Spatial processes and Galois/concept lattices

There is a fair amount of interest in the use of Galois or Concept Lattices in the social sciences as a way of representing complex data structures; it does not seem to be widely appreciated that these structures are homologous to those arising from a discrete noncompensatory multidimensional response process, as originally outlined by Coombs. Here I formalize the spatial intuition underlying Coombs’s proposal, and show how, for any set of data, we may recreate the set of all possible minimal spatial representations. Copyright Springer Science+Business Media Dordrecht 2014