Individual differences in plasticity in speech perception under cognitive load


Purpose: It has been suggested that speech perception is an inherently attention demanding process (Magnuson & Nusbaum, 2007) and limited attentional resources (e.g., cognitive load) have been shown to have disruptive effects on speech recognition (Mattys & Wiget, 2011). Although listeners may have reduced comprehension of speech under adverse conditions such as cognitive load, they may also use adaptive strategies to compensate for their increased cognitive load, achieving good understanding of speech. The goal of the present study is to better understand how and to what extent listeners’ speech perception abilities are modulated by cognitive load when listeners are engaged in a dual task and what makes some listeners better adapters to the demands of cognitive load. More specifically, the present study examines whether individuals differ in the extent to which they adjust their cue weighting strategies in the utilization of multiple acoustic cues under cognitive load. This study also investigates how individual differences in the adjustments of cue weighting strategies under cognitive load are related to individuals’ cognitive abilities (Janse & Adank, 2012; Tamati et al., 2013). Methods: Fifty-four native English listeners were engaged in a dual task in which they completed a speech categorization (two-alternative forced choice identification; 2AFC) task with a concurrent visual search task, after a baseline 2AFC task. Speech categorization stimuli were created from productions of *head* and *had* recorded by a native speaker of English. Five steps varying in formant frequency (created with TANDEM-STRAIGHT) were crossed with 5 duration steps ranging from 120 ms to 360 ms (PSOLA in Praat). Visual stimuli for the dual task were modeled after previous work (e.g., Bosker et al., 2017). The stimuli consisted of odd-one-out arrays (13 × 13) of colored shapes in which about half of the arrays (i.e., 13 out of 25 arrays) contain a black diamond (the target) adapted from Bosker et al. (2017). Participants also completed cognitive tasks examining working memory capacity (Backward Digit Span and Reading Span) and inhibitory control (Go/No-go and Stroop). Results: Results revealed that listeners overall showed increased cue weights under cognitive load, which may be interpreted as compensatory cue weighting strategies for adapting to phonetic categories under cognitive load. The increased cue weights under cognitive load may also be interpreted as an active cognitive process in speech perception in which listeners flexibly adjust their cue weighting strategies in response to the demands of a situation (Heald & Nusbaum, 2014). However, there were considerable differences across individuals in the extent to which these adaptive cue weighting strategies manifest. That is, some listeners showed an increased (or decreased) reliance on spectral quality whereas others showed an increased (or decreased) reliance on vowel duration under cognitive load. Crucially, these individual differences in changes in cue weights were associated with individual differences in cognitive abilities. Specifically, listeners with better inhibitory control showed more adaptive spectral changes while listeners with better working memory showed more adaptive duration changes. Conclusions: These findings suggest that individual listeners use different cue weighting strategies for utilizing multiple acoustic cues in speech perception to cope with cognitive load, and these differences are at least partially accounted for their cognitive abilities. Thus, the current study provides insights into the interplay between speech and cognitive processes, highlighting the potential role of cognitive abilities in speech perception.