cyclic causal discovery
Definition
Cyclic causal discovery refers to the process of inferring causal relationships from observational data when those relationships contain feedback loops or reciprocal influences, rather than assuming simple one-directional causality. Unlike acyclic causal discovery methods that rely on Directed Acyclic Graphs (DAGs), cyclic causal discovery methods are designed to identify cycles in causal structures—such as X causes Y and Y causes X—which are theoretically critical for understanding psychological phenomena like psychopathology. Constraint-based cyclic causal discovery methods use patterns of statistical (in)dependence to infer causal structures and can be applied using single observational datasets, though they present greater conceptual and practical difficulties in fitting and interpretation than their acyclic counterparts. Under certain conditions, cyclic causal models fit to cross-sectional data may be interpreted as reflecting causal relations between equilibriums or resting states of dynamic systems.
Sources: Park et al. (2024)
Related Terms
Applications
Cyclic Causal Discovery and Psychopathology
Cyclic causal relationships are theoretically essential for understanding psychopathology, as mental disorders are characterized by feedback loops and reciprocal relationships between symptoms.
Sources: Park et al. (2024)
Cyclic Causal Discovery and Statistical Network Models
Constraint-based cyclic causal discovery methods offer an approach for capturing causal structures relevant to psychological phenomena.
Sources: Park et al. (2024)
