[Comp-neuro] CFP: CogSci 2010 Workshop - Compositional Connectionism in Cognitive Science II: The Localist / Distributed Dimension

Ross Gayler r.gayler at gmail.com
Sat May 1 02:23:49 CEST 2010

CogSci 2010 Workshop - Compositional Connectionism in Cognitive Science II:
The Localist / Distributed Dimension
Portland, Oregon, USA 11 August, 2010 
Dear list members,
The aim of this workshop is to bring together researchers working with 
a wide range of compositional connectionist models, independent of 
application domain (e.g. language, logic, analogy, web search), with a 
focus on what commitments (if any) each model makes to localist or 
distributed representation.  We therefore welcome and encourage 
submissions from both localist and distributed modelers, as well as 
those whose work bypasses this distinction or challenges its 
importance.  We expect vigorous and exciting debate on this issue.
Topics of interest include:
* What do we mean by "localist" / "distributed" in terms of the 
relationship between connectionist units and the items they represent?
* To what extent is the term "connectionist" still valid; i.e., has 
the distributed / localist dimension supplanted the symbolic / 
connectionist dimension as the major axis of difference among 
cognitive models?
* How plausible and feasible is "holistic" computation, in which an 
entire structure is manipulated with sensitivity to its constituent 
parts without being decomposed into those parts, and does this 
feasibility depend on whether the representation is localist / 
* Are temporal-synchrony-firing models necessarily localist?
* What constraints can neuroscience research bring to the distributed 
/ localist debate, and what can this debate contribute to the 
interpretation of neuroscientific research?
* Are some cognitive functions more plausibly seen as localist, and 
others more plausibly distributed?
* Do distributed (or localist) models scale more easily than localist 
(distributed) models to realistically large problems?
* If two connectionist models, one distributed and the other localist, 
both account reasonably well for the same phenomenon, how can we judge 
between them?
* What mathematical principles (fractals, holography, chaos, etc.) can 
be borrowed from physics and other sciences to shed light on the 
nature of connectionist mental representations?
Our plenary speakers are Chris Eliasmith (U. Waterloo) and John Hummel 
(U. Illinois), for whom we plan to allocate roughly two hours of the 
day-long workshop.  This leaves time for around 12 additional paper 
presentations (20 minutes talk plus 10 minutes for questions).
My co-organizer Ross Gayler and I invite you to submit a one-page 
abstract on a topic that you consider relevant to these issues. 
Acceptance will be based on the abstracts, with 6 page conference 
papers to be delivered at the workshop.
The CogSci 2010 proceedings will only contain a brief description of 
the workshop, written before the speakers are known.  However, Ross 
and I plan to seek a publication venue for the accepted papers.  We 
have already been approached by the editor of a well-established and 
respected journal for the possibility of a special issue.  We expect 
standard six-page conference papers for the workshop but will try to 
be flexible for the published versions based on the more open-ended 
publication plans.
It would be helpful for us to know right away the number of people 
interested in submitting an abstract.  Please send me an email 
immediately if you intend submitting an abstract.
Although we are not tied to the CogSci conference publication 
schedule, Ross and I have found it useful to set a deadline of 15 May 
for the submission of abstracts, the same as the camera-ready deadline 
for the conference, and a deadline of 04 August for the papers, 
allowing us to hand out pre-prints of the papers at the workshop.
In sum, here are the relevant dates:
Now:      Informal letter of intent to submit abstract
15 May:   One-page abstract with references
04 Aug:   Six-page paper
Again, if you have any inclination to submit something, please let me 
know right away.  Even if you don't, we hope to see you at the 
workshop in Portland on 11 August.  Details of registration and 
accommodation can be found via the workshop website: 

Simon D. Levy
Associate Professor and Department Head
Computer Science Department
407 Parmly Hall
Washington&  Lee University
Lexington, VA 24450
540-458-8419 (voice)
540-458-8255 (fax)
levys at wlu.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20100501/a8c51292/attachment.html

More information about the Comp-neuro mailing list