Feedback in the Acquisition of a Miniature Artificial Language

Abstract

The role of explicit informational feedback about semantic and syntactic errors was investigated in an experiment where university students were required to learn a miniature artificial language. A computer presented a series of simple geometric objects on a visual display screen, and alternately provided the proper description of the configuration (a model) or accepted and corrected the descriptions of the subject. The four experimental conditions were that (a) both semantic and syntactic errors were corrected; (b) only semantic; (c) only syntactic; (d) no errors were corrected. The results indicated that the subjects did not effectively utilize the feedback they received, and under all conditions learned the language from the model sentences which were presented.