Re: [ontolog-forum] Ontologies about social and psychological issues

[Simon Spero <sesuncedu@xxxxxxxxx>]

On Oct 11, 2013 12:19 PM, "John F Sowa" <sowa@xxxxxxxxxxx> wrote:

> The new methods have revolutionized cognitive science.  They make it respectable to define ontologies and scientific theories that refer
> to previously unobservable processes.
[...]
> But the implications for cognitive science are clear:  neural processes in humans and other animals are observable.  That makes it legitimate to use existential quantifiers to refer to neural processes and to their implications and correlations with all forms of behavior -- including language.
[...]
> That kind of evidence is important for ontologies about intentions, emotions, and other social and psychological issues.

None of this evidence makes it any more legitimate to require admitting the entities of naïve/folk psychology into ones Ontology. Conversely, none of this evidence prohibits one from acting as if such entities exist if they allow one to make simple explanations and make useful  predictions.

It seems rather unlikely that the brain parses sentences through a backtracking process of unifying hierarchical typed feature structures. However that does not mean that PSGs / SBCGs are incapable of making predictions about acceptability or possible semantics.

Learning theory, neuroscience, biochemistry, molecular biology, and evolutionary theory can rule out many hypotheses (e.g. a specially evolved "language" organ).

Neuroscience can support more specific hypotheses; for example, the two-stream model of visual processing. It can also give evidence for certain areas of the brain having been selected to be good at specific kinds of functions (e.g. the FFA is very good at learning to distinguish the faces of conspecifics in primates).

Theories of language evolution and the evolution of language guided by neuroscience and ANNs have given some interesting results, but the models are so simplified that extrapolations are taken only cautiously. Models that show that languages that are difficult to learn tend to lose those features, and that show some resemblance to creole formation/regularization are suggestive of possible neural patterns.

fMRI studies have shown that the rate of blood flow in the brain is not constant and uniform over time.

They seem to show that many brain regions are activated during a task, and that many are activated for many different tasks. The finer the image resolution, the more distributed the processing.

They have also shown that journalists (and people) like pretty pictures, and that the New York Times would be more accurate if it followed the Brain and Behavioral Science model of publication.

Studies have shown that adding spurious neurological explanations to arguments appear to increase the credibility of the argument, and that fMRI images have the strongest biasing effect.

For discussion and links to some important papers, see e.g.:
http://cedarsdigest.wordpress.com/2012/11/26/neurobabble-inflated-credibility-currency/

For an interesting article by Elman on the existence of the traditional metal lexicon see e.g. http://crl.ucsd.edu/~elman/Papers/Mental%20Lexicon2011.pdf

_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/  
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/  
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/ 
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J    (01)