Referential effects on articulatory learning.

Abstract

The purpose of this investigation was to examine the effects of referent changes on the articulatory learning of familiar words. Two groups of school-age children learned correct articulatory productions of /s/ in /saIn/ and /saId/ in sequences ranging from single-word imitation to spontaneous production in connected discourse. One group was presented with new definitions and pictorial referents when learning the utterances /saIn/ and /saId/, while the other group learned /saIn/ and /saId/ in the familiar semantic context. The results suggest that the articulatory response /s/, in /saIn/ and /saId/, may be learned more effectively by an approach that assigns a nonsense referent to /saIn/ and /saId/, rather than an approach that maintains the original meaningful referent while training correct /s/ productions. Correct /s/ productions learned in this nonsense context, however, do not readily generalize to original semantic contexts. Additional training must be provided to establish these correct productions when the phonemic configurations, /saIn/ and /saId/, are under the stimulus control of the original, meaningful referent. Lastly, it appears that /s/ productions learned in the original context are generalized more easily to other meaningful words than /s/ productions learned in the nonsense context.