The Royal Netherlands Academy of Arts and Sciences (KNAW) has awarded the C.L. de Carvalho-Heineken Prize for Cognitive Science 2014 to professor James McClelland, Director of the Center for Mind, Brain, and Computation at the University of Stanford (United States).
Professor McClelland received this award for his important and fundamental contributions to the use of neural networks to model cognitive processes of the brain.
James (‘Jay’) McClelland was born in Cambridge (U.S.) on 1 December 1948. He studied at Columbia University in New York City and got his Ph.D. in Cognitive Psychology from the University of Pennsylvania in Philadelphia in 1975.
The year before, McClelland had taken up a research position at the University of California at San Diego, while occasionally doing guest professorships at Harvard University and the Massachusetts Institute of Technology in Cambridge (U.S.).
In 1984, McClelland transferred to Carnegie Mellon University in Pittsburgh (U.S.). His seminal work Parallel Distributed Processing: Explorations in the Microstructure of Cognition, however, was published in 1986 together with David E. Rumelhart and other former colleagues from San Diego. Today, the book has been cited over 19,000 times.
McClelland became Lucie Stern Professor in the Social Sciences at the University of Stanford (U.S.) in 2006. He also leads Stanford’s Center for Mind, Brain, and Computation, which he has founded.
McClelland has received honorary doctorates from the Université Libre de Bruxelles and the New Bulgarian University in Sofia. He is member of the U.S. National Academy of Sciences and President of the Psychology Section of the American Association for the Advancement of Science.
Up until the 80’s of the previous century, the human brain was often compared to a simple computer. Cognition scientists modelled the brain as a processor that retrieves information from a memory, performs calculations and saves the results to a memory once again.
By publishing a book that is now seen as seminal to the field, James McClelland and colleagues ended this era in 1986. More comprehensively than others before them, they presented a new, universal model for cognitive processing. They called it ‘parallel distributed processing’ (PDP). Right away they showed how such a model could be applied to a wide range of neuropsychological problems.
The PDP model, these days also known as ‘connectionism’, was not based on computer technology but rather on the complex biological structure of the brain. In the brain, countless nerve cells communicate continuously through constantly changing interconnections. This ‘neural network’ of nodes and changing connections carries out many calculations at the same time. It is able to ‘learn’ by strengthening or weakening particular connections. Information does not have to be stored at a particular location but may reside in the overall state of the network.
With their universal model, McClelland and his colleagues not just changed the way we visualize the brain. They also provided the field of cognitive science with a set of mathematical principles that could be applied to research. Using these principles, scientists could rephrase countless research questions as clear, testable hypotheses.
McClelland himself used his universal model to try and explain how our brains are capable of performing complex cognitive processes such as learning to understand words.
Today, McClelland is still an influential and effective advocate of the ‘parallel distributed processing’ concept. Thanks in part to his persistence and his persuasiveness, the once controversial model has become highly influential and has inspired a generation of scientists, not just in cognitive science but in psychology and other adjacent fields as well.