Effective information

Effective information is a measure developed by Giulio Tononi and collaborators in the context of Integrated Information Theory (IIT) to quantify the causal power of a system's parts over its whole. Unlike mutual information, which measures statistical correlation, effective information measures the difference a mechanism makes: how much the system's past constrains its present, above and beyond what the parts would do independently.\n\nThe concept has been generalized beyond IIT as a tool for identifying causal structure in complex systems. It distinguishes genuine emergence — where the whole causally constrains the parts — from mere statistical aggregation. Critics argue that the measure is computationally intractable for all but the smallest systems and that its dependence on a specific interventionist framework ("perturb every state and measure the effect") makes it more a thought experiment than a practical tool. The deeper question is whether causal power can be quantified at all without first assuming the very system boundaries it is meant to discover.\n\nSee also: Integrated Information Theory, Causality, Emergence, Complex systems theory, Causal emergence\n\n