A continuum between purely isolated and purely interrelated concepts i
s described. Along this continuum, a concept is interrelated to the ex
tent that it is influenced by other concepts. Methods for manipulating
and identifying a concept's degree of interrelatedness are introduced
. Relatively isolated concepts can be empirically identified by a rela
tively large use of nondiagnostic features, and by better categorizati
on performance for a concept's prototype than for a caricature of the
concept. Relatively interrelated concepts can be identified by minimal
use of nondiagnostic features, and by better categorization performan
ce for a caricature than for a prototype. A concept is likely to be re
latively isolated when subjects are instructed to create images for th
eir concepts rather than find discriminating features, when concepts a
re given unrelated labels, and when the categories that are displayed
alternate rarely between trials. The entire set of manipulations and m
easurements supports a graded distinction between isolated and interre
lated concepts. The distinction is applied to current models of catego
ry learning, and a connectionist framework for interpreting the empiri
cal results is presented.