This book aims to understand human cognition and psychology through a comprehensive computational theory of the mind, namely, a "cognitive architecture."

• Table of Contents


• Preface


• Chapter 1. What is A Cognitive Architecture?


• 1.1. A Theory of the Mind and Beyond


• 1.2. Why Computational Models/Theories?


• 1.3. Questions about Computational Models/Theories


• 1.4. Why a Computational Cognitive Architecture?


• 1.5. Why CLARION?


• 1.6. Why This Book?


• 1.7. A few Fundamental Issues


• 1.7.1. Ecological-Functional Perspective


• 1.7.2. Modularity


• 1.7.3. Multiplicity of Representation


• 1.7.4. Dynamic Interaction


• 1.8. Concluding Remarks


• Chapter 2. Essential Structures of the Mind


• 2.1. Essential Desiderata


• 2.2. An Illustration of the Desiderata


• 2.3. Justifying the Desiderata


• 2.3.1. Implicit-Explicit Dichotomy and Synergistic Interaction


• 2.3.2. Separation of the Implicit-Explicit and the Procedural-Declarative Distinction


• 2.3.3. Bottom-up and Top-down Learning


• 2.3.4. Motivational and Metacognitive Control


• 2.4. Four Subsystems of CLARION


• 2.4.1. Overview of the Subsystems


• 2.4.2. The Action-Centered Subsystem


• 2.4.3. The Non-Action-Centered Subsystem


• 2.4.4. The Motivational Subsystem


• 2.4.5. The Metacognitive Subsystem


• 2.4.6. Parameters of the Subsystems


• 2.5. Accounting for Synergy within the Subsystems of CLARION


• 2.5.1. Accounting for Synergy within the ACS


• 2.5.2. Accounting for Synergy within the NACS


• 2.6. Concluding Remarks


• Chapter 3. Subsystems, Modules, and Algorithms I: The Action-Centered and Non-Action-Centered Subsystems


• 3.1. The Action-Centered Subsystem


• 3.1.1. Background


• 3.1.2. Representation


• 3.1.2.1. Representation in the Top Level


• 3.1.2.2. Representation in the Bottom Level


• 3.1.2.3. Action Decision Making


• 3.1.3. Learning


• 3.1.3.1. Learning in the Bottom Level


• 3.1.3.2. Learning in the Top Level


• 3.1.4. Level Integration


• 3.1.5. An Example


• 3.2. The Non-Action-Centered Subsystem


• 3.2.1. Background


• 3.2.2. Representation


• 3.2.2.1. Overall Algorithm


• 3.2.2.2. Representation in the Top Level


• 3.2.2.3. Representation in the Bottom Level


• 3.2.2.4. Representation of Hierarchies


• 3.2.3. Learning


• 3.2.3.1. Learning in the Bottom Level


• 3.2.3.2. Learning in the Top Level


• 3.2.4. Memory retrieval


• 3.2.5. An Example


• 3.3. Knowledge Extraction, Assimilation, and Transfer


• 3.3.1. Background


• 3.3.2. Bottom-Up Learning in the ACS


• 3.3.2.1. Rule Extraction and Refinement


• 3.3.2.2. Independent Rule Learning


• 3.3.2.3. Implications of Bottom-up Learning


• 3.3.3. Top-down Learning and Assimilation in the ACS


• 3.3.4. Transfer of Knowledge from the ACS to the NACS


• 3.3.5. Knowledge Extraction in the NACS


• 3.3.6. Transfer of Knowledge from the NACS to the ACS


• 3.3.7. An Example


• 3.3.7.1. Learning about "Knife "


• 3.3.7.2. Learning about "Knife " within CLARION


• 3.3.7.3. Learning More Complex Concepts in CLARION


• 3.4. General Discussion


• 3.4.1. More on the Two Levels


• 3.4.2. More on the Two Learning Directions


• 3.4.3. Controversies


• 3.4.4. Summary


• Appendix


• A.1. Response Time


• A.1.1. Response Time of the ACS


• A.1.2. Response Time of the NACS


• A.2. Learning in MLP (Backpropagation) Networks


• A.3. Learning in Auto-associative Networks


• A.4. Representation of Conceptual Hierarchies


• Chapter 4. Subsystems, Levels, and Algorithms II: The Motivational and Metacognitive Subsystems


• 4.1. Introduction


• 4.2. The Motivational Subsystem


• 4.2.1. Essential Considerations


• 4.2.2. Drives


• 4.2.2.1. Primary Drives


• 4.2.2.2. Secondary Drives


• 4.2.2.3. Approach versus Avoidance Drives


• 4.2.2.4. Drive Strength


• 4.2.3. Goals


• 4.2.4. Modules and Their Functions


• 4.2.4.1. Initialization Module


• 4.2.4.2. Preprocessing Module


• 4.2.4.3. Drive Core Module


• 4.2.4.4. Deficit Change Module


• 4.3. The Metacognitive Subsystem


• 4.3.1. Essential Considerations


• 4.3.2. Modules and Their Functions


• 4.3.2.1. Goal Module


• 4.3.2.2. Reinforcement Module


• 4.3.2.3. Processing Mode Module


• 4.3.2.4. Input/output Filtering Modules


• 4.3.2.5. Reasoning/learning Selection Modules


• 4.3.2.6. Monitoring Buffer


• 4.3.2.7. Other MCS Modules


• 4.4. General Discussion


• 4.4.1. Reactivity versus Motivational Control


• 4.4.2. Scope of the MCS


• 4.4.3. Need for the MCS


• 4.4.4. Information Flows Involving the MS and the MCS


• 4.4.5. Concluding Remarks


• Appendix: Additional Details of the MS and the MCS


• A.1. Change of Drive Deficits


• A.2. Determining Avoidance versus Approach Drives, Goals, and Behaviors


• A.3. Learning in the MS


• A.4. Learning in the MCS


• A.4.1. Learning Drive-Goal Connections


• A.4.2. Learning New Goals


• Chapter 5. Simulating Procedural and Declarative Processes


• 5.1. Modeling the Dynamic Process Control Task


• 5.1.1. Background


• 5.1.2. Task and Data


• 5.1.3. Simulation Setup


• 5.1.4. Simulation Results


• 5.1.5. Discussion


• 5.2. Modeling the Alphabetic Arithmetic Task


• 5.2.1. Background


• 5.2.2. Task and Data


• 5.2.3. Top-down Simulation


• 5.2.3.1. Simulation Setup


• 5.2.3.2. Simulation Results


• 5.2.4. Alternative Simulations


• 5.2.5. Discussion


• 5.3. Modeling the Categorical Inference Task


• 5.3.1. Background


• 5.3.2. Task and Data


• 5.3.3. Simulation Setup


• 5.3.4. Simulation Results


• 5.3.5. Discussion


• 5.4. Modeling Intuition in the Discovery Task


• 5.4.1. Background


• 5.4.2. Task and Data


• 5.4.3. Simulation Setup


• 5.4.4. Simulation Results


• 5.4.5. Discussion


• 5.5. Capturing Psychological "Laws "


• 5.5.1. Uncertain Deductive Reasoning


• 5.5.1.1. Uncertain Information


• 5.5.1.2. Incomplete Information


• 5.5.1.3. Similarity


• 5.5.1.4. Inheritance


• 5.5.1.5. Cancellation of Inheritance


• 5.5.1.6. Mixed Rules and Similarities


• 5.5.2. Reasoning with Heuristics


• 5.5.2.1. Representativeness Heuristic


• 5.5.2.2. Availability Heuristic


• 5.5.2.3. Probability Matching


• 5.5.3. Inductive Reasoning


• 5.5.3.1. Similarity between the Premise and the Conclusion


• 5.5.3.2. Multiple Premises


• 5.5.3.3. Functional Attributes


• 5.5.4. Other Psychological "Laws "


• 5.5.5. Discussion of Psychological "Laws "


• 5.6. General Discussion


• Chapter 6. Motivational and Metacognitive Simulations


• 6.1. Modeling Metacognitive Judgment


• 6.1.1. Background


• 6.1.2. Task and Data


• 6.1.3. Simulation Setup


• 6.1.4. Simulation Results


• 6.1.5. Discussion


• 6.2. Modeling Metacognitive Inference


• 6.2.1. Task and Data


• 6.2.2. Simulation Setup


• 6.2.3. Simulation Results


• 6.2.4. Discussion


• 6.3. Modeling Motivation-Cognition Interaction


• 6.3.1. Background


• 6.3.2. Task and Data


• 6.3.3. Simulation Setup


• 6.3.4. Simulation Results


• 6.3.5. Discussion


• 6.4. Modeling Human Personality


• 6.4.1. Background


• 6.4.2. Principles of Personality Within CLARION


• 6.4.2.1. Principles and Justifications


• 6.4.2.2. Explaining Personality within CLARION


• 6.4.3. Simulations of Personality


• 6.4.3.1. Simulation 1


• 6.4.3.2. Simulation 2


• 6.4.3.3. Simulation 3


• 6.4.4. Discussion


• 6.5. Accounting for Human Moral Judgment


• 6.5.1. Background


• 6.5.2. Human Data


• 6.5.2.1. Effects of Personal Physical Force


• 6.5.2.2. Effects of Intention


• 6.5.2.3. Effects of Cognitive Load


• 6.5.3. Two Contrasting Views


• 6.5.3.1. Details of Model 1


• 6.5.3.2. Details of Model 2


• 6.5.4. Discussion


• 6.6. Accounting for Emotion


• 6.6.1. Issues of Emotion


• 6.6.2. Emotion and Motivation


• 6.6.3. Emotion and the Implicit-Explicit Distinction


• 6.6.4. Effects of Emotion


• 6.6.5. Emotion Generation and Regulation


• 6.6.6. Discussion


• 6.7. General Discussion


• Chapter 7. Cognitive Social Simulation


• 7.1. Introduction and Background


• 7.2. Cognition and Survival


• 7.2.1. Tribal Society Survival Task


• 7.2.2. Simulation Setup


• 7.2.3. Simulation Results and Analysis


• 7.2.3.1. Effects of Social and Environmental Factors


• 7.2.3.2. Effects of Cognitive Factors


• 7.2.4. Discussion


• 7.3. Motivation and Survival


• 7.3.1. Simulation Setup


• 7.3.2. Simulation Results and Analysis


• 7.3.2.1. Effects of Social and Environmental Factors


• 7.3.2.2. Effects of Cognitive Factors


• 7.3.2.3. Effects of Motivational Factors


• 7.3.3. Discussion


• 7.4. Organizational Decision Making


• 7.4.1. Organizational Decision Task


• 7.4.2. Simulations and Results


• 7.4.2.1. Simulation I: Matching Human Data


• 7.4.2.2. Simulation II: Extending Simulation Temporally


• 7.4.2.3. Simulation III: Varying Cognitive Parameters


• 7.4.2.4. Simulation IV: Introducing Individual Differences


• 7.4.3. Discussion


• 7.5. Academic Publishing


• 7.5.1. Academic Science


• 7.5.2. Simulation Setup


• 7.5.3. Simulation Results and Analysis


• 7.5.4. Discussion


• 7.6. General Discussion


• 7.6.1. Theoretical Issues in Cognitive Social Simulation


• 7.6.2. Challenges


• 7.6.3. Concluding Remarks


• Chapter 8. Some Important Questions and Their Short Answers


• 8.1. Theoretical Questions


• 8.2. Computational Questions


• 8.3. Biological Connections


• Chapter 9. General Discussions and Conclusions


• 9.1. A Summary of the Cognitive Architecture


• 9.2. A Discussion of the Methodologies


• 9.3. Relations to Some Important Notions


• 9.4. Relations to Some Existing Approaches


• 9.5. Comparisons with Other Cognitive Architectures


• 9.6. Future Directions


• 9.6.1. Directions for Cognitive Social Simulation


• 9.6.2. Other Directions for Cognitive Architectures


• 9.6.3. Final Words on Future Directions


• References

Dr. Ron Sun is Professor of Cognitive Sciences at Rensselaer Polytechnic Institute. A well-known cognitive scientist, Ron Sun explores the fundamental structures of the human mind. He aims for the synthesis of many intellectual ideas into a coherent model of the human mind. The goal is to come up with a cognitive architecture that captures a variety of psychological processes and provides unified explanations of a wide range of data and phenomena.