is regarded as an implicit premise in relevance theory. A Bridging Inference does not directly yield expected cognitive effects in its own right but contributes to relevance by providing access to the intended explicature, which in turn yields cognitive effects.
This definition was taken from: T. Matsui. Modeling and Using Context (Proceedings of the conference CONTEXT 2001, Dundee, UK, July), volume 2116 of Lecture Notes in Artificial Intelligence, chapter Experimental Pragmatics: Towards Testing Relevance-Based Predictions about Anaphoric Bridging Inferences, pages 248–258. Springer, Berlin, 2001.