ELEANOR ROSCH
Biography
Eleanor Rosch is currently a Professor of the Graduate School in the Department of Psychology, Program in Cognitive Science, and the Groups in Religious Studies and Buddhist Studies at the University of California, Berkeley. Her research in concepts and categorization challenged the once accepted Aristotelian view of concepts and word meaning as classical sets that could be combined by the logic of classical operations and offered evidence for the processing of concepts in many contexts in relation to their prototypical good examples. A second line of research has focused on what the teachings and practices of the traditions that our culture calls religions (such as Buddhist mindfulness) have to tell psychology about the mind. She is co-editor of the book Cognition and Categorization and co-author of The Embodied Mind: Cognitive Science and Human Experience.
Talk
Concepts: Are They Logically Fuzzy
Abstract: Concepts, categories, and word meaning had a long history of being understood as classical sets in which members could be combined by the operations of classical logic. However, beginning in the 1970s, my work and those of others has established that category membership is actually a gradient with members judged to have different degrees of goodness of example and with the boundaries of the category often indeterminate. Such gradients are psychologically important because they correlate with all the major dependent variables studied in psychology. Does this mean that concepts and word meaning are fuzzy? If so, conceptual combination should follow specifiable operations of fuzzy logic. However, there is much evidence that they do not; the understanding of conceptual combinations appears to depend on an individual’s mental encyclopedia of real world knowledge rather than a fixed and specifiable mental dictionary. We are left with some interesting questions: Do we want to call concepts and categories fuzzy sets because they have membership gradients even though one cannot perform fuzzy logical operations on them? And could fuzzy logic perform useful practical functions for limited domains of concepts in specific situations even if it does not explain concepts theoretically?