What's the difference between a concept and a proposition?

[acanas]

Hi,

I don't have as much a problem with the term "concept" as with the "term" proposition, which creates a lot of confusions.
You are right about "have", and that is part of learning to build better maps, which is use more precise linking phrases.
The hierarchical vs. network discussion is interesting. There is a hierarchy of concepts within a domain, and this is the hierarchy that Ausubel refers to and on which Novak based the use of hierarchy, but it doesn't necessary mean a hierarchical topology of the Cmap. Look for example at Cyclic maps... they are not physically hierarchical (http://cmap.ihmc.us/Publications/Resear ... 20Maps.pdf).

[tifialf]

Hi David
I think that the problems you have envised have more to do with basic complexity of language's logic that emerge when one tries to render the speech in cmap form. I have tried to speak of this matter in this research http://cmc.ihmc.us/cmc2008papers/cmc2008-p215.pdf (see in particular the case of “Plastids are found in plant cells only” at pages 5-6. Hollow Bones are part of Birds is surely more precise than Birds have Hollow Bones, because of "have" is too vague, moreover it let to unwanted (?) generalization that all Bird's bones are cave. But if you use "Hollow Bones are part of Birds" in a Cmap you are compelled to use Birds as a subordinate concept and "hollow bones" as inclusive concept, and that could introduce a sort of random development in inclusive relations (I'm not speaking of the graphical setting of the Cmap).
If you refer to simple - basic relations, as A is part of B; A is a (kind of) B; A affects B,
all of them can be easily reverted as: B comprehend/contains/includes A (es. Birds comprehend hollow bones); B includes subclass A; B is affected by A.
Yet there are more complex relations between concepts that aren't easily translated in cmap propositions without loosing the flexibility that they have in ordinary language. In such cases I think that the problem is not about learning, but about the relation between language-meaning-making-changing and concept mapping that isn't enough undisclosed yet and maybe oversimplified.

Alfredo

[dgroos]

Hi Alfredo,

Thanks for your further thoughts. Your comment:

But if you use "Hollow Bones are part of Birds" in a Cmap you are compelled to use Birds as a subordinate concept and "hollow bones" as inclusive concept, and that could introduce a sort of random development in inclusive relations (I'm not speaking of the graphical setting of the Cmap).

confused me. Do you mean that the arrow would need to point from 'hollow bones' towards 'birds'? I don't see that as an issue as I don't use a top-to-bottom structure. But then you say that you aren't speaking of the graphical aspects of the cmap so I don't think I'm right on interpreting what you are saying. Can you elaborate this point?

David

[tifialf]

Yes, this is an issue, independently by the hierarchic - top dow structure.
the problem is that without a prevalent inclusion criteria, the cmap propositions could be regulated only by proximity relations (the same relations that we find in the original text in a precise and wanted sequential order) or simple associative bonds. This could be an index of scarce de-construction and reconstruction of knowledge domains and related conceptual structures. Forcing the use of a subset of general linking phrase types, on the other hand, doesn't warrant by itself that such destructuring-reconstructuring task is made. In this case we could renounce to an important criteria to recognize the re-creation of a system of meanings by the learner. Secondly, we could have cmaps that would be difficult to read (remember that passing from normal text to c-map you renounce to the sequential ordering criteria, provided the network shaped c-map may haven't yust a single reading order. So the inclusion criteria that can be inferred or rendered by the linear text, shouldn't be directly denied to the cmap, in my opinion.

Alfredo