Influence of Semantic Context on Speech-in-Noise Performance: Evaluating the Quick Speech-in-Noise Test

Purpose: Sentences are encoded with sematic context, which facilitates aulogic ability to navigate background noise, or speech-in-noise (SIN), conditiTo examine how semantic context contributes to performance on one monly used SIN test, the Quick Speech-in-Noise Test (QuickSIN) by EtymResearch, Inc. (henceforth “QuickSIN”), we use a novel experimental paradthat isolates semantic information. Method: Ten college-aged monolingual participants with typical hearing listeto 72 sentences delivered in 0, 5, 10, 15, 20, or 25 dB SNR followechoice between two visual words. One word was related to the overall senmeaning, and the other word was unrelated. The reaction time (RT) to coselect related targets was measured to index usage of semantic informationResults: Participant’s RTs to select a correct response were compared acdifferent signal-to-noise ratios (SNRs). We found that less favorable noise ctions (0, +5 dB SNR) elicited a greater usage of semantic information than favorable noise conditions (20, 25 dB SNR). Transformed RT data were lyzed with nonparametric tests that assessed the homogeneity of variawithin responses to each SNR condition. Results indicated that participaRTs were consistently varied within each SNR condition, except to sentenc+20 dB SNR, indicating an imbalance in the degree of semantic context uthe sentences in that SNR level. Conclusions: Respondents to the QuickSIN use semantic context to facilprocessing especially at less favorable SNR levels, which is consistent research supporting a greater role of semantic information during suboptlistening conditions. Differences in context use across noise conditions metest performance also reflects language processing and should be considfor updated tests of speech-in-noise performance. Critically, responses to tences at the +20 dB SNR used in the QuickSIN are so inconsistently vatheir degree of semantic usage as to prohibit a clinical interpretation alonthe other conditions. These findings warrant the development of a quicadminister SIN test with stimuli that are balanced for semantic expectancavoid language effects.