Surprise is a widely used concept describing a range of phenomena from unexpected events to behavioral responses. We propose a measure of surprise, to arrive at a new framework for surprise-driven learning. There are two components to this framework: (i) a confidence-adjusted surprise measure to capture environmental statistics as well as subjective beliefs, (ii) a surprise-minimization learning rule, or SMiLe-rule, which dynamically adjusts the balance between new and old information without making prior assumptions about the temporal statistics of the environment. We apply our framework to a dynamic decision-making task and a maze exploration task to demonstrate that it is suitable for learning in complex environments, even if the environment undergoes gradual or sudden changes. Our proposed surprise-modulated belief update algorithm provides a framework to study the behavior of humans and animals encountering surprising events.