Selective auditory attention in adults: Effects of rhythmic structure of the competing language
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Research to date has provided limited evidence related to how selective auditory attention is affected by: 1) differences between the vocal and/or speaking characteristics of different masking talkers and 2) rhythmic properties of the language of the competing (or masking) speech signal. The current study addressed these gaps in existing research by measuring sentence reception thresholds (sSRTs) for 50 monolingual English-speaking adults in conditions where the competing speech was spoken in a language with a stress-based rhythm (i.e., English or German), a syllable-based rhythm (i.e., Spanish or French), or a mora-based rhythm (i.e., Japanese) (Abercrombie, 1967; Pike, 1945; both as cited in Nazzi, Jusczyk, & Johnson, 2000). In all conditions, the target speech signal consisted of meaningful English sentences spoken by a male talker; whereas, the competing speech signal consisted of continuous speech from 1 male and 1 female native speaker of the language assigned to that condition. For each language, 2 different 2-talker masking signals were created (i.e., labeled Masker A and Masker B), with a different male-female pair of native speakers used for each signal. An average sSRT was obtained in 10 types of competing speech conditions, including 1 Masker A condition and 1 Masker B condition for each of the 5 languages.
Analysis of results revealed that for each language, there was a significant difference between subjects’ performance in the Masker A and Masker B conditions. These findings suggest that the vocal/speaking characteristics of different mixed-gender 2-talker maskers (i.e., 1 male and 1 female masking talker) can provide different degrees of masking effectiveness, regardless of whether the masking talkers are speaking the listeners’ native language or an unfamiliar language. A significant effect of the language of the competing speech was also seen, with significantly greater listening difficulty associated with conditions where the competing speech was spoken in English, German, or Japanese, as compared to Spanish or French. This pattern of results leads to 2 primary conclusions: 1) the underlying rhythmic structure of the language of the competing speech has a greater impact on masking effectiveness than the meaning of the competing speech, and 2) competing speech spoken in a language from the listener’s native rhythmic category (i.e., stress-based) is more difficult to ignore than competing speech spoken in a language from a non-native rhythmic category (i.e., syllable-based), at least for some languages. Together, these results extend the findings of previous studies by providing evidence that exposure to the rhythm of the native language affects not only how a listener processes speech in quiet (e.g., Cutler, Mehler, Norris, & Segui, 1983; Sebastián-Gallés, Dupoux, Costa, & Mehler, 2000), but also how he/she processes a target speech signal in the presence of speech from 2 masking talkers.