作者: Mark Steyvers , Robert L. Goldstone , Alan Kersten , Jesse Spencer-Smith
DOI:
关键词: Set (psychology) 、 Artificial intelligence 、 Perceptual learning 、 Cognitive science 、 Vocabulary 、 Concept learning 、 Perception 、 Visual Objects 、 Perceptual system 、 Direct and indirect realism 、 Psychology
摘要: Confusions arise when 'stable' is equated with 'foundational.' Spurred on by the image of a house`s foundation, it tempting to think that something provides effective support extent rigid and stable. We will argue considering role perception in grounding our concepts, exactly opposite true. Our perceptual system supports ability acquire new concepts being flexibly tuned these concepts. Whereas we learn are certainly influenced representations, representations also learned In keeping one central themes this book, behavioral adaptability completely consistent representationalism. fact, most straightforward account experimental results concept learning can produce changes 'vocabulary' features, used subsequent tasks. This chapter reviews theoretical empirical evidence vocabularies describe visual objects adapted demands their user. extend arguments made elsewhere for adaptive (Goldstone, Schyns, & Medin, 1998; Goldstone, Thibaut, 1998), discuss research from laboratory illustrating specific interactions between conceptual learning. computer simulations provide accounts using neural network models. These models have detectors become increasingly set features The bulk be organized around mechanisms human learning, mechanisms.