Explaining Basic Categories: Feature Predictability and Information

  title={Explaining Basic Categories: Feature Predictability and Information},
  author={James E. Corter and Mark A. Gluck},
  journal={Psychological Bulletin},
The category utility hypothesis holds that categories are useful because they can be used to predict the features of instances and that the categories that tend to survive and become preferred in a culture (basic-level categories) are those that best improve the category users' ability to perform this function. Starting from this hypothesis, a quantitative measure of the utility of a category is derived. Application to the special case of substitutive attributes is described. The measure is… 

Figures and Tables from this paper

Predicting the Basic Level in a Hierarchy of Concepts
This paper examines to what extent the basic level can be learned from data, and test the utility of three types of concept features, that were inspired by thebasic level theory: lexical features, structural features and frequency features.
Is it a Fruit, an Apple or a Granny Smith? Predicting the Basic Level in a Concept Hierarchy
It is argued that applications that use concept hierarchies - such as knowledge graphs, ontologies or taxonomies - could significantly improve their user interfaces if they `knew' which concepts are the basic level concepts.
Basic-level categories: A review
Application of Heidegger’s notion of handiness as a framework for understanding the relational nature of basic-level categories is discussed.
Optimal categorization
A simplicity principle in unsupervised human categorization
Knowledge categorization affects popularity and quality of Wikipedia articles
It is shown that Wikipedia articles assigned to coarse-grained categories garner more attention from Wikipedia editors, but receive lower evaluations; the negative relation between attention and quality implied by this result is consistent with current theories of social categorization.
Occam and Bayes in predicting category intuitiveness
  • E. Pothos
  • Mathematics
    Artificial Intelligence Review
  • 2009
Two models of unsupervised categorization are considered, the simplicity model and the rational model, and their predictions were assessed with a series of artificial datasets, such that each dataset was designed to reflect a simple intuition about human categorization processes.
Inference using categories.
The results from 4 experiments suggest that category membership influences inference even when similarity information contradicts the category label and that category labels and category features are 2 different things.
Category coherence and category-based property induction
Bayesian models of category acquisition and meaning development
This thesis focuses on categories acquired from natural language stimuli, using nouns as a stand-in for their reference concepts, and their linguistic contexts as a representation of the concepts’ features, and presents a Bayesian model which jointly learns categories and structured featural representations.


Context theory of classification learning.
A context theory of classificatio n is described in which judgments are assumed to derive exclusively from stored exemplar information, and the main idea is that a probe item acts as a retrieval cue to access information associated with stimuli similar to the probe.
Not the Path to Perdition: The Utility of Similarity-Based Learning
Four reasons why robust machine learning must involve the integration of similarity-based and explanation-based methods are presented, and it is argued that it may not always be practical or even possible to determine a causal explanation.
The role of theories in conceptual coherence.
It is proposed that concepts are coherent to the extent that they fit people's background knowledge or naive theories about the world and to structure the attributes that are internal to a concept.
Features of Similarity
The metric and dimensional assumptions that underlie the geometric representation of similarity are questioned on both theoretical and empirical grounds and a set of qualitative assumptions are shown to imply the contrast model, which expresses the similarity between objects as a linear combination of the measures of their common and distinctive features.
Correlated symptoms and simulated medical classification.
Across four experiments, subjects proved to be sensitive to configural information and tended to choose the case that preserved the correlation over the cases that broke the correlation, even when the case with correlated symptoms contained fewer typical symptoms.
Objects, parts, and categories.
Data is presented showing that the basic level differs qualitatively from other levels in taxonomies of objects and of living things and an explanation for why so many measures converge at that level is presented.
Basic-level superiority in picture categorization
Order of acquisition of subordinate-, basic-, and superordinate-level categories.
MERVIS, CAROLYN B., and CRIsAFI, MARIA A. Order of Acquisition of Subordinate-, Basic-, and Superordinate-Level Categories. CHILD DEVELOPMENT, 1982, 53, 258-266. The present research was concerned
Learning at the knowledge level
The paper analyzes two classes of learning programs, called symbol level learning and nondeductive knowledge level learning, and speculates on the possibility of developing coherent theories of each.