edVirtus Reference

What Are Causal-Loop and Influence Diagrams?

Causal-Loop and Influence Diagrams: Mapping Hypotheses About Circular Causation

Causal-loop diagrams and influence diagrams are graphical tools for representing hypothesised causal relationships in a system. Variables are connected by directed arrows, and link markings indicate how a change in one variable is expected to affect another when other relevant conditions are held constant.

A positive link means that the affected variable changes in the same direction as the cause, relative to what it otherwise would have done. A negative link means it changes in the opposite direction. These signs describe local causal influence, not whether a change is desirable or undesirable.

When a chain of influences returns to its starting variable, it forms a feedback loop. Reinforcing loops amplify change, while balancing loops oppose change or move a system toward a goal. Delays may be marked because they can generate overshoot, oscillation, or a misleading impression that an action has no effect.

The diagrams help teams expose mental models, compare explanations, and identify possible feedback structures before numerical formulation. An influence diagram may also show decision, information, or structural relationships without requiring every path to form a closed loop.

A diagram is a hypothesis, not proof. Ambiguous variable names, links that conceal several mechanisms, omitted accumulations, and diagrams containing every conceivable factor can reduce usefulness. Important claims should be supported by evidence and translated into stock-and-flow structure when simulation is required.

Used carefully, these diagrams provide a compact language for discussing why behaviour persists and where intervention might act. Their greatest value often lies in the reasoning and negotiation required to construct them, rather than in the finished picture alone.