The Effect of Background Language and Speaker Differences on Serial Recall of Aurally Presented Single-Talker Stimuli

Abstract

Previous studies suggest that hearing a sound during the completion of a working memory task negatively affects recall performance. This is true for a wide variety of competing background sounds including multitalker babble, unfamiliar-language multitalker babble, white noise, and reversed speech. It is also suggested that the closer the background speaker is to the target speaker physically, the more difficult it is to separate the two speech streams, which results in lowered recall performance. The present study investigates serial recall performance of English words in single-talker background speech when the background language is either the listeners' native language (English) or an unknown language (Korean). Moreover, the effect of speaker differences is tested by using the same (bilingual) speaker for both foreground and background in one condition and introducing a novel foreground speaker in a second condition. An effect of background language did not emerge overall, but on some lists, Korean background speech is less distracting than English background speech. The comparison of recall performance between conditions revealed that there is not a significant difference overall when the background speaker is either the same as or different from the foreground speaker. On individual lists, when the background speaker is physically different from the foreground speaker, listeners recall significantly more words than when the background and foreground speaker are the same. These results support previous work and provide evidence for theories on target-masker separation.