Analyzing multidimensional formal dynamic models in psychology: A tutorial using graphical tools

Area of Science:

• Clinical Psychology
• Mathematical Psychology
• Dynamical Systems Theory

Background:

• Formal theories in clinical psychology use mathematical representations (e.g., differential equations) to enhance deductive power and generate hypotheses.
• Current reliance on simulations for evaluating formal models lacks precise links between mathematical properties and dynamic outcomes.
• Discrepancies between model predictions and clinical observations are difficult to resolve with simulation-based approaches alone.

Purpose of the Study:

• To introduce formal mathematical techniques for graphical analysis of dynamic psychological models.
• To demonstrate the utility of phase plane and bifurcation analysis for understanding model behavior.
• To provide an accessible tool (deBif R package) for applying these graphical methods.

Main Methods:

• Phase plane analysis to identify stable and unstable equilibria in dynamical systems.
• Bifurcation analysis to map parameter regimes to qualitatively different model outcomes.
• Application of these methods to formal dynamic models of panic disorder and suicidal ideation.

Main Results:

• Graphical analysis provides a precise link between mathematical model properties and dynamic outcomes.
• Phase plane and bifurcation analyses reveal critical parameter values and system behaviors.
• The deBif R package facilitates the application of these advanced analytical techniques.

Conclusions:

• Graphical model analysis offers superior insight into psychological systems compared to simulations alone.
• These methods aid in identifying discrepancies between models and clinical data, guiding model improvement.
• The study underscores the importance of graphical tools for investigating hypothesized mechanisms in psychological dynamics.