(PP1321) The Effect of Listener Group and Masker Condition on Auditory Working Memory

Title: The effect of listener group and masker condition on auditory working memory


Objectives: The goal of this study was to determine the effect of listener group and masker condition on auditory working memory. Auditory working memory was evaluated using the Digit Span Forward Recall test in four listening conditions: four-talker babble forward (4TF), four-talker babble backwards (4TB), conversational masker forward (CF), and conversational masker backwards (CB). We hypothesized that participants with no self-reported speech-in-noise difficulties will perform better on an auditory working memory task in the presence of noise. We also hypothesized that the semantic content of the masker would have a negative effect on auditory working memory.

Rationale: Roup et al (2021) reported that listeners with greater self-perceived hearing difficulty performed poorer on the Listening in Spatialized Noise-Sentences Test (LiSN-S). Anoop and Kumar (2019) found that linguistic information in maskers correlated with greatest deterioration of working memory performance. Carhart et al (1969) used the term “semantic interference” to refer to this effect. Building on these previous studies, the current investigation sought to determine the effect of listener group and masker conditions on auditory working memory studies.

Design: Twenty-two native English speakers with normal pure-tone thresholds (≤ 25 dB HL, 0.25 – 4.0 kHz) participated in this study. Participants were asked “How would you rate your overall level of difficulty with speech perception in noisy environments?” in a pre-testing questionnaire. Participants who reported no difficulty were assigned to the control group while participants who responded with slight or moderate difficulty were assigned to the King-Kopetzky syndrome (KKS) group (Hinchcliffe, 1992). The Digit Span Test was used in forward recall to test auditory working memory (Wechsler, 1955). It was administered in 4TF, 4TB, CF, and CB conditions. The digits and maskers were presented at 65 dBA (0 dB SNR) and test stimuli were presented using supra-aural headphones in a sound-treated booth. All listening conditions and Digit Span lists were randomized.

Results: A repeated measures ANOVA was conducted to investigate the effects of participant group (control, KKS) and listening condition on Digit Span scores. Although the control group performed better on the Digit Span test in all listening conditions except in CB, this analysis revealed that the main effect was not statistically significant (F(1,21)=0.0145 for participant group, p<0.5687). However, a statistically significant main effect was found for listening condition on Digit Span scores (F(1,21)=1.920 for listening condition, p<0.0001). A post hoc analysis using one sample t-tests revealed statistically significant differences between all condition comparisons except conversational masker conditions. Evidence of semantic interference on working memory scores was demonstrated; performance for 4TB was statistically better than the 4TF condition (mean difference = 1.17, p=0.0042).

Conclusions: The results of the present investigation revealed no effect for participant group (control, KKS) on Digit Span scores. A significant main effect for listening condition was found. Evidence of semantic interference was demonstrated for the four-talker babble listening conditions. This suggests that working memory ability may be compromised in background noise containing semantic information.

Kathryn Fennie, East Carolina University, fenniek19@students.ecu.edu