
Bowerman, M. and Choi, S. (2001). Shaping meanings for language:
universal and language-specific in the acquisition of spatial semantic
categories. In M. Bowerman and S. C. Levinson (Eds) Language
acquisition and conceptual development, (pp. 475-511). Cambridge:
Cambridge University Press.
I only found the latter. It is accessible from
http://citeseer.ist.psu.edu/557631.html.
Indeed, language has an enormous effect on human categorisation. Even
if the same language is used: if people under investigation come from
different cultural backgrounds this will result in different
conceptualisations of the same thing. Hence, in my research on the
conceptualisation of land use I asked respondents of their cultural
origins in order to establish if there are any large differences in
their conceptualisations, even though this isn't the focus of my
survey.
The paper you send me on affordances
(http://scholar.google.com/url?sa=U&q=http://www.spatial.maine.edu/~max/place.pdf)
is very useful, despite being more theoretical than practical.
Nevertheless, it highlighted some useful aspects that should be
considered within a model for GIS. Especially the aspect of actions and
functions which result from the object's affordances are interesting. I
assume for you the aspect of affordances in the area of human-computer
interaction is interesting, besides the affordances of data for your
emergency management application in terms of environment, task and user
requirements?
Tina.
Vlad Tanasescu
Thanks for the citeseer ref, will certainly use it and it's not
directly accessible from Google Scholar. Not sure about the references,
Melissa Bowerman just gave the short refs (names and years) in her
keynote at GIScience and just overflew your link yet (but it is about
Korean, English and spatial prepositions so sounds good).
I totally agree with you about the intricate relationships between
language and categorization, but wonder at the implications of this on
the notion of 'thing'. The definition I prefer about ontologies is from
Guarino [1]: "a logical theory accounting for the intended meaning of a
formal vocabulary, i.e. its ontological commitment to a particular
conceptualization of the world" and from what I understand (the paper
being quite technical) a conceptualization is simply the way we think
about a part of the world, i.e. a categorization (but it's difficult
to be sure since Guarino refers to conceptualization also for
philosophical systems which imo are again a different thing). Therefore
if an ontology refers to our conceptualization and not to things
themselves maybe it make sense to talk about things differently and to
follow a bit Kant when he decides that we see everything through the
(rough) sieve of our categories and that we don't have real access to
the exterior/eternal/unique nature of things? It's not just abstract
rambling because getting comfortable with the notion that objects are
not so fundamental helps different, and maybe more useful, definitions
for a thing. For example one based on its affordances, i.e. on its uses
in a given context. Same for the representation of a thing; it seems
difficult to explain that we simply need different representations
depending on different contexts, because we think that an object is an
object, and that of course slight differences in interpretation can
exist but the object is still the same, so we talk about
generalisation, which is basically size, instead of making
representation - an essential aspect of a thing - context dependent
too. So that's what I am aiming for in eMerges [2], which is not only
about emergency management (only that's the only data we have access
to) but about how to navigate through spatial data sources in a
cognitively sound way according to the context.
However I am having the same problem in defining context as the ones
found in the paper of affordances abound finding all the affordances
which can define a place, task qualified as 'formidable' :)... seems
that there are too many parameters and too many directions. There is
not as much work about context in GIS, or I couldn't find much...
Best,
Vlad
[1] http://www.loa-cnr.it/Papers/FOIS98.pdf
[2] http://irs-test.open.ac.uk/sgis-dev/
Tina Thomson
object by differentiating it from the general ontological
conceptualisation of the world. However, with ontologies there seems to
be a mixture of definitions out there depending on the purpose. I mean
some refer to ontologies as an informal conceptual system, a
specification of a conceptualisation, a representation of such a
conceptual system via logical theory, or a hierarchical structured set
of terms for describing a domain that can be used as a skeletal
foundation for a knowledge base, and so forth (dont have the refs at
hand at the moment). I guess all these definitions have in common that
ontologies are used for specifying a conceptualisation, and that's
where the ambiguities start. A conceptualisation can relate to
anything: to maybe an object itself, its properties, its uses, roles,
or affordances, which again can differ according to situations. This is
highly subjective and does depend on the context, i.e. the situation in
which the conceptualisation is being used. Apparently humans have a
strong desire for categorising everything, perhaps yes this is partly
due to our cognitive system what we call our brain. I dont know, I am
not an expert in this area. However, I do agree that we can refer to
the same thing, but mean something totally different at the same time;
equal to different representation of the same thing according to
differing contexts, you are right. From what I understand, in your case
study you want to represent or use only data that is relevant to the
given situation and environment, hence reffering to these as
affordances: a particular situation may afford this or that.
>From a philosophical point of view an ontology deals with the study of
existence, in AI it is mainly concerned with modelling and representing
knowledge of the real world. Regarding the latter the problem lies with
conceptualising from a human perspective, as one pitfall may be the
infiltration of human knowledge with erroneous beliefs [1]. The other
problem is if we model a conceptualisation just purely from a human
perspective, which may be too general or too broad in order to be
linked to the data itself. That's why I want to use a two way approach
in my study for building my ontology, a top-down approach from the
human conceptualisation of a land use domain and a bottom-up approach
from the data itself, because the data itself is already a
representation, or more precisely an objectification of the real world.
And of course the other necessity is to generalise from the subjective
to the objective, as a conceptualisation needs to be true for all of
its representing situations.
So, in my study I want to model context as well, but only the spatial
context within the topographic map with an ontology that describes the
spatial relations and configurations of the topographic features among
other things. I assume in your case context refers to much more, i.e.
that of the user, the emergency situation, and the location. Modelling
context and putting it into a computable form is quite challanging, and
I havent come across anything in the literature yet. Indeed, there are
attempts of calculating configurations within a database [2], but those
remain at a very small number of objects and fail as soon as databases
become complex (probably not so interesting for you). Regarding the
definition of context and context-awareness, I dont know if you have
come across [3]. It is definitly advisable to precisely define context.
That's all from me for now. Maybe some of these thoughts are useful...
I still have a pile of books on my desk which I need to go through,
they are all about the mind, categorisation, conceptualisation, and
cognitive science. Maybe I come across something interesting, and I
will let you know.
Tina.
[1]ontology.buffalo.edu/bfo/BeyondConcepts.pdf
[2]www.inf.udec.cl/~andrea/papers/giscience02_2.pdf
[3]http://scholar.google.com/url?sa=U&q=http://www.it.usyd.edu.au/~bob/IE/99-22.pdf