Abstract
The role of errors in children's discrimination learning was explored. 108 second grade children mastered simultaneous intradimensional discriminations (line tilts) or interdimensional discriminations (line tilt and dot) in either an errorful or errorless fashion. Errorful learners acquired the discriminations with a trial and error procedure. Errorless learning was experimentally produced by use of a progressive S— fading procedure. Following acquisition all children received generalization tests along the line-tilt continuum. The post-discrimination generalization gradients for children trained on the intradimensional tasks demonstrated negative peak-shift effects and no positive peak-shift effects. The S— post-discrimination generalization gradients for children trained on the interdimensional tasks were flat indicating no S— control. No differences were noted in the post-discrimination generalization gradients for the errorful and errorless learners. It was concluded that young children can master a simultaneous discrimination without noticeable S— control and that making errors or responding to S— during simultaneous discrimination acquisition is not a sufficient condition for the establishment of S— dimensional control.
Get full access to this article
View all access options for this article.
