Logo of 2002 conference

Cogsci 2002 Tutorial Programme

There are six tutorials this year on Wednesday 7 August (rooms to be announced on the day). They cost $45 (about 32 pounds or 51 Euros) for a half-day tutorial and $90 for a whole day tutorial. Payment can be made using the registration site on the conference page, or can be paid for on the day (if space is available).

The program includes handouts, and a tea and a coffee break (including biscuits). Lunch is available on campus in the Johnson Center, which we recommend you use. There will be a meeting of the tutorial committee meeting and get-together after the tutorials, location to be announced on the day.

Registration for tutorial attendees will be from 8.30 am on 7 August in the lobby of the Johnson Center on the Fairfax Campus (main campus) of George Mason University (where the conference is held). It should take less than 5 minutes to get from the tutorial desk to the tutorial rooms.

9.30 - 12.30 is the morning session, and 2-5 pm is the afternoon session.

In addition to the descriptions here, the materials for several of these tutorials is available as well.

Multiple Perspectives on Consciousness for Cognitive Science
Carlson, Half-day (morning)
in the Johnson Center

APEX/CPM-GOMS: An Architecture for Modeling Human Performance in Applied HCI Domains
Remington et al., Half-day (morning)
Johnson Center/Enterprise

How to Build Intelligent Interactive Agents Using Soar
Jones et al., Full-day
Johnson Center/Enterprise

ACT-R Tutorial
Lebiere, Half-day (morning)
in the Johnson Center

Functional Imaging of the Brain - Developing a Synergy of Cognitive Neuroscience Behavior and Modeling
Walter Schneider, Half-day (afternoon)
in the Johnson Center

A Cognitive Approach to Designing Human Error Tolerant Interfaces
Wood & Byrne, Half-day (afternoon)
in the Johnson Center

Back to registration

========================================================================

Multiple Perspectives on Consciousness for Cognitive Science

Half-day tutorial (morning)
in the Johnson Center

Rich Carlson
Penn State University
cvy@psu.edu

The huge contemporary literature on consciousness spans multiple disciplines, including psychology, philosophy, and neuroscience. Cognitive scientists may be interested in understanding consciousness as a phenomenon to be explained, as a source of evidence to constrain theory, and as a link between cognitive theory and everyday experience. This tutorial will introduce participants to major proposals about consciousness, and their empirical and methodological implications. The perspectives to be considered include:

1. Philosophy

1.1. Daniel Dennett's "Multiple drafts" proposal
1.2. David Rosenthal's "Higher Order Thought" (HOT) theory
1.3. David Chalmers on "easy" and "hard" problems of consciousness
1.4. John Searle's analysis of intentionality

2. Neuroscience

2.1. Gerald Edelman's "Dynamic core" proposal
2.2. Francis Crick's "Temporal binding" hypothesis
2.3. Antonio Damasio's "Core consciousness/extended consciousness" distinction

3. Psychology

3.1. Bernard Baars' "Global workspace" and related views
3.2. Attention and consciousness
3.3. Metacognition and consciousness
3.4. Richard Carlson's "Experienced cognition" view

These perspectives will be considered in relation to core questions about consciousness, including:

1. How can subjectivity and agency be accommodated in a scientific theory of consciousness?
2. How can conscious and nonconscious or unconscious processes and representations be systematically distinguished?
3. How can conscious mental states be assessed?
4. How can dissociations and impairments of consciousness be understood?

This tutorial will provide participants with starting points for understanding and contrasting current perspectives on consciousness. Brief orienting readings and a more comprehensive bibliography will be provided.

Rich Carlson received his Ph.D. in cognitive psychology from the University of Illinois in 1984 and has been on the faculty of Penn State University since 1985. His research program is concerned with the relation between consciousness and learning, especially skill acquisition. He has published a number of empirical articles on these topics. His book, "Experienced Cognition" (Erlbaum, 1997), describes a framework for understanding consciousness that draws on ecological and situated views of cognition as well as standard theoretical views from cognitive psychology. His web site is psych.la.psu.edu/faculty/carlson.htm

========================================================================

APEX/CPM-GOMS: Modeling Human Performance in Applied HCI Domains

Half-day tutorial (morning)
Johnson Center/Enterprise

Roger Remington
NASA Ames Research Center
Moffett Field, CA 94035
mfreed@arc.nasa.gov

Bonnie John - Carnegie Mellon University

Michael Matessa - NASA Ames Research Center

Alonso Vera - NASA Ames Research Center

Michael Freed- NASA Ames Research Center

This tutorial introduces participants to CPM-GOMS modeling using APEX, a tool for applied human performance modeling. APEX-CPM is intended to be of value to both cognitive science researchers and HCI professionals. It is also valuable for teaching students about task-analysis, user-modeling, and computational cognitive modeling in general. The tutorial will teach participants how to represent GOMS, KLM, and CPM-GOMS task analyses in APEX-CPM, and to refine models based on output in the form of automatically generated PERT charts. We will also discuss recent improvements aimed at making it practical to model in more complex HCI domains. These include capabilities that (a) facilitate representation of simulation environments and (b) allow modelers to draw on a set of reusable building blocks both for cognitive/task modeling and for physical environment modeling. The tutorial will primarily consist of a guided tour through activities supplemented with presentations. Participants will work in pairs, supervised by at least four presenters, on applied modeling problems. This tutorial should be of particular value to people interested in developing, using, and/or teaching engineering models of human performance in HCI contexts. Some background in cognitive modeling and computer programming is recommended. Tutorial participants will be given a CD containing the APEX-CPM code, a world-building tool-kit, and a number worked example models.

Roger Remington is a Senior Research Psychologist at NASA Ames Research Center. He holds a Ph.D. in Psychology from the University of Oregon.

Bonnie John is an Associate Professor in the Institute of Human-Computer Interaction and Carnegie-Mellon University. She holds a Ph.D. in Psychology from Carnegie Mellon University.

Michael Matessa is a Research Psychologist at NASA Ames Research Center. He holds a Ph.D. in Cognitive Psychology from Carnegie Mellon University.

Alonso Vera is a Research Scientist at NASA Ames Research Center. He holds a Ph.D. in Experimental Psychology from Cornell University.

Michael Freed is a Research Scientist at NASA Ames Research Center. He holds a Ph.D. in Computer Science from Northwestern University.

==========================================================

How to Build Intelligent Interactive Agents Using Soar

Full-day tutorial
Johnson Center/Enterprise

Randolph M. Jones
Soar Technology, Inc., and Colby College
rjones@soartech.com

Robert E. Wray, III, Soar Technology, Inc.

Amy E. Henninger, Soar Technology, Inc.

Scott Wood, Soar Technology, Inc.

Ronald S. Chong, George Mason University

Soar has been under development for over two decades as an architecture for building intelligent systems and human behavior models. Recent research and development activity with Soar has emphasized building competent, autonomous agents that interact with realistic and complex simulated environments. This tutorial will teach some of the methods that we use to design and engineer such behavior models. Instead of focusing on strict cognitive modeling, this tutorial will discuss the complexities that autonomous behavior and real-time interaction impose on a model. It will not involve intensive programming of intelligent agents, but will concentrate on higher level issues of task analysis, knowledge representation, process-oriented modeling, and knowledge acquisition. Many of these activities are useful to learn even if one does not use Soar to implement models, but the tutorial will also demonstrate the ways that Soar approaches and informs (and sometimes exacerbates) these tasks. To this end, tutorial participants will study and tweak a variety of interactive behavior models, and learn techniques for representing knowledge and behavior in Soar. They will also gain experience with some of the new development tools that support Soar modeling. Tutorial participants do not need an extensive background in programming.

Presenters: Each of the presenters has research and industry experience building interactive intelligent agent systems. Many of these have been "believable" agents with large amounts of knowledge developed within Soar. Soar Technology, Inc., uses Soar and other software paradigms to create intelligent and usable software for a variety of defense applications. All of the presenters also have experience developing interactive models of human behavior for various purposes, such as improving intelligent agents, improving human-computer interaction, understanding learning in problem solving, and studying human error in interactive tasks.

========================================================================

ACT-R

Half-day tutorial
Johnson Center/Enterprise

Christian Lebiere
Carnegie-Mellon University
cl@andrew.cmu.edu

ACT-R is a cognitive theory and simulation system for developing cognitive models. It assumes cognition emerges through the interaction of a procedural memory of productions with a declarative memory of chunks and independent modules for external perception and actions. The ACT-R 4.0 version of the theory was detailed in the book "The Atomic Components of Thought" by John R. Anderson and Christian Lebiere, published in 1998 by Lawrence Erlbaum. Since its release in 1997, ACT-R 4.0 has supported the development of over 100 cognitive models published in the literature by many different researchers. These models cover topics as diverse as driving behavior, implicit memory, learning backgammon, metaphor processing, and emotion. We have recently developed a new version, ACT-R 5.0 that extends ACT-R 4.0 to be more interruptible, to achieve greater across-task parameter consistency, to have better mechanisms of production learning, and to be more in correspondence with our knowledge of brain function. While the new system extends the capabilities of ACT-R 4.0, it involves relatively few changes and is actually simpler. This short tutorial will provide an overview of ACT-R, as it is specified in the 5.0 version, and some of its applications. It will not assume a prior background in ACT-R 4.0.

Christian Lebiere is a Research Scientist in the Human-Computer Interaction Institute at Carnegie-Mellon University. He received his B.S. in Computer Science from the University of Liege (Belgium) and his M.S. and Ph.D. from the School of Computer Science at Carnegie Mellon University. During his graduate career, he worked on the development of connectionist models, including the Cascade-Correlation neural network learning algorithm. Since 1990, he has worked on the development of the ACT-R hybrid cognitive architecture and is co-author with John R. Anderson of the 1998 book "The Atomic Components of Thought". His main research interest is cognitive architectures and their applications to psychology, artificial intelligence, human-computer interaction, decision-making, game theory, and computer-generated forces.

========================================================================

Functional Imaging of the Brain -- Developing a Synergy of Cognitive Neuroscience Behavior and Modeling

Half-day tutorial (afternoon)
in the Johnson Center

Walter Schneider
University of Pittsburgh
Pittsburgh 152260
wws@pitt.edu

The last ten years have produced an explosive growth in brain imaging technology and findings. The combination of fMRI, ERP, DTI, and PET enable non-invasive research on humans obtaining millimeter and millisecond resolution of activation, connection tracing, and mapping of transmitter systems. This large effort (1000+ papers per year) is providing detailed data of the biology of cognition and having a large impact on the conceptualization of cognitive science. There is a "grand challenge" to the field to relate the biology and mechanisms of human thought. There is a critical need for comprehensive behavioral, theoretical, and modeling efforts to interpret the findings. This tutorial will provide an introduction to the brain imaging methods stressing both the potential and limitations of the existing methods. We will describe the challenges that cognitive science methods may be particularly beneficial to resolving. We will look at efforts to relate modeling (e.g., ACT-R, LSA) and activation data. We will provide guidelines on how to get into brain imaging via collaboration or direct imaging.

Walter Schneider (B.A. Psychology, U. Illinois 1971; Ph.D. Psychology Indiana U. 1975). He is a fellow of the American Psychology Association and AAAS, is know for his classic work on automaticity and skill acquisition, published some of the first papers on fMRI in humans, and has developed software systems for empirical and brain imaging systems used in 2,000 laboratories. His current research focuses on brain imaging and modeling of learning, attention, and language processing and the modeling of skill acquisition and control/automatic processing. His web site is www.pitt.edu/~schlab/People/walt.htm

========================================================================

A Cognitive Approach to Designing Human Error Tolerant Interfaces

Half-day tutorial (afternoon)
in the Johnson Center

Scott D. Wood
Soar Technology, Inc.
Ann Arbor, MI 48105
swood@soartech.com

Michael Byrne
Rice University
byrne@acm.org

Designing for human error is a major challenge for developers of safety-critical and mission-critical systems. Human error is of particular concern for banking, commerce, medicine, military, and any system where tasks are performed with high frequency. Most approaches to error-tolerant design use either general design guidelines or treat humans as just another error-prone system component. This tutorial approaches error tolerant design from a cognitive perspective, focusing on practical techniques for improving your system's ability to deal with inherent human limitations. Participants will learn the basics of human error, how to classify error types, a framework error for tolerant design, how to deal with multiple aspects of error in design, and how to form a multilayered defense against error. The tutorial will include a mix of lecture, discussion, and several hands-on exercises. No prior experience modeling or designing for human error is required.

Scott D. Wood is a Senior Scientist at Soar Technology, a research and development company that develops systems utilizing knowledge-rich human behavior, cognitive modeling, and information visualization for military and intelligence applications. He has over 10 years of research and industry experience in the areas of software development, e-business, cognitive modeling, and human-computer interaction. His research includes computational modeling of human error, human error prediction techniques, and applying those techniques to mission-critical interfaces. Dr. Wood has designed and implemented a variety of error-tolerant interfaces for web banking and business process reengineering. His other experience includes human-performance modeling using the EPIC architecture, and optimizing workflows and interface usability through task analysis. His current book, "Usability for the Web", is a roadmap for creating highly usable web sites and applications. He earned a Ph.D. in Computer Science and Engineering from the University of Michigan, Ann Arbor.

Michael D. Byrne is a faculty member of the Psychology Department at Rice University. His primary research interests include computational modeling of human cognition and performance, human error, visual attention, and human-computer interaction. He is currently funded by NASA to model human peformance, particularly pilot errors, in the commercial aircraft cockpit. He has worked with several cognitive architectures including CAPS, SPAN, and ACT-R, and is the primary developer of ACT-R/PM. He received a Ph.D. in Experimental Psychology and an M.S. in Computer Science from Georgia Tech.

==========================================================

Also available, Tutorial booklets from the CogSci 1999 conference, and Tutorial booklets from the 2001 Conference.

last updated 16 Jan 03 -FER

~ ~ ~ ~ ~ ~ ~ ~

Main Conference Page -- Back to registration -- Programme -- Travel

Cognitive Science Society