Please use this identifier to cite or link to this item:
Title: Covariational analysis of the correspondence between performance and verbal reports: A method of replication [An�lisis covariacional de la correspondencia entre ejecuci�n y reportes verbales: Un m�todo de replicaci�n]
Author: Martinez, H.
Carrillo, K.
Ortiz, G.
Issue Date: 1998
Abstract: Three studies investigated the effects of response feedback and different criterions of task progression on human relational learning. Forty-five college students were exposed to various phases of concept learning involving discrete trials. On each trial, the subject had to decide whether the presented item was an instance of the concept currently trained. Each study also included transfer tests with untrained concept instances. It was expected that the subjects would carry out the necessary covariations and display high accuracy on learning and transfer trials. However, the observed performances proved highly variable and inconsistent, with only weak evidence that subjects learned the expected relations. Analysis of individual data suggested that the subjects carried out covariations different from the ones planned by the experimenters. One experimenter tried to duplicate each subject's performance on a trial-by-trial basis, using the subject's past responses and verbal reports as information. Using this procedure, a high degree of correspondence with the data could be achieved. Although the advantages of this kind of procedure remain to be worked out, the achieved level of predictivity suggests (a) that subjects who do not show learning of an experimental task do not respond randomly but just carry out covariations different from the expected ones, and (b) that it is possible to identify these covariations.
Appears in Collections:Producción científica UdeG

Files in This Item:
There are no files associated with this item.

Items in RIUdeG are protected by copyright, with all rights reserved, unless otherwise indicated.