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

Perceptual categorization: connectionist modelling and decision rules

F W Jones1, A J Wills, I P McLaren

  • 1Psychological Laboratory, University of Cambridge, U.K. fwj1000@cus.cam.ac.uk

The Quarterly Journal of Experimental Psychology. B, Comparative and Physiological Psychology
|April 9, 1998
PubMed
Summary

This study explores how connectionist networks model human associative learning. The winner-take-all model best predicts response probabilities and potential response latencies in categorization tasks.

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

The human jaws and dentition.

The Dental record·2014
Same author

Associations and propositions: the case for a dual-process account of learning in humans.

Neurobiology of learning and memory·2013
Same author

Modeling human sequence learning under incidental conditions.

Journal of experimental psychology. Animal behavior processes·2013
Same author

Implementing video conferencing in mental health practice.

Journal of psychiatric and mental health nursing·2012
Same author

Dissociating expectancy of shock and changes in skin conductance: an investigation of the Perruchet effect using an electrodermal paradigm.

Journal of experimental psychology. Animal behavior processes·2012
Same author

Animal Learning and Cognition: A neural network approach.

Trends in cognitive sciences·2011

Area of Science:

  • Cognitive Science
  • Computational Neuroscience
  • Machine Learning

Background:

  • Connectionist networks are popular for modeling human associative learning.
  • The conversion of network output activations to response probabilities is under-explored.

Purpose of the Study:

  • To investigate and dissociate models of decision processes in associative learning.
  • To compare different models including ratio rules and winner-take-all architectures.

Main Methods:

  • Two categorization experiments were designed to differentiate decision models.
  • Analogues of experiments were simulated using a single-layer, feed-forward, delta-rule network.

Main Results:

  • The exponential ratio rule and winner-take-all models accurately predicted mean response results.

Related Experiment Videos

  • The winner-take-all model showed potential for predicting response latencies, unlike the exponential ratio rule.

    • Conclusions:

    • The winner-take-all architecture is a promising model for associative learning, potentially explaining both response probabilities and latencies.
    • Further research is needed to validate winner-take-all predictions for response latencies under stricter conditions.