Abstract
One of the most important and usually implicitly assumed models of animal concept discrimination–-the feature learning theory–-was explicitly tested with pigeons. In an experiment that employed a novel method of housing and testing, pigeons were trained to discriminate between two sets of Brunswik faces that differed along four dimensions, each represented by three values. The category rule was defined as an additive combination (sum) of the four feature values and thus resembles the polymorphous m-out-of-n rule. In the course of training, the birds tended to classify members of the categories according to their feature sum. This result was confirmed by the fact that the weightings assigned to each feature were similar in the final stage of training. A multiple regression analysis of the response rates enabled the classification decisions of the pigeons to be predicted from the linear feature model in accordance with feature learning theory.
Get full access to this article
View all access options for this article.
