The effects of noninformative stimulus elements on pattern recognition by pigeons.
Item
-
Title
-
The effects of noninformative stimulus elements on pattern recognition by pigeons.
-
Identifier
-
AAI9130312
-
identifier
-
9130312
-
Creator
-
Donis, Francisco J.
-
Contributor
-
Adviser: Eric G. Heinemann
-
Date
-
1991
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Psychology, Experimental
-
Abstract
-
Pigeons (columba livia) were trained to discriminate between simple forms ("targets"), displayed on the screen of a computer monitor. In several experiments identical forms ("flankers", and) accompanied each of the targets. In experiments done with human subjects some flankers have been found to enhance the discrimination of the target forms under some circumstances (the "context superiority effect"), and impair the discrimination under others (the "distractor effect"). The purpose of the present study were to isolate the conditions giving rise to either context superiority or distractor effects in pigeons, and to test predictions from the Heinemann and Chase model of pattern recognition by pigeons.;In all the experiments reported here only one stimulus, randomly selected from the set of stimuli presented for discrimination, was shown on each trial. One of two choice responses (a peck on one of two disks displayed on the monitor) was rewarded and counted as a correct response for each stimulus. The independent variable manipulated was the distance between the target patterns and the flankers, and the dependent variable was the proportion of correct responses.;In one experiment it was found that adding flankers to two previously learned patterns (and) produced a decrement in their discrimination. In the rest of the experiments flankers accompanied the target patterns from the beginning of training. In most of these experiments the patterns to be discriminated were white presented in an environment that was otherwise totally dark. Among the patterns studied were two oblique lines, and the same lines to which an identical L-shaped right angle form was added. In contrast to the "context superiority effect" found with human subjects in this situation, the pigeons studied in the present experiments discriminated significantly better between the oblique lines when presented alone than when presented with context.;In other experiments each target consisted of two small squares. One pair of targets used may be described as the end points of two vertical lines differing by 5 mm in their placement along the horizontal axis on the screen, and the other pair as the end points of two oblique lines.;The subjects were unable to discriminate the position of the "vertical line" targets when shown alone, but were able to discriminate the two targets when accompanied by flankers. An opposite effect was found for the "oblique line" targets which were well discriminated when presented alone but poorly discriminated when accompanied by flankers.;Thus, the flankers can produce either a context superiority effect or a distractor effect depending on the discriminability of the target patterns. The magnitude of each of these effects was approximately a linear function of the logarithm of target-flanker separation.
-
Type
-
dissertation
-
Source
-
PQT Legacy CUNY.xlsx
-
degree
-
Ph.D.
