In Cognitive Linguistics, 'conceptual blending' was proposed by Gilles Fauconnier and Mark Turner as a fundamental cognitive operation underlying much of everyday thought and language.
They described it as the process by which human cognition combines particular elements and relations of originally separate input mental spaces –"small conceptual packets constructed as we think and talk'"– that share some common structure, into a blended space, in which new elements and relations emerge, and novel inferences can be drawn. For instance, a `houseboat' or a `boathouse' are not merely the intersection of the concepts of `house' and `boat'. Instead, the concepts `houseboat' and `boathouse' selectively integrate different aspects of the source concepts in order to produce two new concepts, each with its own distinct internal structure. Leer mas...
DESCRIPTION
Although Fauconnier and Turner did not aim at a computational model of conceptual blending, and did not develop the sufficient details for this cognitive operation to be captured algorithmically, still their theory has proven to be a useful framework for studying creative thinking using a computational approach. That is the reason why a number of researchers in the field of Computational Creativity have recognised the potential value of conceptual blending for guiding the implementation of creative systems.
At IIIA we have developed uniform model of computational conceptual blending based on the notion of 'amalgam' from Case-Based Reasoning (CBR), which computes the combination of two cases in the CBR problem-solving process into a new 'blended' case through a process of generalisation and unification. We applied our model to computationally tackle the process of concept invention in mathematical creativity, music harmonisation and ontological modelling.