PARADIGM CHANGE IN COGNITIVE SCIENCES

Abstract

Since the 1950s, the dominant paradigm in the cognitive sciences has been cognitivism, which emerged as an alternative to behaviorism, and predominantly views cognitive processes as various kinds of “computations” similar to those performed by the computer. Despite significant advances made in the last quarter of the 20th century within this paradigm, it does not satisfy many scientists because it could not adequately explain some features of cognitive processes. Connectionism, which emerged somewhat later, recognizes the role of computational processes, but as their basis considers a neural network, which is a much better model of brain functioning than Turing-type computations. Neural networks, unlike the classical computer, demonstrate robustness and flexibility in the face of real-world problems, such as increased input noise, or blocked parts of the network. They are also well suited for tasks requiring the parallel resolution of multiple conflicting constraints. Despite this, the analogy between the functioning of the human brain and artificial neural networks is still limited due to radical differences in system design and associated capabilities. Parallel to the paradigms of cognitivism and connectionism, the notions that cognition is a consequence of purely biological processes of interaction between the organism and the environment have developed. These views, which have become increasingly popular in recent years, have taken shape in various currents of the so-called enactivism. This review compares the theoretical postulates of cognitivism, connectionism, and enactivism, as well as the predictive coding paradigm and the free energy principle.