<?xml version="1.0"?>
<!-- name="generator" content="blosxom/2.0" -->
<!DOCTYPE rss PUBLIC "-//Netscape Communications//DTD RSS 0.91//EN" "http://my.netscape.com/publish/formats/rss-0.91.dtd">

<rss version="0.91">
  <channel>
    <title>Notebooks   </title>
    <link>http://bactra.org/notebooks</link>
    <description>Cosma's Notebooks</description>
    <language>en</language>

  <item>
    <title>Dynamics in Cognitive Science</title>
    <link>http://bactra.org/notebooks/2004/12/17#dynamics-cognition</link>
    <description>
&lt;P&gt;Since the early 1990s, some people have gotten very excited about the idea
that dynamical systems theory can be used to model &lt;a
href=&quot;cognitive-science.html&quot;&gt;cognitive processes&lt;/a&gt;.  As somebody trained
in &lt;a href=&quot;chaos.html&quot;&gt;nonlinear dynamics&lt;/a&gt;, I applaud this development,
since, if successful, it will enhance my material and academic prospects.
Sadly, when they do things like purporting to explain &lt;a
href=&quot;judgment.html&quot;&gt;decision-making&lt;/a&gt; with a low (8) dimensional model with
no noise, I grow deeply suspicious.  Worse, many of these same people believe
that dynamics gives them an account of cognition which is &lt;em&gt;incompatible&lt;/em&gt;
with traditional models, whether of the (Newell-&lt;a href=&quot;simon.html&quot;&gt;Simon&lt;/a&gt;)
symbol-processing or &lt;a href=&quot;neural-nets.html&quot;&gt;connectionist&lt;/a&gt; sort, and in
fact one which is fundamentally non-&lt;a
href=&quot;computation.html&quot;&gt;computational&lt;/a&gt;.  As somebody trained in the &lt;a
href=&quot;symbolic-dynamics.html&quot;&gt;symbolic aspects of nonlinear dynamics&lt;/a&gt;, and
who uses that math to study the &lt;a
href=&quot;computational-mechanics.html&quot;&gt;intrinsic computation carried out by
dynamical systems&lt;/a&gt;, I have to wonder what they're talking about.

&lt;P&gt;&lt;em&gt;To do&lt;/em&gt;: Find something interesting to say about this by December,
when abstracts are due for the Potsdam workshop
on &lt;a
href=&quot;http://www.ling.uni-potsdam.de/~peter/TandemWorkshop/Workshop2005.html&quot;&gt;Dynamical
Systems Approaches to Language and Symbol
Grounding&lt;/a&gt;.  &lt;strong&gt;Update&lt;/strong&gt;, December 2005&lt;/strong&gt;: Well, I don't
know if what I found to say &lt;em&gt;was&lt;/em&gt; interesting, but you can read the
abstract &lt;a
href=&quot;http://www.agnld.uni-potsdam.de/%7Eallefeld/Workshop/abstracts.html#shalizi&quot;&gt;here&lt;/a&gt;.

&lt;P&gt;&lt;em&gt;See also&lt;/em&gt;:
	&lt;a href=&quot;ai.html&quot;&gt;Artificial Intelligence&lt;/a&gt;;
	&lt;a href=&quot;development-dynamics.html&quot;&gt;Notes on a Lecture on &quot;Origins of
an Embodied Cognition: Moving, Perceiving, and Thinking in Infancy&quot;&lt;/a&gt;;
	&lt;a href=&quot;philosophy-of-mind.html&quot;&gt;Philosophy of Mind&lt;/a&gt;

&lt;ul&gt;Recommended:
	&lt;li&gt;&lt;a href=&quot;http://www.santafe.edu/~chaos/&quot;&gt;James P. Crutchfield&lt;/a&gt;,
&quot;Dynamical Embodiments of Computation in Cognitive Processes&quot;, &lt;cite&gt;Behavioral
and Brain Sciences&lt;/cite&gt; &lt;strong&gt;21&lt;/strong&gt; (1998): 635--637 [&lt;a
href=&quot;http://www.santafe.edu/sfi/publications/wpabstract/199802016&quot;&gt;Online&lt;/a&gt;]
	&lt;li&gt;Chris Eliasmith, &quot;Attractive and In-discrete: A Critique of Two
Putative Virtues of the Dynamicist Theory of Mind&quot;, &lt;cite&gt;Minds and
Machines&lt;/cite&gt; &lt;strong&gt;11&lt;/strong&gt; (2001): 417--426 [&lt;a
href=&quot;http://dx.doi.org/10.1023/A:1017542632653&quot;&gt;Journal link&lt;/a&gt;, &lt;a
href=&quot;http://www.arts.uwaterloo.ca/~celiasmi/Papers/ce.press.attractive.mm.html&quot;&gt;HTML
preprint&lt;/a&gt;]
	&lt;li&gt;&lt;a href=&quot;http://www.phil.cmu.edu/faculty/glymour/&quot;&gt;Clark
Glymour&lt;/a&gt;, &quot;Goethe to van Gelder: Comments on 'Dynamical Systems' Models of
Cognition&quot; [&lt;em&gt;Abstract&lt;/em&gt;: &quot;The 'dynamical systems' model of cognitive
processing is not an alternative computational model. The proposals about
'computation' that accompany it are either vacuous or do not distinguish it
from a variety of standard computational models. I conclude that the real
motivation for van Gelder's version of the account is not technical or
computational, but is rather in the spirit of &lt;em&gt;natur-philosophie&lt;/em&gt;.&quot;
Available as a (boo, his) RTF document, &lt;a
href=&quot;http://philsci-archive.pitt.edu/archive/00000139/&quot;&gt;PHIL-SCI 139&lt;/a&gt;]
	&lt;li&gt;&lt;a href=&quot;http://www.cs.pdx.edu/~mm/&quot;&gt;Melanie Mitchell&lt;/a&gt;
		&lt;ul&gt;
		&lt;li&gt;&quot;A complex-systems perspective on the 'computation vs.
dynamics' debate in cognitive science&quot;, in M. A. Gernsbacher and S. J. Derry
(eds.), &lt;cite&gt;Proceedings of the 20th Annual Conference of the Cognitive
Science Society&lt;/cite&gt; (Cogsci98), pp. 710--715 [&lt;a
href=&quot;http://www.cs.pdx.edu/~mm/cogsci98.pdf&quot;&gt;PDF&lt;/a&gt;]
		&lt;li&gt;&quot;Theories of structure vs. theories of change&quot;,
&lt;cite&gt;Behavioral and Brain Sciences&lt;/cite&gt; &lt;strong&gt;21&lt;/strong&gt; (1998): 645--646
[&lt;a href=&quot;http://www.cs.pdx.edu/~mm/van-gelder-commentary.pdf&quot;&gt;PDF&lt;/a&gt;]
		&lt;/ul&gt;
	&lt;/ul&gt;

&lt;ul&gt;Dis-recommended:
	&lt;li&gt;Marco Giunti, &lt;cite&gt;Computation, Dynamics and Cognition&lt;/cite&gt;
[Actually, there's a lot of good stuff in this book, but I think it's quite
unhelpful &lt;em&gt;on this topic&lt;/em&gt;.  For instance, while he correctly points out
that all computational systems &lt;em&gt;are&lt;/em&gt; dynamical systems, he takes a
ridiculously restrictive view of computation, especially of analog computation.
By his definition, y=sin(x) is an uncomputable function, and I think even y=x,
the identity function, is uncomputable.]
	&lt;li&gt;T. van Gelder, &quot;The dynamical hypothesis in cognitive
science&quot;, &lt;cite&gt;Behavioral and Brain Sciences&lt;/cite&gt; &lt;strong&gt;21&lt;/strong&gt;
(1998): 615--665 [On which the papers by Crutchfield and Mitchell in the same
volume of &lt;cite&gt;BBS&lt;/cite&gt; are commentary.  Frankly the rest of van Gelder's
work, e.g. the book he edited with R. Port called &lt;cite&gt;Mind as Motion&lt;/cite&gt;,
is no more impressive to me.]
	&lt;/ul&gt;
	
&lt;ul&gt;To read:
	&lt;li&gt;Peter beim Graben, &quot;Incompatible implementations of physical symbol
systems&quot;, &lt;cite&gt;Mind and Matter&lt;/cite&gt; &lt;strong&gt;2&lt;/strong&gt; (2004): 29--51 [&lt;a
href=&quot;http://www.mindmatter.de/mmpdf/graben.pdf&quot;&gt;PDF via Dr. beim Graben&lt;/a&gt;]
	&lt;li&gt;Istavan S. N. Berkeley, &quot;What the #$*%! is a
Subsymbol?&quot;, &lt;cite&gt;Minds and Machines&lt;/cite&gt; &lt;strong&gt;10&lt;/strong&gt; (2000): 1--13
	&lt;li&gt;Rick Dale and Michael J. Spivey, &quot;From apples and oranges to
symbolic dynamics: a framework for conciliating notions of cognitive
representation&quot;,
&lt;a href=&quot;http://dx.doi.org/10.1080/09528130500283766&quot;&gt;&lt;cite&gt;Journal of
Experimental &amp; Theoretical Artificial Intelligence&lt;/cite&gt; &lt;strong&gt;17&lt;/strong&gt;
(2005): 317--342&lt;/a&gt;
	&lt;li&gt;&lt;a href=&quot;http://www.faculty.iu-bremen.de/hjaeger/&quot;&gt;Herbert
Jaeger&lt;/a&gt;, &quot;From Continuous Dynamics to Symbols&quot;, in W. Tschacher and
J.-P. Dauwalder (eds.), &lt;cite&gt;Dynamics, Synergetics, Autonomous Agents&lt;/cite&gt;
(Singapore: World Scientific, 1999) pp. 29--48 [&lt;a
href=&quot;http://www.faculty.iu-bremen.de/hjaeger/pubs/jaeger.97.symbols.pdf&quot;&gt;PDF&lt;/a&gt;]
	&lt;li&gt;Teed Rockwell, &quot;Attractor Spaces as Modules: A Semi-Eliminative
Reduction of Symbolic AI to Dynamic Systems Theory&quot;, &lt;a
href=&quot;http://dx.doi.org/10.1007/s11023-004-1344-7&quot;&gt;&lt;cite&gt;Minds and
Machines&lt;/cite&gt; &lt;strong&gt;15&lt;/strong&gt; (2005): 23--55&lt;/a&gt;
	&lt;li&gt;Whit Schonbein, &quot;Cognition and the Power of Continuous Dynamical
Systems&quot;, &lt;a href=&quot;http://dx.doi.org/.1007/s11023-004-1345-6&quot;&gt;&lt;cite&gt;Minds and
Machines&lt;/cite&gt; &lt;strong&gt;15&lt;/strong&gt; (2005): 57--71&lt;/a&gt;
	&lt;/ul&gt;
</description>
  </item>
  </channel>
</rss>