The influence of training structure and instructions on generalized stimulus equivalence classes and typicality effects

UNCW Author/Contributor (non-UNCW co-authors, if there are any, appear on document)
Kelly N. Stanley (Creator)
Institution
The University of North Carolina Wilmington (UNCW )
Web Site: http://library.uncw.edu/
Advisor
Carol Pilgrim

Abstract: Three experiments were conducted using different training structures and two sets of instructions in an attempt to systematically replicate previous findings of equivalence and typicality effects. Although each experiment used a different training structure, all training included the use of nonsense syllables and eight sets of abstract stimuli. The abstract stimuli differed along a number of features, but only four of these features were consistent across the trained classes. Each stimulus included one, two, three, and four of the class consistent features which defined the typicality of the stimulus. In Experiment 1, subjects were given a detailed set of instructions and used a many-to-one training structure where the abstract stimuli were samples and the nonsense syllables were comparisons. In Experiments 2 and 3, subjects were given either detailed or minimal instructions and used a one-to-many training structure where one of the abstract shapes served as the sample and the syllables and other abstract shapes were the comparisons. Symmetry and transitivity probe tests were used to assess the formation of equivalence while novel probe tests, which presented new abstract stimuli, tested the subject’s ability to generalize the class consistent features to the new examples. Subjects in Experiment 1 showed generalized equivalence but failed to show typicality effects in latency data. Subjects in Experiment 2 failed to demonstrate equivalence. Results from Experiment 3 showed the formation of generalized equivalence and typicality in subjects that received the detailed instructions. These findings reveal that language-like equivalence classes and typicality effects can be produced from more than one training structure.

Additional Information

Publication
Thesis
A Thesis Submitted to the University of North Carolina at Wilmington in Partial Fulfillment Of the Requirements for the Degree of Master of Arts
Language: English
Date: 2009
Keywords
Cognitive learning, Reinforcement (Psychology), Stimulus generalization
Subjects
Stimulus generalization
Reinforcement (Psychology)
Cognitive learning

Email this document to

This item contains the following parts:

TitleLocation & LinkType of Relationship
Title page, table of contents and abstracthttp://libres.uncg.edu/ir/uncw/f/stanleyk2004-1.pdfThe described resource includes the related resource either physically or logically.
Referenceshttp://libres.uncg.edu/ir/uncw/f/stanleyk2004-3.pdfThe described resource includes the related resource either physically or logically.