Association for Behavior Analysis International

The Association for Behavior Analysis International® (ABAI) is a nonprofit membership organization with the mission to contribute to the well-being of society by developing, enhancing, and supporting the growth and vitality of the science of behavior analysis through research, education, and practice.

Search

33rd Annual Convention; San Diego, CA; 2007

Event Details


Previous Page

 

Paper Session #206
International Paper Session - Categorization
Sunday, May 27, 2007
1:30 PM–2:20 PM
Madeleine AB
Area: EAB
Chair: Ronald G. Weisman (Queens University)
 
Visual Categorization in Pigeons: Comparing Rule Based, Information Integration, and Associative Models.
Domain: Basic Research
MARK E. BERG (University of Canterbury), Randolph C. Grace (University of Canterbury)
 
Abstract: Eight pigeons were trained to categorize Gabor stimuli (sine wave gratings with contrast modulated by a circular Gaussian filter) presented on a touch-screen LCD panel that varied in terms of frequency and orientation. In the rule-based condition, accurate performance depended only on frequency, whereas in the information integration condition both frequency and orientation were relevant. After baseline training in each condition, transfer tests were conducted with novel stimuli. Results showed that all pigeons were able to learn the baseline category discriminations, although individual differences in performance were evident. Analyses are reported that will compare predictions of rule based and information integration models for the transfer tests, as well as predictions of an associative model based on reinforcement learning that we propose. The major goal is to test whether pigeons’ categorization performance can be described in terms of a single system, in contrast to humans’ performance under analogous conditions which has interpreted as providing evidence for multiple systems of category learning (Ashby & Maddox, 2004).
 
Teaching Concepts: Examples from Category Perception, Propagation, and Equivalence for the Natural Communication Signals.
Domain: Basic Research
RONALD G. WEISMAN (Queens University), Laurie Bloomfield (University of Alberta), Tara Farrell (University of Alberta), Marc Avery (University of Alberta ), Christopher Sturdy (University of Alberta)
 
Abstract: The most important perception is not that two things are different but rather that they are similar. The authors summarize progress in research on how songbirds (oscines) categorize their acoustic communication signals. Category perception for the learned songs and calls of oscines are well described by three rules: The exemplars of different vocal categories are more easily discriminable one from another than exemplars of the same vocal category. Vocal categorization transfers to novel exemplars of the same categories. Changes in outcome propagates back from novel to the original exemplars of a category demonstrating that songbirds are nearly effortless equivalence learners. Here we show that vocal communication is the conjoint adaptation of production and perception, so that the biacoustic features and auditory perception of note types and larger units such as songs and calls determine their categorization. Similarities to concepts in human speech are discussed.
 
 

BACK TO THE TOP

 

Back to Top
ValidatorError
  
Modifed by Eddie Soh
DONATE
{"isActive":false}