Understanding the contribution of a Spatial Data Infrastructures (SDI) to society through improved economic, social and environmental outcomes implies an integrated and dynamic approach. To address this challenge, we propose a research framework that gathers theories and findings from a study concerning 45 French institutional SDIs. Its goal is to provide a logical structure for a holistic analysis of the complex components that affect the production and use of geographical information, their relationships, the dynamics of these relationships and the resulting outcomes. Two published frameworks, the Press-Pulse Dynamics and the Institutional Analysis and Development, are used as boundary objects during a three-workshop process gathering the research team. The result of this collective process leads to the design of an integrated conceptual framework for SDI research. It describes five main components of SDI (external drivers, a social component, patterns of interactions, a technical and informational component and outcomes), their relationships and research hypotheses. Guided by external press and internal pulse dynamics, the iterative framework redresses  This work is licensed under the Creative Commons Attribution-Non commercial Works 3.0 License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/3.0/ or send a letter to Creative Commons, 55 the balance of SDI components in favor of the social block and links this block to the technical and informational component through two bridges: patterns of interactions and outcomes. It shows how social norms and representations of different types of actors affect collective and individual actions, which then impact the technical and informational component of SDIs. In turn, technical and informational artifacts influence outcomes, thereby modifying human actions and initiating feedback that impacts the original dynamics and processes. We provide such a type of conceptual activity for assimilating large amounts of social and technical knowledge to strengthen our understanding of SDIs functioning and dynamics.< Réduire