Connectionism

Connectionism is a computational approach to the study of the mind that emerged in the context of cognitive science in the second half of the 1980s. However, its origins can be traced back to cybernetics. The connectionism of the 1980s arose in opposition to classical cognitive science and symbolic artificial intelligence. It is characterized by the adoption of parallel computational models inspired by the structure of the nervous system (artificial nerral networks). Today’s deep neural networks are, in all respects, a direct development of connectionist models. However, while connectionism originated as a project aimed at understanding the human mind, deep learning is primarily an enterprise with engineering and applied goals. Nevertheless, the recent striking successes of machine learning techniques have revived interest in neural networks as potential models of biological cognition and have revitalized many of the theoretical debates of past decades.