Abstract
Kittur et al. (2004, 2006) and Jung and Hummel (2011, 2014) showed that people have great difficulty learning relation-based categories with a probabilistic (i.e., family resemblance) structure, in which no single relation is shared by all members of a category. Yet acquisition of such categories is not strictly impossible: in all these studies, roughly half the participants eventually learned to criterion. What are these participants doing that the other half are not? We hypothesized that successful participants were those who divided the nominal categories into two or more sub-categories, each of which individually had a deterministic structure. We report three experiments testing this hypothesis: explicitly presenting participants with hierarchical (category and sub-category) structures facilitated the acquisition of otherwise probabilistic relational categories, but only when participants learned the subordinate-level (i.e., deterministic) categories prior to learning the nominal (i.e., probabilistic) categories and only when they were permitted to view multiple exemplars of the same category simultaneously. These findings suggest that one way to learn natural relational categories with a probabilistic structure [e.g., Wittgenstein’s (1953), category game, or even mother] is by learning deterministic subordinate-level concepts first and connecting them together under a common concept or label. They also add to the literature suggesting that comparison of multiple exemplars plays an instrumental role in relational learning.
Highlights
One of the most generally accepted assumptions in the literature on categorization and category learning is that categories and exemplars are mentally represented as lists of features and that the process of assigning exemplars to categories is based on comparing their features
EXPERIMENT 2 If deterministic subordinate-level learning is to facilitate probabilistic basic-level learning, it seems necessary for the subordinate-level learning to temporally precede the basic-level learning2
In the basic-level first with comparison condition of Experiment 2, participants were trained to classify exemplars at the probabilistic basic level before classifying them at the deterministic subordinate level. This experiment investigated the effect of subordinatelevel comparison without subordinate-level category learning: in the basic-level only with comparison condition of this experiment, participants viewed pairs of exemplars that would have belonged to the same subordinate-level category, but only learned to classify them at the basic level
Summary
One of the most generally accepted assumptions in the literature on categorization and category learning is that categories and exemplars are mentally represented as lists of features and that the process of assigning exemplars to categories is based on comparing their features (for reviews, see Murphy, 2002; Kittur et al, 2006). As pointed out by Barsalou (1983), Gentner (1983), Murphy and Medin (1985) and others, one limitation of this view is that many concepts and categories are based, not on the literal features of their exemplars, but on relations— either relations among an exemplar’s features (e.g., arranged in one way, the parts of a folding bed form a bed, but arranged in another, they form a couch; Biederman, 1987; Hummel and Biederman, 1992) or relations between the exemplar and other objects in the world If they are learned in different ways, little or nothing we know about the acquisition of feature-based categories will necessarily apply to the case of relational concepts and categories
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call