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

A hybrid model of categorization.

J R Anderson1, J Betz

  • 1Department of Psychology, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213-3890, USA. ja@cmu.edu

Psychonomic Bulletin & Review
|February 19, 2002
PubMed
Summary

This study integrates exemplar-based random walk (EBRW) and rule-plus-exception (RULEX) models within the ACT-R cognitive architecture. The combined system successfully models category learning by mixing both strategies, reproducing experimental data.

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

Searching for Chameleon Dark Energy with Mechanical Systems.

Physical review letters·2022
Same author

Developmental Ethanol-Induced Sleep Fragmentation, Behavioral Hyperactivity, Cognitive Impairment and Parvalbumin Cell Loss are Prevented by Lithium Co-treatment.

Neuroscience·2017
Same author

Thermocapillary convection around gas bubbles: an important natural effect for the enhancement of heat transfer in liquids under microgravity.

Annals of the New York Academy of Sciences·2002
Same author

Late effects of therapy in 94 patients with localized rhabdomyosarcoma of the orbit: Report from the Intergroup Rhabdomyosarcoma Study (IRS)-III, 1984-1991.

Medical and pediatric oncology·2000
Same author

Comparison of voxel- and volume-of-interest-based analyses in FDG PET scans of HIV positive and healthy individuals.

Journal of nuclear medicine : official publication, Society of Nuclear Medicine·2000
Same author

Aggressive surgery is unwarranted for biliary tract rhabdomyosarcoma.

Journal of pediatric surgery·2000

Area of Science:

  • Cognitive Science
  • Computational Neuroscience
  • Psychology

Background:

  • Category learning is traditionally explained by either exemplar-based or rule-based models.
  • Existing models like exemplar-based random walk (EBRW) and rule-plus-exception (RULEX) offer distinct explanations for classification behavior.

Purpose of the Study:

  • To investigate the integration of exemplar-based and rule-based category learning strategies within a unified cognitive architecture.
  • To demonstrate that a single architecture can accommodate and combine different learning mechanisms.

Main Methods:

  • Implemented variations of the exemplar-based random walk (EBRW) and rule-plus-exception (RULEX) models within the ACT-R cognitive architecture.
  • Developed a computational framework allowing for the mixing of exemplar and rule-based strategies.

Main Results:

  • The combined ACT-R system successfully reproduced behavioral data, including latency, learning rates, and generalization patterns, from three key category learning experiments.
  • Demonstrated that ACT-R can effectively integrate similarity and practice mechanisms, akin to EBRW, while also incorporating a strategy selection theory.

Conclusions:

  • Both exemplar-based and rule-based strategies can be effectively combined within the ACT-R cognitive architecture for category learning.
  • ACT-R provides a unified framework that reconciles different approaches to category learning and explains strategy selection.

Related Experiment Videos