Gibson’s Theory of Perceptual Learning

This article describes the key ideas of the influential psychologist Eleanor J. Gibson, developed over 70 years of research with infants, children, adults, and a wide range of nonhuman species. Gibson’s ecological approach to perceptual learning and development describes how perception – extracting meaningful information from the environment to guide actions adaptively – improves with experience, the acquisition of new means of exploration, and the development of new perception– action systems. Learning to Perceive or Perceiving to Learn? Eleanor Jack Gibson built her theory of perceptual learning over a 70-year research career. She published her first paper on perceptual learning in 1932 (Gibson et al., 1932) and her last book in 2002 (Gibson, 2002). There is a clear thread from beginning to end, but she was not dogmatic in her ideas; her theories were always informed by data, and data collection was often inspired by real life and serendipity. Her theory is consistent with but not identical to James Gibson’s (1979) ecological approach to perception. Although the Gibsons were married and shared many arguments and ideas about perceptual learning and development, they wrote only five articles together (Gibson, 2002). Thus, in this article, ‘Gibson’ refers to Eleanor Gibson unless otherwise noted. Gibson’s 1969 book, Principles of Perceptual Learning and Development, described her theory in detail and jump-started a new field of inquiry. However, as new methods and findings became available, notably new ways of studying perception in young infants, Gibson questioned things that she had once taken for granted and broadened her domain of inquiry. In later writings, Gibson critiqued her 1969 account for failing to capture how infants learn to detect the perceptual information for guiding action adaptively as their perceptual-motor systems are developing. In fact, she considered infancy to be the perfect place to study perceptual learning (Gibson, 1992) and she was instrumental in building the field of infant perception (Pick, 1992). On Gibson’s (1969) account, perceptual learning entails an increased ability to extract relevant information from a stimulus array as the result of experience. The traditional view of perceptual learning, dating back to Bishop Berkeley in the 1700s, is that animals must learn to perceive; the information at sensory receptors is impoverished and meaningless and thus a complete percept requires learning. In Gibson’s view, the information at receptors is sufficient to support complete percepts from the start, and thus animals need not learn to perceive; rather, they perceive to learn (Gibson, July 1989). Perceptual learning is the key to knowledge and where it all begins. Gibson (1992) maintained that a theory of perceptual learning must answer basic questions: What is learned and what is the function? What instigates learning and what terminates the process? As for the question of mechanism, Gibson held that meaningful explanations of psychological processes must be at the level of behavior. She always maintained that the job of a perception psychologist is to describe and explain perception at the level of individual behavior, not in terms of the underlying neurophysiology (Gibson and Pick, 2000). Perceptual processes should be consistent with what we know about neural mechanisms and physiology, but these facts should not drive research into perceptual processes. Rather, the appropriate research program should begin with a consideration of what there is to be perceived in an ecological context – what animals need to perceive so as to act adaptively in the natural environments in which they evolved and in the changing environments in which they develop. Starting Assumptions Gibson’s theory grows out of her commitment to several related starting assumptions that guided her thoughts about perceptual learning and shaped the trajectory of her research career. She believed that J.J. Gibson’s (1966, 1979) ecological approach to perception is the proper starting place for a theory of perceptual learning and development (Gibson and Pick, 2000). Animals must be considered doing the things that they naturally do in their particular ecological niche. Their activities are guided by perceptual information and such information is available in the ambient arrays of energy to which their perceptual systems are sensitive. Because perception involves the whole animal, nested in its environment, her approach was necessarily a systems approach. The total system includes animals in the environment that surrounds them in the here and now and which has shaped them in the course of evolution and development, and which they in turn influence (Gibson, 1992). Because of her focus on animals in their species-typical environments, hers was a comparative approach. In contrast to the traditional practice of focusing nearly exclusively on humans and the chambered eye, she maintained that perception psychologists should consider the perceptual systems and actions of varied species as they build theories of perceptual learning and development in humans (Gibson and Walk, 1960; Walk and Gibson, 1961). All animals have solved the problem of using perceptual information to guide activity: houseflies with compound eyes, fish using International Encyclopedia of the Social & Behavioral Sciences, 2nd edition, Volume 10 http://dx.doi.org/10.1016/B978-0-08-097086-8.23096-1 127 International Encyclopedia of the Social & Behavioral Sciences, Second Edition, 2015, 127–134 Author's personal copy