Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Learning nonlinearly separable categories by inference and classification.

Takashi Yamauchi1, Bradley C Love, Arthur B Markman

  • 1Department of Psychology, Texas A&M University, College Station 77843, USA. tya@psyc.tamu.edu

Journal of Experimental Psychology. Learning, Memory, and Cognition
|May 23, 2002
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Pharmacy operations within the Japan Disaster Relief (JDR) medical team during the 2025 Myanmar Earthquake: Practices, challenges, and lessons learned.

American journal of disaster medicine·2026
Same author

Action control in emotion regulation: when reappraisal may carry hidden costs.

Frontiers in psychology·2026
Same author

A Case of Thromboembolic Cerebral Infarction Occurring Perioperatively in Acute Lower Limb Embolic Ischemia.

Cureus·2026
Same author

The HIROSHIMA Study: A High-Volume, Institutional, Retrospective, Observational Study of Hidden Non-Incisional Suture Double Eyelid Surgery for Multivariate Analysis of Crease Loss.

Archives of plastic surgery·2026
Same author

Distinct Hippocampal Mechanisms Support Concept Formation and Updating.

The Journal of neuroscience : the official journal of the Society for Neuroscience·2026
Same author

Confidence-weighted integration of human and machine judgments for superior decision-making.

Patterns (New York, N.Y.)·2026
Same journal

Testing the predictions of a distinctiveness model of memory: The production effect in backward recall.

Journal of experimental psychology. Learning, memory, and cognition·2026
Same journal

On the impact of adjacency on transposed-word effects under serial presentation.

Journal of experimental psychology. Learning, memory, and cognition·2026
Same journal

It's time to opt out: Metacognitive analysis of time regulation under uncertainty.

Journal of experimental psychology. Learning, memory, and cognition·2026
Same journal

The role of statistical learning in attentional guidance during search through naturalistic scenes.

Journal of experimental psychology. Learning, memory, and cognition·2026
Same journal

Representing objects and features in long-term memory: A case for direct feature-feature binding.

Journal of experimental psychology. Learning, memory, and cognition·2026
Same journal

Crossmodal correspondences influence adaptation during rule-based category learning of objects.

Journal of experimental psychology. Learning, memory, and cognition·2026
See all related articles

Learning categories through classification is easier than through inference, especially for complex, nonlinearly separable categories. This is because category prototypes are less representative for these types of categories.

Area of Science:

  • Cognitive Psychology
  • Machine Learning Theory

Background:

  • Category learning research distinguishes between classification and inference learning.
  • Classification highlights discriminative features, while inference focuses on abstract category summaries (prototypes).
  • Prior studies used linearly separable categories, yielding different results.

Purpose of the Study:

  • To investigate how category structure (linearly vs. nonlinearly separable) affects classification and inference learning.
  • To test if the typical benefits of inference learning diminish with complex category structures.

Main Methods:

  • Participants learned categories using either classification or feature inference tasks.
  • Nonlinearly separable categories were employed to challenge existing models.

Related Experiment Videos

  • Performance was compared between the two learning procedures.

    • Main Results:

    • Learning nonlinearly separable categories via inference was significantly more difficult than via classification.
    • Category prototypes were poor summaries for nonlinearly separable categories.
    • Classification learning was more robust to category complexity.

    Conclusions:

    • The effectiveness of inference learning is contingent on category structure.
    • Cohesive category structures are crucial for successful inference learning.
    • Classification learning may be more advantageous for complex, nonlinearly separable categories.