INTRODUCTION TO COGNITIVE SCIENCE

2007 Fall

Instructor: Péter Érdi, Henry R. Luce Professor of Complex Systems Studies

Office:Olds/Upton 208B

Office hours: by appointment

Phone: (269)337-520

email: perdi@kzoo.edu

URL: http://cc.kzoo.edu/~perdi/

Topic:

Cognitive science is the interdisciplinary study of mind and the nature of intelligence. It is a rapidly evolving field that deals with information processing, intelligent systems, complex cognition, and large-scale computation. The scientific discipline encompasses the overlapping areas of neuroscience, psychology, computer science, linguistics and philosophy. Students will learn the basic physiological and psychological mechanisms and computational algorithms underlying different cognitive phenomena.

Goal:

The first goal is to teach WHY cognitive science became a popular and efficient discipline to investigate natural and artificial information processing devices. The second goal is to give an introduction to the historical development of the field. The third goal is to show HOW to analyze the performance of cognitive systems. The course is designed mostly for psychology and computer science students, but other students interested in interdisciplinary thinking might take the class.

Prerequisite:

PSYC 101 and COMP 105 or permission.

Course Structure:

Each week a topic will be discussed. There will be some weekly reading

assignments. During the term it will be possible to give reports on readings.

Exams:

There will be a sixty minute long written midterm and a final oral examination. Active participation in class activities are strongly encouraged. Class discussion is a vital part of learning in this style of course. Reports on readings and discussion forums might determine the grades up to 40%. Written and oral reports on a group project are a pre-requirement for taking the final examination. Active participation in events organized by the Center for Complex System Studies is encouraged, and will also be considered in assigning the final grade.

Special Excuse

I received and invitation to give a talk on conference on biological computation BIOCOMP2007 http://biocomp.unina.it/2007/index.html to be held in Southern Italy at the end of September. Since I could not resist to accept the invitation, classes will start on October 1^st, Monday. Classes canceled will be rescheduled for some evenings.

Book required:

Paul Thagard: Mind : Introduction to Cognitive Science.

Cambridge, MIT Press. (MIT Press, 1996), (2nd edition, 2005). (I will refer to it as T:M)

Supporting literature:

Eichenbaum H: The Cognitive Neuroscience of Memory: An Introduction. Oxford University Press, 2002

Johnson-Laird, P.N.:Mental Models. Cambridge: Cambridge University Press. Cambridge, Mass.:Harvard University Press. 1983

Polk T. and Seifert., C.M., Cognitive Modeling, Eds. MIT Press 2002

Ward LM: Dynamical Cognitive Science. MIT Press , 2001

Mind Readings: Introductory Selections on Cognitive Science

by Paul Thagard (Editor) MIT Press, 1998)

[MIT Encyclopedia of Cognitive Science]

Cognitive Science Resources

http://cogsci.uwaterloo.ca/courses/resources.html

Weekly Topics:

1. Cognitive science: a true interdisciplinary field

A very good summary can be found at:

http://plato.stanford.edu/entries/cognitive-science/ or

http://www.science.uva.nl/~seop/entries/cognitive-science/

Disciplines:

Philosophy

Psychology

Computer science and AI

Neuroscience

Linguistics

Anthropology

Overview:

History

Methods

Representation and computation

Theoretical approaches

Philosophical relevance

Why we are NOT using a single textbook?

Two different approaches exist:

I. Assumptions about the representation of the knowledge in the mind

(logic, rules etc.)

II. Cognitive functions (learning, memory, emotions etc.)

Reading: T:M Chapter 1.

This class integrates the two approaches.

Key question: how can the different disciplines interact to undestand mind?

2. Knowledge Representation: Logic, Rule-based systems and others

Mind might contain mental representations:

Formal logic

Rules

Concepts

Analogies

Images

Connections

Formal logic

· People have mental representations similar to sentences in predicate logic.

· People have deductive and inductive procedures that operate on those sentences.

· The deductive and inductive procedures, applied to the sentences, produce the inferences.

(You certainly should know the most important inferences:

Modus ponens: 1. If P, then Q . 2. P. Therefore, Q.

Modus tollens: 1. If P, then Q. 2. Q is false. Therefore, P is

false. (indirect proof)

The scope and limits of the approach of formal logic (and deductive reasoning. Two papers for group discussion:

Philip N. Johnson-Laird: Human reasoning and rationalitywww.pul.it/irafs/CD%20IRAFS'02/texts/Johnson%20Laird.pdf

Rips, L. J.:. Two kinds of reasoning. Psychological Science, 12, 129-134, 2001

to be downloaded from http://www.psych.northwestern.edu/~rips/

Readings: T:M Chapter 2 and 3.

Rules

· People have mental rules.

· People have procedures for using these rules to search a space of possible solutions, and procedures for generating new rules.

· Procedures for using and forming rules produce the behavior.

Rule-based systems:

manipulation and transformation of symbols

Rule-based programes for AI and cognitive science:

0. ELIZA http://en.wikipedia.org/wiki/ELIZA

1. Newell and Simon, GPS 1950s-60s

2. Expert systems, 1970s-90s. Most corporations.

3. ACT 1983. John Anderson.

4. SOAR, Newell and his students, 1980s, John E. Laird (Univ. Michigan)

5. Prolog: logic programming

Strength and weakness of the rule-based systems

Rules of Language: from Chomsky to Pinker

Noam Chomsky:

http://en.wikipedia.org/wiki/Noam_Chomsky

Steven Pinker: "Rules of Language

http://www.sciencemag.org/cgi/content/abstract/253/5019/530

3. Neural Networks and Connectionism

Logic and rules are the main forms of mental representations of the classical, symbolic approach. McCulloch-Pitts neurons connects logic

to networks. How to calculate with MCP neurons and networks?

http://www.mind.ilstu.edu/curriculum/modOverview.php?modGUI=212

The Perceptron.

Connectionism offered and alternative to this symbolic approach.

The most celebrated book of the connectionist alternative is:

A first reading on connectionism:

http://en.wikipedia.org/wiki/Connectionism

Second reading:

Connectionism: an introduction

http://www.mind.ilstu.edu/curriculum/modOverview.php?modGUI=76

Can Computers Have Emotions?http://www.inf.ed.ac.uk/events/hotseat/panel_statements.html#andy_position

Affective computing in the MIT:

Picard, R. W. (2003), "Affective Computing: Challenges," Int. Journal of Human-Computer Studies, Vol. 59, Issues 1-2, July 2003, pp. 55-64.

The neural basis of psychonalysis: from Sigmund Freud to Eric Kandel

Amit Etkin, M.Phil, Ph.D., Christopher Pittenger, M.D., Ph.D., H. Jonathan Polan, M.D. and Eric R. Kandel, M.D.:Toward a Neurobiology of Psychotherapy: Basic Science and Clinical Applications .

J Neuropsychiatry Clin Neurosci 17:145-158, May 2005

T:M Chapter 10. and 11.

9. The Mind as a Dynamical System

- Cognitive agents are dynamical systems

– Cognitive agents can be understood by dynamical system theory

van Gelder, T. J. (1998) The dynamical hypothesis in cognitive science. Behavioral and Brain Sciences, 21, 1-14

http://www.arts.unimelb.edu.au/~tgelder/Publications.html

Dynamical cognitive neuroscience: experiment and models

T:M Chapter 12

10. Summary and Outlook

Cognitive science is an interdisciplinary study of mind and intelligence.

Levels, methods, the need of integration.

Open problems

Institutions, graduate programs

T:M Chapter 14.