Broadly speaking, this form of causality is one in which the cause reproduces itself over time. It can be symbolised in the following way: A <–> B (i.e., A effects B, just as B effects A) in contrast to linear causality A –> B (i.e., A is the antecedent to or causes B). Also, it is the rate at which [A effects B] changes as [B effects A]. Examples of circular causality from cybernetics are negative (deviation-reducing) and positive (deviation-amplifying) feedback. All told, it boils down to instances of self-regulation, a simple example based on negative feedback being a thermostat to maintain a constant room temperature. Going beyond such simple systems to self-organizing complex systems, circular causality is inferred in the two-way relationship between order and control parameters (i.e., an order parameter at the macroscopical level constrains the behavior of a control parameter at the microscopical level, but in turn a control parameter can change the state of an order parameter when it exceeds some critical value). This interpretation in terms of brain functioning is expressed by Walter J. Freeman as a top-down macroscopic state (e.g., attention) simultaneously influencing the actions of individual neurons at the microscopical level that in turn create and sustain the macroscopical state bottom up. A similar interpretation based on the principles of self-organization can be found in James J. Gibson‘s ecological psychology in which perception and action are depicted as being inseparably linked or coupled via circular causality.
See Causality (in philosophy), Control parameter, Cybernetics, Downward causation (or macrocausation), Ecological psychology, Feedback, Order parameter, Perception-action coupling, Self-organization, System