Commentary: Conceptual Heterogeneity versus Developmental Homogeneity (on Chairs and Bears and Other Such Pairs)

Matching based on biological categories in Orangutans (Pongo abelii) and a Gorilla (Gorilla gorilla gorilla)

“…The authors found that humans easily discriminated concepts at all levels of abstraction, whereas both monkeys and pigeons had difficulty with the intermediate level discrimination. This finding was somewhat surprising because this intermediate level corresponds to the basic level that Rosch presumed was easiest for humans to learn (Keil, 1988; Mervis & Rosch, 1981; Rosch et al, 1976). Ross et al (2003) suggest that children acquire such concepts by the age of six years.…”

“…It has been suggested that the formation of superordinate categories relies less upon perceptual feature analysis and more on an understanding of how the category coheres, across significant perceptual variance (Spalding & Ross, 2000). Superordinate categories are thus thought of as being more conceptually based or “abstract” relative to basic level categories, and it has been speculated that language is necessary for the formation of these later categories (Benelli, 1988; Keil, 1988; Nelson, 1988; Premack, 1983). Organisms may be inherently preprogrammed to distinguish between items at the level of the basic concept perceptually (Cerella, 1979; Eimas & Quinn, 1994).…”

The Cognitive Basis of Uncertainty

“…In support of our account regarding equivalence classes are the findings that pre schoolers seem to organize their concepts around prototypes [Rosch, 1978], and that there occurs a 'characteristic-to-defining shift' in classification between the ages of 5 and 9 [Keil and Batterman, 1984; though sec Keil, 1989, for qualifications regarding this shift], Keil and Batterman found kindergar teners more likely to use characteristic fea tures as membership criteria (e.g.. 'gray hair' for 'grandfather'), whereas older children were more likely to use defining features (e.g., 'father of my parent'). Byrnes and Beilin [1987] asked children to sort the 8 causal sequence pairs employed in their study.…”

Discovering Linguistic Differences: Domain Specificity and the Young Child’s Awareness of Multiple Languages

“…Two areas, or domains, of knowl edge have been of particular interest: (1) the young child's ability to discriminate between and infer properties of natural kinds -par ticularly living kinds [Carey, 1985;Keil, 1986Keil, , 1988Gelman and Markman, 1987;Atran and Sperber, 1987], and (2) the young child's propensity to distinguish between certain kinds of social judgments -specifi cally moral, social-conventional and per sonal concepts [Turiel, 1983;Turiel and Davidson, 1986;Smetana, 1985: Song et al, 1987Nucci, 1981], Taken together, this re search lends support to three general conclu sions. First, different conceptual domains appear to follow distinct patterns of concept acquisition [Keil, 1988]; second, these do mains form the basis for rich inferential structures [Carey, 1985;Gelman and Markman, 1987]; and third, surprisingly general and uniform knowledge about the domains results from the young child's relatively brief, varying and casual exposure to phe nomenal cues [Keil, 1986;Smetana, 1985;Song et al, 1987;Atran and Sperber. 1987;Turiel and Davidson, 1986].…”

“…The evidence, however, suggests that the domain specificity associated with natural kind cate gories is structurally quite different from that associated with social cognition. Judg ments about natural kinds entail the concep tual discrimination of limited varieties, that is, the child's (and the adult's) more or less spontaneous recognition that the world con sists of a limited variety of kinds character ized in terms of their underlying essences [Atran, 1987;Keil, 1988], Moral and socialconventional judgments, in contrast, involve the child's mapping of distinct sorts of inter pretations onto a common set of social events, rather than representing a distinct way of classifying the social events them selves [Nucci, 1981, p. 120].…”