Dissertation Defenses

Dissertation Defenses

Graduate College Home

For Future Students

For Current Students

For Faculty and Staff

About the Graduate College

Directory

Graduate Center for
Research and Retention

Theses and Dissertations

Events

Forms

Have a Question?
Ask the Graduate
College at our new
email address:
GRAD-Info@wmich.edu

Graduate Catalog

Doctoral Dissertation Announcement


Candidate: Matthew Porritt

Degree of: Doctor of Philosophy

Department: Psychology

Title: Fluency in Pigeons: The Effects of Response Rate on Learning

Committee:
Dr. Alan Poling, Chair
Dr. Brad Hoitema
Dr. Eric Fox
Dr. Kent Johnson

Date: Wednesday, March 21, 2007 1:30 p.m. – 3:30 p.m. 3713 Wood Hall

Abstract:
Instructional procedures that build high response rates are purported to have significant benefits over procedures that do not focus on fast responding (Binder, 1996). However, much of the research exploring rate-building procedures is confounded by two variables: rate of reinforcement and number of exposures (Doughty, Chase, & O’Shields, 2004). In six experiments, the present study examined the effects of three variables; rate of reinforcement, number of exposures, and rate of responding on learning under rate-building procedures in pigeons. In Experiment 1, rate-building produced significantly better learning outcomes when compared to a rate-controlled condition with the number of exposures held constant. In Experiments 2 and 3, significantly higher rates of reinforcement during rate-building conditions failed to produce differences in learning outcomes when response rates and the number of exposures were held constant. In Experiment 4, rate-building produced significantly better learning outcomes, despite holding rate of reinforcement, and the number of exposures were constant across rate-building and rate-controlled conditions. In Experiment 5, the arrangement of rate-controlled experimental procedures was changed to demonstrate the effect of decreasing latencies between responses within a trial (chain of responses) while controlling trial spacing. When rate-building was arranged within each trial, learning outcomes were significantly better than when rate of responding was controlled within and between trials.  Experiment 6 demonstrated that the effect of rate-controlled conditions was (over) due to decreased response rates and not an artifact of procedural manipulations. Overall, the results suggest that training responses to high rates enhances learning and support the use of rate-building procedures.

 

 

The Graduate College
Western Michigan University
Kalamazoo MI 5242 USA
(269) 387-8212 | (269) 387-8232 Fax