The role of syntax in maintaining the integrity of streams of speech

Abstract

This study examined the ability of listeners to utilize syntactic structure to extract a target stream of speech from among competing sounds. Target talkers were identified by voice or location, which was held constant throughout a test utterance, and paired with correct or incorrect (random word order) target sentence syntax. Both voice and location provided reliable cues for identifying target speech even when other features varied unpredictably. The target sentences were masked either by predominantly energetic maskers (noise bursts) or by predominantly informational maskers (similar speech in random word order). When the maskers were noise bursts, target sentence syntax had relatively minor effects on identification performance. However, when the maskers were other talkers, correct target sentence syntax resulted in significantly better speech identification performance than incorrect syntax. Furthermore, conformance to correct syntax alone was sufficient to accurately identify the target speech. The results were interpreted as supporting the idea that the predictability of the elements comprising streams of speech, as manifested by syntactic structure, is an important factor in binding words together into coherent streams. Furthermore, these findings suggest that predictability is particularly important for maintaining the coherence of an auditory stream over time under conditions high in informational masking.

The Effects of Uncertainty in Level on Speech-on-Speech Masking

“…Although performance was still above chance for the RWL condition, in general, (from Figures 3 and 4), it was substantially poorer than for either of the stable speech level conditions, suggesting that listeners benefitted significantly from a known and constant level even when the consistency was only within a trial (e.g., the similarity of results between FL and RSL conditions). This highlights the importance of speech streams conforming to expectations over time for target "stream maintenance" under competing speech (see Kidd et al, 2014, for analogous evidence of the benefit of conforming to a known syntax).…”

“…In typical multiple-talker communication situations, the listener has many different cues that may assist with solving the CPP by enhancing the perceptual segregation of sound sources and aiding in the focus of attention on the target while ignoring competing masker speech or other unwanted sources of environmental "noise." These cues include both auditory and visual information (e.g., lip reading, Woodhouse et al, 2009), as well as higher-level linguistic factors such as syntactic and semantic context (e.g., Brouwer et al, 2012; Kidd et al, 2014; review in Mattys et al, 2012). Spatial separation of sound sources may also afford a substantial improvement in speech recognition performance in multiple-source sound fields compared to conditions where competing sources appear to emanate from the same spatial location (i.e., "colocated" with the target source; e.g., Best et al, 2013a, 2013b; Hawley et al, 2004; Marrone et al, 2008).…”

Performance on stochastic figure-ground perception varies with individual differences in speech-in-noise recognition and working memory capacity

“…More importantly, whereas language processing is absent for SFG tasks, it is vital for speech-in-noise recognition. For speech-in-noise tasks that involve recalling or identifying spoken words embedded in sentences, in addition to low-level acoustic perception processes, listeners engage in higher-level comprehension processes that include lexical access (Carroll et al, 2016) and to some extent, syntactic knowledge (Kidd et al, 2014). For instance, Coulter et al (2021) found that higher contextual constraint had a larger positive effect on speech recognition in noisy compared to quiet conditions.…”

Auditory and language contributions to neural encoding of speech features in noisy environments

“…High-context sentences can be better understood in noisy environments than low-context sentences or random words (Miller et al, 1951). Furthermore, prior knowledge about the content of a sentence can help the listeners to understand it in noise (Helfer and Freyman, 2005;Jones and Freyman, 2012;Kidd Jr et al, 2014;Yang et al, 2007). In the present study, we further demonstrate that language knowledge can even facilitate a low-level auditory task, i.e., detecting repeated sound patterns, at an intermediate SNR levels, i.e., -6 dB and -9 dB (Fig.…”