Effect of number of forced-choice primary generalization test trials on the establishment of dimensionally-based perceptual classes.
Item
-
Title
-
Effect of number of forced-choice primary generalization test trials on the establishment of dimensionally-based perceptual classes.
-
Identifier
-
AAI9830758
-
identifier
-
9830758
-
Creator
-
Reeve, Kenneth Francis.
-
Contributor
-
Adviser: Lanny Fields
-
Date
-
1998
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Psychology, Experimental | Psychology, Behavioral
-
Abstract
-
In Experiment 1, 20 college students learned two identity conditional discriminations with two squares that were filled with pixels comprising either 23 or 77% of the square's interior. This was followed by 0, 152, 456, or 760 two-choice generalization test trials. The samples were 19 squares with fill percentages that ranged from 23-77%. Fill23 and Fill77 were comparisons. Next, a generalization test with a third comparison, Neither, was presented. During three-choice generalization testing, both the area under the generalization gradients, and the range of test fill percentages that occasioned the exclusive selection of Fill23 or Fill77, were direct functions of the number of prior two-choice generalization trials. Next, a discriminability test revealed that Fill23 and Fill77 were discriminable from the test fill values. Thus, a range of filled squares functioned as members of dimensionally defined perceptual classes of low- or high-fill stimuli. In Experiment 2, five new students received 760 two-choice generalization trials for fill percentage, followed by three-choice generalization testing. Students were then trained to select a glyph in the presence of Fill23 and another glyph in the presence of Fill77. Next, a three-choice generalization test with the 19 fill stimuli as samples and the glyphs as comparisons was presented. These gradients overlapped with three-choice generalization gradients previously obtained with fill patterns as comparisons. Finally, a discriminability measure showed that adjacent stimuli along the dimension of fill were discriminable from each other. Thus, each perceptual class functioned as a transfer network for a new response.
-
Type
-
dissertation
-
Source
-
PQT Legacy CUNY.xlsx
-
degree
-
Ph.D.
