An Analysis of Reasons for Speech-in-Noise Deficits Based on Clinical Use of LiSN-S

Authors: Harvey Dillon1,2

1Macquarie University
2University of Manchester

Background: The Listening in Spatialized Noise Sentences (LiSN-S) test is used in clinics across English-speaking countries to assess understanding of sentences in noise, primarily to assess people with listening difficulties despite normal hearing thresholds. LiSN-S comprises four subtests, differing in whether the target and competing speech come from the same or different directions, and whether they are spoken by the same or different talkers. This structure allows some reasons for deficits to be inferred.

Methods: The age-corrected results of nearly 3000 clinical applications of LiSN-S from multiple countries were analysed.

Results: In almost half the cases, speech-in-noise performance in the High-cue condition (different voices, different locations) was more than one population standard deviation (SD) below the mean for their age and for more than half of those, the deficit exceeded two SDs. Of the latter, ability to use spatial cues to segregate the target from competition was also more than two SDs below normal in 43% of cases. The distribution of this spatial ability was highly asymmetrical (see figure). It could be decomposed into a sum of two Gaussian distributions with 46% of cases falling into a distribution (mean = 0.3, SD = 0.93) very close to that expected for people with normal ability (i.e. mean = 0; SD = 1). The remaining 56% fell into a (displaced) normal distribution with mean =-2.2 and SD = 2.23. The High-Cue condition could similarly be decomposed into “normal” and “abnormal” distributions, with about half the cases falling into each.

Conclusions and implications: Spatial Processing Disorder, which is fully remediable in people with normal hearing thresholds, is the cause of almost half of the speech-in-noise deficits identified with LiSN-S. Currently Spatial Processing Disorder, can be identified only with LiSN-S, or its language-independent alternative, Listening in Spatialized Noise – Universal (LiSN-U).