2001/12/31 by Angelo Cangelosi, Stevan Harnad · 2 citations
Social Sciences · Psychology · #Language and cultural evolution #Language, Metaphor, and Cognition #Child and Animal Learning Development
paper · doi:10.1075/eoc.4.1.07can
openalex publication_date 2001/12/31 · openalex created_date 2016/06/24 · openalex updated_date 2026/07/02
Using neural nets to simulate learning and the genetic algorithm to simulate\nevolution in a toy world of mushrooms and mushroom-foragers, we place two\nways of acquiring categories into direct competition with one another: In (1)\n"sensorimotor toil,” new categories are acquired through real-time, feedbackcorrected, trial and error experience in sorting them. In (2) "symbolic theft,” new\ncategories are acquired by hearsay from propositions – boolean combinations of\nsymbols describing them. In competition, symbolic theft always beats\nsensorimotor toil. We hypothesize that this is the basis of the adaptive advantage\nof language. Entry-level categories must still be learned by toil, however, to avoid\nan infinite regress (the “symbol grounding problem”). Changes in the internal\nrepresentations of categories must take place during the course of learning by toil.\nThese changes can be analyzed in terms of the compression of within-category\nsimilarities and the expansion of between-category differences. These allow\nregions of similarity space to be separated, bounded and named, and then the\nnames can be combined and recombined to describe new categories, grounded\nrecursively in the old ones. Such compression/expansion effects, called\n"categorical perception" (CP), have previously been reported with categories\nacquired by sensorimotor toil; we show that they can also arise from symbolic\ntheft alone. The picture of natural language and its origins that emerges from this\nanalysis is that of a powerful hybrid symbolic/sensorimotor capacity, infinitely\nsuperior to its purely sensorimotor precursors, but still grounded in and dependent\non them. It can spare us from untold time and effort learning things the hard way,\nthrough direct experience, but it remain anchored in and translatable into the\nlanguage of experience