Chapter
THE CURRENT STATE OF COMPUTATIONAL PSYCHIATRY
The Potential Impact of Computational Psychiatry and What Is Needed to Get There
Integration Across Levels of Analysis
Computational Phenotyping and Biomarker Refinement
WHAT COMPUTATIONAL PSYCHIATRY NEEDS TO SUCCEED?
Generating Multilevel Data
Mapping Categorical Versus Continuous Alterations: RDOC Versus DSM
Conceptually Separate “Big Data Analytics” From “Theory”
Sharing, Standardization, and Reproducibility
Ground-Truth Datasets for Benchmarking of Models
Linking Theoretical Approaches
Training and Development of the Scientific Workforce
Facilitating Dialogue Between Computational and Experimental Neuroscience
I
- APPLYING CIRCUIT MODELING TO UNDERSTAND PSYCHIATRIC SYMPTOMS
1 - Cortical Circuit Models in Psychiatry: Linking Disrupted Excitation–Inhibition Balance to Cognitive Deficits As ...
1.2 ROLES FOR BIOPHYSICALLY BASED NEURAL CIRCUIT MODELING IN COMPUTATIONAL PSYCHIATRY
1.3 LINKING PROPOSITIONS FOR COGNITIVE PROCESSES
1.4 ATTRACTOR NETWORK MODELS FOR CORE COGNITIVE COMPUTATIONS IN RECURRENT CORTICAL CIRCUITS
1.5 CIRCUIT MODELS OF COGNITIVE DEFICITS FROM ALTERED EXCITATION–INHIBITION BALANCE
1.6 CRITICAL ROLE OF EXCITATION–INHIBITION BALANCE IN COGNITIVE FUNCTION
1.7 FUTURE DIRECTIONS IN NEURAL CIRCUIT MODELING OF COGNITIVE FUNCTION
1.7.1 Integrating Cognitive Function With Neurophysiological Biomarkers
1.7.2 Incorporating Further Neurobiological Detail
1.7.3 Informing Task Designs
1.7.4 Studying Compensations and Treatments
2 - Serotonergic Modulation of Cognition in Prefrontal Cortical Circuits in Major Depression
3 - Dopaminergic Neurons in the Ventral Tegmental Area and Their Dysregulation in Nicotine Addiction
3.1 NICOTINE, DOPAMINE, AND ADDICTION
3.2 MODELING RECEPTOR KINETICS
3.2.1 Ligand–Receptor Interaction
3.2.1.1 A Two-Gate Receptor Model
3.2.1.2 Steady-State Receptor Current
3.2.1.3 Temporal Dynamics of the Receptor Current
3.2.2 Competition and Cooperation Between Ligands, or Between Ligand and the Endogenous Transmitter
3.2.2.1 Competitive Inhibition
3.2.2.3 Positive Allosteric Modulation
3.2.3 Miscellaneous and Secondary Receptor Effects
3.3 CIRCUIT MODELS OF THE VENTRAL TEGMENTAL AREA
3.3.1 Reorganization of the Ventral Tegmental Area in Addiction
3.3.2 Circuit Simulations of the Normal and Reorganized Ventral Tegmental Area
3.4 MODELING TONIC VERSUS PHASIC DOPAMINE RELEASE
3.4.1 Modeling the Firing Pattern of Dopamine Neurons
3.4.2 Effects of Nicotine on the Dopamine Cell's Spike Pattern
APPENDIX A: THE DOPAMINE NEURON MODEL
II
- MODELING NEURAL SYSTEM DISRUPTIONS IN PSYCHIATRIC ILLNESS
4 - Computational Models of Dysconnectivity in Large-Scale Resting-State Networks
4.1.1 The Study of Large-Scale Brain Connectivity
4.2 RESTING-STATE FUNCTIONAL CONNECTIVITY AND NETWORKS IN FUNCTIONAL MAGNETIC RESONANCE IMAGING
4.3 DYNAMIC FUNCTIONAL CONNECTIVITY
4.4 MEASURING STRUCTURAL CONNECTIVITY
4.5 EFFECTIVE CONNECTIVITY
4.6 TOPOLOGICAL ANALYSIS OF THE NETWORKS
4.7 COMPARING CONNECTIVITY AMONG GROUPS
4.7.1 Clinical Applications of Large-Scale Resting State Connectivity
4.8 MODELING THE LARGE-SCALE BRAIN ACTIVITY-I: LINKING STRUCTURE AND FUNCTION
4.9 MODELING THE LARGE-SCALE BRAIN ACTIVITY-II: ADDING DYNAMICS INTO THE EQUATION
5 - Dynamic Causal Modeling and Its Application to Psychiatric Disorders
5.1 INTRODUCTION TO DYNAMIC CAUSAL MODELING
5.1.1 Dynamic Causal Models for Functional Magnetic Resonance Imaging
5.1.2 Dynamic Causal Models for Electrophysiological Data
5.1.3.1 Model Family Comparison
5.1.4 Model Parameter Estimates: Physiological and Clinical Interpretations
5.1.4.1 Posterior Parameter Estimates
5.1.4.3 Generative Embedding
5.1.5 Other Variants of Dynamic Causal Modeling
5.2 APPLICATION OF DYNAMIC CAUSAL MODELING IN PSYCHIATRY
5.2.1 Using Dynamic Causal Modeling to Understand Mechanism of Behavioral/Cognitive Dysfunction
5.2.2 Using Dynamic Causal Modeling to Investigate Synaptic Dysfunction
5.2.3 Using Dynamic Causal Modeling to Dissect Spectrum Disorders
5.2.4 Current Dynamic Causal Modeling Limitations
6 - Systems Level Modeling of Cognitive Control in Psychiatric Disorders: A Focus on Schizophrenia
6.2 MECHANISMS OF CONTROL: PROACTIVE AND REACTIVE
6.2.1 Proactive Versus Reactive Control Deficits in Schizophrenia
6.2.2 Computational Models of Proactive and Reactive Control
6.2.2.1 Connectionist Modeling of Proactive Control in Schizophrenia: Guided Activation Framework
6.2.2.1.1 Using Proactive Control Models to Make Predictions About Dorsolateral Prefrontal Cortex Activity
6.2.2.2 Reactive Control—When to Engage or Upregulate Control
6.2.2.2.1 Performance Monitoring and Reactive Control in Schizophrenia
6.2.2.3 Relationships Between Proactive Control and Reactive Control in Schizophrenia
6.3 UPDATING CONTROL REPRESENTATIONS—DOPAMINE, THE STRIATUM AND A GATING MECHANISM
6.3.1 Dopamine and Gating in Schizophrenia
6.4 COGNITIVE CONTROL, VALUE, AND EFFORT ALLOCATION
6.4.1 Cognitive Control, Utility, and Exploitation Versus Exploration
6.4.2 Model-Based Learning and a Decision-Making as a Form of Cognitive Control
6.4.3 Cognitive Control Impairments as a Core Feature of Psychopathology
6.5 SUMMARY AND FUTURE DIRECTIONS
7 - Bayesian Inference, Predictive Coding, and Computational Models of Psychosis
7.1 HIERARCHICAL MODELS AND PREDICTIVE CODING
7.2 PSYCHOSIS, SYNAPTIC GAIN, AND PRECISION
7.3 COMPUTATIONALLY MODELING THE FORMATION OF DELUSIONS
7.4 MODELING THE MAINTENANCE OF DELUSIONS
7.5 CONCLUSIONS AND FUTURE DIRECTIONS
III
- CHARACTERIZING COMPLEX PSYCHIATRIC SYMPTOMS VIA MATHEMATICAL MODELS
8 - A Case Study in Computational Psychiatry: Addiction as Failure Modes of the Decision-Making System
8.1 THE MACHINERY OF DECISION-MAKING
8.2 ADDICTION AS FAILURE MODES OF DECISION-MAKING SYSTEMS
8.3 BEYOND SIMPLE FAILURE MODES
8.4 RELIABILITY ENGINEERING
8.5 IMPLICATIONS FOR TREATMENT
9 - Modeling Negative Symptoms in Schizophrenia
9.1 INTRODUCTION: NEGATIVE SYMPTOMS IN SCHIZOPHRENIA
9.2 DOPAMINE SYSTEMS AND PREDICTION ERRORS
9.3 MODELING IN REWARD-RELATED DECISION TASKS
9.4 PROBABILISTIC STIMULUS SELECTION—COMBINED ACTOR-CRITIC/Q-LEARNING
9.4.4 Combined Actor-Critic/Q-Learning
9.5 TIME CONFLICT—TEMPORAL UTILITY INTEGRATION TASK
9.5.2 Time Conflict Model
9.6 PAVLOVIAN BIAS—EXTENDED Q-LEARNING
9.6.2 Pavlovian Bias Model
9.7 DIRECT ADDITION OF WORKING MEMORY TO REINFORCEMENT LEARNING MODELS
9.7.2 Reinforcement Learning and Working Memory Model
10 - Bayesian Approaches to Learning and Decision-Making
10.2 MARKOV DECISION PROBLEMS
10.2.2 Solving the Bellman Equation
10.2.2.1 Model-Free Temporal Difference Prediction Error Learning
10.2.2.2 Phasic Dopaminergic Signals
10.3.1 General Considerations
10.4 DISSECTING COMPONENTS OF DECISION-MAKING
10.4.2 Pavlovian Influences
10.4.3 Model-Based and Model-Free Decision-Making
11 - Computational Phenotypes Revealed by Interactive Economic Games
11.2 REINFORCEMENT LEARNING SYSTEMS AND THE VALUATION OF STATES AND ACTIONS
11.3 REACHING TOWARD HUMANS
11.4 COMPUTATIONAL PROBES OF PSYCHOPATHOLOGY USING HUMAN SOCIAL EXCHANGE: HUMAN BIOSENSOR APPROACHES
11.5 EPILOGUE: APPROACH AND AVOIDANCE IS NOT RICH ENOUGH
Computational Psychiatry
：Mathematical Modeling of Mental Illness
Disclaimer: Any content in publications that violate the sovereignty, the constitution or regulations of the PRC is not accepted or approved by CNPIEC.
