The influence of frequency of exposure on the learning of a phrase structural grammar

Abstract

Letter pairs generated by an elementary grammar were presented to college student Ss for recall. Ss were given 14 trials to learn 18 pairs, using a free recall technique. These Ss did not reach learning asymptote. Another group of Ss was given 12 pairs for 20 trials. Most Ss did not learn the pairs in these trials. The intrusions in the recall protocols suggest that the Ss did not learn the grammar from which the pairs were generated, but they learned the position of the letters in the pairs and generated sequences accordingly. Some Ss seem to have learned the presented pairs directly and these Ss tended to learn the pairs presented better. The authors concluded that mere exposure with intent to learn is not sufficient for the learning of grammatical structure.