The standard logical approaches to belief revision or scientific theory change assume either that the reality under investigation is static or at least that the ontic changes are un-correlated with the doxastic/epistemic change. But in numerous situations, the very act of learning new information may change the reality that is being learnt. Such situations were studied in Quantum Physics, Economics and Social Science, but have not been much investigated from the perspective of Philosophy of Science and the Logic of Theory Change. An example is the way in which an introspective agent changes her beliefs when learning new higher-order information, i.e. information that may refer to her own beliefs. A similar situation arises when a scientist learns about a phenomenon by performing measurements that perturb the very phenomenon under study. In Quantum Mechanics, this property that “observation causes perturbation” (the so-called observer effect) lies at the basis of most applications in quantum communication. We find similar examples in Psychology when a psychological test changes the very facts under investigation. More complex such scenarios of correlated information change occur in groups of communicating agents, whenever some agents’ knowledge about the others’ belief changes influence their own beliefs.What these examples have in common is that the very act of learning (individually or in group) can influence the results of learning, by changing the phenomena under study. In this project we develop a new unified logical setting to handle these different types of correlated information change in a multi-agent context. This setting is based on bringing together the insights and methods of Dynamic Epistemic Logic, Quantum Logic, Belief Revision Theory, Truth Approximation and Learning Theory. We plan to investigate applications of this setting to various areas of philosophy, ranging from social epistemology to philosophy of information and philosophy of science.