Fatalities and economic losses caused by floods are dramatically increasing in many regions of the world, and there is serious concern about future flood risk given the potentially negative effects of climatic and socio-economic changes. Over the past decades, numerous socio-economic studies have explored human responses to floods—demographic, policy and institutional changes following the occurrence of extreme events. Meanwhile, many hydrological studies have investigated human influences on floods, such as changes in frequency, magnitude, and spatial distribution of floods caused by urbanization or by implementation of risk reduction measures. Research in socio-hydrology is providing initial insights into the complex dynamics of risk resulting from the interplay (both responses and influences) between floods and people. Empirical research in this field has recently shown that traditional methods for flood risk assessment cannot capture the complex dynamics of risk emerging from mutual interactions and continuous feedback mechanisms between hydrological and social processes. It has also been shown that, while risk reduction strategies built on these traditional methods often work in the short term, they might lead to unintended consequences in the longer term. Besides empirical studies, a number of socio-hydrological models have been recently proposed to conceptualize human/flood interactions, to explain the dynamics emerging from this interplay, and to explore possible future trajectories of flood risk. Understanding the interplay between floods and societies can improve our ability to interpret flood risk changes over time and contribute to developing better policies and measures that will reduce the negative impacts of floods while maintaining the benefits of hydrological variability.