Coping with adversity: Individual differences in the perception of noisy and accented speech

Abstract

During speech communication, both environmental noise and nonnative accents can create adverse conditions for the listener. Individuals recruit additional cognitive, linguistic, and/or perceptual resources when faced with such challenges. Furthermore, listeners vary in their ability to understand speech in adverse conditions. In the present study, we compared individuals' receptive vocabulary, inhibition, rhythm perception, and working memory with transcription accuracy (i.e., intelligibility scores) for four adverse listening conditions: native speech in speech-shaped noise, native speech with a single-talker masker, nonnative-accented speech in quiet, and nonnative-accented speech in speech-shaped noise. The results showed that intelligibility scores for similar types of adverse listening conditions (i.e., with the same environmental noise or nonnative-accented speech) significantly correlated with one another. Furthermore, receptive vocabulary positively predicted performance globally across adverse listening conditions, and working memory positively predicted performance for the nonnative-accented speech conditions. Taken together, these results indicate that some cognitive resources may be recruited for all adverse listening conditions, while specific additional resources may be engaged when people are faced with certain types of listening challenges.

Serial Recall Predicts Vocoded Sentence Recognition Across Spectral Resolutions

“…Because our speech task had a single vocoded stream with no competing streams, the role of attention was likely minimized. The cognitive abilities that facilitate speech recognition depend on the speech task being performed (Heinrich et al, 2015;McLaughlin et al, 2018). For example, our lack of a correlation between attention and sentence recognition differs from speech in noise results from the study of Heinrich et al (2015).…”

“…Recognizing speech in noise requires segregating auditory streams from one another, attending the relevant stream, and identifying speech cues in the partially masked relevant stream (for a review, see Shinn-Cunningham & Best, 2008). Speech recognition performance is not correlated across speech-in-noise maskers and speech-in-speech maskers or between native and nonnative speech recognition (McLaughlin et al, 2018), which indicates that the skills required for listening in different adverse conditions depend on the adverse condition (see also Mattys et al, 2012, for a review). The extent to which these cognitive factors are relied upon also depends on the complexity of the speech task.…”

Task-evoked pupillary response to completely intelligible accented speech

“…In particular, working memory may be key for resolving perceptual ambiguities in accented speech. Evidence indicates that working memory capacity is positively correlated with intelligibility for accented speech perception in young and older adults (McLaughlin et al, 2018;Janse & Adank, 2012).…”

“…Under this executive recruitment account, even when an L2-accented speaker is completely intelligible (i.e., all of the words in their speech can all be correctly identified) speech processing should nonetheless require greater cognitive load (defined here as the degree to which cognitive resources are recruited at a given moment to meet processing demands; Pichora-Fuller et al, 2016). In particular, this increased cognitive load during speech processing may be attributable to the recruitment of executive functions including working memory (McLaughlin, Baese-Berk, Bent, Borrie, & Van Engen, 2018) and inhibition (Janse & Adank, 2012;Banks, Gowen, Munro, & Adank, 2015), which positively correlate with individual listeners' recognition of and adaptation to accented speech, respectively.…”

Sentence Context Facilitation for Children's and Adults' Recognition of Native- and Nonnative-Accented Speech

“…Future studies could attempt to equate performance in the native and nonnative conditions through the use of environmental degradation or by increasing the cognitive demands of the task. However, adding noise to speech so that scores are not at ceiling and a sentence context benefit can be observed may introduce other factors that will need to be considered: the perception of speech in noise or babble may involve the recruitment of different cognitive resources or processing strategies relative to processing of speech in quiet (McLaughlin, Baese-Berk, Bent, Borrie, & Van Engen, 2018).…”