Connectionism is an approach to the study of human mental processes and cognition that uses mathematical models called connectionist networks, or artificial neural networks, built from many simple interconnected units, standing as an alternative to traditional symbolic accounts of the mind. The approach has developed through successive waves: an initial wave beginning in 1943 with Warren McCulloch and Walter Pitts applying formal mathematical methods to neural circuitry and continuing through Frank Rosenblatt's 1958 perceptron; a second wave from the late 1980s following the 1987 publication of Parallel Distributed Processing, which introduced hidden layers and other refinements; and a current wave built on deep learning.
Facts
Core ClaimMental processes can be modeled as patterns of activation spreading across networks of many simple interconnected units, rather than as manipulation of explicit symbolic rules. 1 Origin YearMarks the first wave of connectionist modeling, when Warren McCulloch and Walter Pitts applied formal mathematical methods to neural circuitry; Frank Rosenblatt's perceptron followed in 1958. FAQs
Why did connectionist research go quiet between its first and second waves?
A 1969 book on the limits of the original perceptron discouraged funders, and most work paused until the mid-1980s.
The first wave of connectionism ended with the 1969 book Perceptrons by Marvin Minsky and Seymour Papert, which set out limitations of the original perceptron idea. The book contributed to discouraging major funding agencies in the United States from investing in connectionist research, and with a few noteworthy deviations most of the research became inactive until the mid-1980s. The second wave followed the 1987 book Parallel Distributed Processing, which added intermediate processors now called hidden layers between the input and output units.
Sources
1. Wikipedia: Connectionism
Wikimedia Foundationlead paragraph
Connectionism is an approach to the study of human mental processes and cognition that utilizes mathematical models known as connectionist networks or artificial neural networks.
History section, first wave
The first wave appeared 1943 with Warren Sturgis McCulloch and Walter Pitts both focusing on comprehending neural circuitry through a formal and mathematical approach.
In Group: Information Processing and Cognitive Architecture Theories, Lead
Connectionism is an approach to the study of human mental processes and cognition that utilizes mathematical models known as connectionist networks or artificial neural networks.
View the Source Reader Challenges (0)
No disputes yet. Spotted an error or a better source? Open the first one.
Sign in to dispute this or suggest a correction.