(PP1318) Speech Recognition in Noise Performances Versus Perceived Listening Difficulty

Objectives: The purpose of this study was to investigate the relationship between speech recognition in noise (SRN) performances and perceived listening difficulty. The AzBio test was used to determine SRN ability and perceived listening difficulty was rated on a five-point scale where 1 equals very poor, and 5 equals excellent SRN ability. 

Rationale: There is no consensus on the rating of speech recognition performance. According to Schoepflin (2012) word recognition score (WRS) categories are: excellent (90-100%), good (78- 88%), fair (66-76%), poor (54-64%), and very poor (< 52%). However, the Veterans Administration (2009) had slightly different WRS categories: good (80-100%), fair (60-78%), poor-fair (46-58%), and poor (0-44%) word recognition ability. It is not clear how these categories relate to the perceived listening difficulty of the listener. It has been reported in the literature that individuals with normal pure-tone thresholds will present with a wide range of word recognition abilities. Holder et al, (2018) reported a 60-percentage point range and Vermiglio et al, (2021) reported a range of 55-percentage points for speech in noise performances in individuals with normal pure-tone thresholds. The purpose of this study was to investigate the relationship between the perceived listening ability using the categories from Schoepflin (2012) on AzBio test performances in 10-talker babble and steady-state noise. 

Design: Thirty-one native English speakers with normal pure-tone thresholds (≤ 25 dB HL, 0.25 – 4.0 kHz) participated in this study (mean=20.9 years). All but one participant was female. The AzBio test was used to evaluate binaural speech perception in noise. The sentences and maskers were presented at 65 dBA (0 dB SNR). The maskers included 10-talker babble and steady-state noise. The test stimuli were delivered using supra-aural headphones in a sound-treated booth. All test conditions were randomized. Following each sentence list, participants were asked to rate their speech recognition ability based on the 5 categories from Schoepflin (2012).

Results: Mean performance for 10-talker babble and steady-state noise conditions were 48.70% and 77.96% respectively. The range in performances for the 10-talker babble masker was 61.05 percentage points and 89.53 percentage points for the steady-state noise condition. Better performances were found in steady-state noise than in 10-talker babble for 93.5% of subjects. (steady-state noise advantage) and 6.5 % of subjects performed better in 10-talker babble. The Spearman Rho correlation was used to quantify the strength of the relationship between AzBio performances and perceived listening ability (coded numerically 1-5). For the steady-state noise condition; a statistically significant relationship was found between AzBio scores vs. perceived listening ability (r = 0.58, p = 0.003). The percentage of subjects in each rating category are as follows; very poor, 3.23%; poor, 12.9%; fair, 22.58%; and good, 61.29%. 

Conclusions: A statistically significant relationship was found between the rating of listening ability vs. AzBio performance in steady-state noise. Participants with better AzBio scores, gave higher ratings of listening ability.