COGS 2010 Semester II
2003
Assignment
Date Due: Thursday
30th October (week 13). Note Extension to
Submission: At
the lecture or hard
copy to my office 78-333 or online to http://submit.itee.uq.edu.au/select.php?coursecode=COGS2010
Marks: 10% of total
assessment. 1 mark per day will be deducted for late submission without prior permission.
Submissions later than 1 week late will not be marked. Contact the lecturer at
least 24 hours in advance if you have a valid request for extension or if you
are ill at the time submit a medical certificate.
The Assignment
The task for this assignment is to write an analytical
report of a computational cognitive model, as described in published papers,
and if possible test a software demo (note that you are not expected to write
one! Just find it from the literature and/or the web). You may write one in
NetLogo to demonstrate an idea if you choose, but note that the aim of the
assessment is the understanding of the model and written report and marks will
be assigned accordingly.
Choose one of the topics listed below (or choose one
of your own, and check it by email with the lecturer). Prepare the report using
the following headings:
Source
q
(Where is the model from?)
Background
q
Identify the phenomenon that
the model designers are trying to model. What level of description is being
used {neuroscience/psychology/linguistic?}
The Model
q
Architecture
q
Task/Environment
q
Simulations Run
Discussion
q
Critical discussion of the
implementation of the model and the extent to which it achieves the goals of
the designers. By this, we mean that you
should describe the model, and then evaluate the quality of the model. That is, is this is a useful approach?
Justify your answer.
References
q
Any conventional referencing
format is acceptable (e.g., APA, IEEE journal format, Alife or Neural
Computation). Marks will be deducted for inadequately referenced work.
The Models
Models can be chosen from neural networks,
evolutionary computation (including evolutionary robotics), complex systems
(applied to cognitive science) or artificial life. Find one for yourself, or choose one of the
following:
A. Developmental, learning and memory models
1.
Elman, J.L., Bates, E.A., Johnson, M.H., Karmiloff-Smith,
A., Parisi, D., and Plunkett, K. (1996). Rethinking Innateness,
Cambridge, Mass: MIT Press. Software is available for some of these models from
the accompanying textbook, Plunkett K.
and Elman, J. (1997). Exercises in Rethinking Innateness, Cambridge,
Mass: MIT Press.
a.
Learning the past tense,
ch3, pp130-147.
b.
Stages and rules in
children's thinking, ch3, pp158- 169.
c.
Imprinting in chicks and
object recognition in children, ch6, pp327-333.
d.
Emergent properties of
networks due to waves of learning, ch6, pp334-340
2.
Memory models, see for
example, McClelland, J.L., McNaughton, B.L., and O'Reilly, R.C. Why there are
complementary learning systems in hippocampus and neocortex: Insights from the
successes and failures of connectionist models of learning and memory.
Psychological Review. 102: 419-457, 1995.
3.
Reading models, see for
example, Plaut, D. C., McClelland, J. L., Seidenberg, M. S., and Patterson, K.
E. (1996). Understanding normal and impaired word reading: Computational
Principles in Quasi-Regular Domains. Psychological Review, 103, 565-115.
B. Evolution of language
1.
Transmission of language
through generations of neural networks.
Hare, M. and Elman, J.L., (1993). From Weared to wore: a connectionist
account of language change In Proceedings of the Fifteenth Conference of the
Cognitive Science Society, Hillsdale, NJ:Erlbaum.
2.
The Talking Heads Experiment
by Luc Steels (the talking heads are robotic agents that watch scenes in front
of them and engage in language games about those scenes). Luc Steels has written many papers. One
recent example is The
Emergence of Grammar in Communicating Autonomous Robotic Agents. Proceedings of ECAI 2000, August 2000.
3.
Work by John Batali (UCSD)
or Simon Kirby (Edinburgh) on the evolution of language. Both are prolific
authors. Select one of their published papers that appeals to you. For example,
J. Batali. Computational
Simulations of the Emergence of Grammar. in Approaches to the Evolution
of Language: Social and Cognitive Bases, J. R. Hurford, M.
Studdert-Kennedy and C. Knight (eds). Cambridge University Press, pages
405-426, 1998.
C. Evolution of cooperation
1.
The Prisoner’s dilemma and
all its variations. For an overview, see
chapter 3 in Ridley, M. (1996). The
Origins of Virtue. NY: Viking,
and select one of the published
papers from the reference list.
2.
Frean, M.R. (1996) The
evolution of degrees of cooperation. Journal
of Theoretical Biology , 182, 549 - 559.
3.
Batali, J. and Kitcher, P. Evolution of
Altruism in Optional and Compulsory Games. Journal of Theoretical
Biology. Vol. 175, pages 161-171. 1995.
D. More models:
1.
Cognitive Neuroscience M.S.Gazzaniga
The Cognitive Neurosciences, MIT Press. 1996. ch 27 "A Model of Visual
Motion Processing in Area MT of Primates"
2.
Music - P.M.Todd and D. G.
Loy, "Music and Connectionism", Cambridge, Mass: MIT Press A
Connectionist Approach to Algorithmic Composition. p. 173-194.
3.
Several good compilations of
connectionist models exist. To find your own paper, good starting points
include:
a.
"The Handbook of Brain
Theory and Neural Networks", by M.A. Arbib, 1995, MIT Press,
b.
McClelland, J. L. Rumelhart,
D.E., & the PDP Research Group (1986). Parallel Distributed Processing:
Explorations in the microstructure of cognition. Vol. 2. Psychological and
Biological Models. Cambridge, MA: MIT Press.
c.
The Cognitive Science
Society Conference and journal http://www.umich.edu/~cogsci/about.html
Cogs2010
Assignment Marking Scheme
|
|
|
|
|
q
Indentification of the source of the model. |
1 |
|
|
q Description of the
phenomenon q
Description of the level of description |
1 |
|
|
q Architecture q Task/Environment q
Simulations Run |
3 |
|
|
q Critical discussion of the
implementation of the model and the extent to which it achieves the goals of the
designers. q Discussion of the insights
it provides for cognitive science q
Discussion of the implications of the work |
2 |
|
|
Any conventional referencing format is acceptable. Marks will be
deducted for inadequately referenced work. q Correctly formatted
reference section q
Correct referencing in text |
1 |
|
|
q Evidence of understanding
of the model q Evidence of original
thought q Clarity of expression q Completeness of coverage |
2 |
|
|
|
|
|
|
Total |
/10 |
|