What is the causal relationship between crisis, learning and change? How does causality works in the current responses to the sovereign debt crisis provided by the European Union? We question the conventional identification of the cause-and-effect relationship provided by theories of crisis management, integration and policy learning. Drawing on models of contingent learning developed within psychology and behavioural and evolutionary economics, we theorise that surprise produces behavioural change via a fast-paced associative cue-outcome mechanism and that policy learning follows change. We then run our exercise in causal identification through a plausibility probe. We show that our argument passes the plausibility probe. Our conclusions on cognition and situational effects on learning suggest a new research agenda, more sensitive to how individuals behave in the real world and more robust in its micro-foundations.