Working memory supports rapid talker and accent accommodation: An individual differences investigation

Under multitalker listening conditions, listeners appear to rapidly accommodate variability in speaker productions. However, evidence indicates that this trial-to-trial accommodation incurs a processing cost and is amplified by accent differences among talkers. The present study investigated how individual listener differences in working memory capacity and attentional control may predict the degree of processing costs (as measured by response time and accuracy) incurred by trial-to-trial switches in English talkers of first language (L1) American accent and second language (L2) Mandarin Chinese accent. Listeners completed a dual-task experiment with a primary listening task and secondary non-linguistic task as well as measures of working memory capacity (the Word Auditory Recognition and Recall Measure) and attentional control (the Trail Making Test and a Stroop Test). Results indicated that, on trials with a talker switch, listeners with smaller working memory capacities were slower to respond to the secondary task and had poorer recognition accuracy for the primary task. The measures of attentional control did not predict switching costs. One marginal interaction emerged between Stroop test scores (typically used to assess selective attention and inhibitory control) and the accent manipulation, such that listeners with better attentional control had marginally better overall recognition accuracy for L2 accent. Working memory capacity, on the other hand, did not predict differences in recognition accuracy between accent conditions. We conclude that working memory plays a critical role in supporting rapid (trial-to-trial) talker accommodation. Attentional control may support processing of L2 accent more generally but not be critical for rapid accommodation.