"Many philosophers claim that the neurocomputational framework of predictive processing entails a globally inferentialist and representationalist view of cognition," says Marco Facchin. "Here, I contend that this is not correct." Quite right. Indeed, I have said in the past that this is what distinguishes connectivism from cognitivist theories such as constructivism and direct instruction. "Cognitive science really seems strongly committed to representationalism," writes Facchim. But there's nothing in cognitive science, even as it is practiced today, that commits us to things like working memory, concept formation and structural representations. This is a preprint; the article is also behind a stupid paywall here. See also: Daniel Williams, Predictive Processing and the Representation Wars; Representation in Cognitive Science by Nicholas Shea: Reply by the Author; and Constant, Clark and Friston, Representation Wars: Enacting an Armistice Through Active Inference.
Today: 2 Total: 102 [Direct link] [Share]
