Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
The interaction between talker variability and selective attention along the auditory pathway
Poster D77 in Poster Session D, Wednesday, October 25, 4:45 - 6:30 pm CEST, Espace Vieux-Port
This poster is part of the Sandbox Series.
Sung-Joo Lim1; 1Binghamton University
Speech processing is less efficient when listeners process speech from variable mixed talkers compared to one consistent talker. Recent behavioral work provides evidence that talker variability is detrimental to speech processing because talker discontinuity evokes involuntary stimulus-driven attentional reorientation to a new auditory object (i.e., talker). However, little is known about the cognitive and neural mechanisms of automatic, stimulus-driven attentional reorientation in the context of speech communication. In this in-progress research, we are using talker variability as a tool to examine how stimulus-driven attentional disruption impact neural encoding of speech at different stages in the auditory processing pathway. Furthermore, we are examining whether the impact of talker variability depends on the goal-directed attention when recognizing speech in the presence of unattended competing speech. Using electroencephalography (EEG) and functional MRI, we are investigating whether the impact of talker discontinuity arises during the early encoding of speech in the auditory brainstem and/or in the higher-order domain-general vs. auditory-specific cortical regions. In a pilot EEG experiment, we recorded n=7 participants’ auditory brainstem responses during passive listening of a target speech syllable (English /da/), randomly intermixed with two non-target syllables (/di/ and /du/). We contrasted the frequency-following responses (FFRs) to the target syllable in contexts where listeners heard either a single talker vs. four mixed talkers, and tested whether this differed for speech in quiet vs. unintelligible babble noise. We hypothesize that talker variability would disrupt neural encoding of speech, and the effect of talker variability would be greater in the presence of background noise. Our preliminary results indicate a trend towards higher neural tracking of the target syllable’s fundamental frequency (F0) and second harmonic component (H2) in a single vs. mixed talker context; however, the FFR strengths across talker or noise contexts do not reach statistical significance. With fMRI, we are examining how talker variability affect cortical activity in auditory areas and the ventral attention network in the brain. We are using sparse-sampling fMRI while participants perform an auditory word-matching task, in which words are heard in single- vs. mixed-talker context, with or without competing speech in the background. By separately localizing the stimulus-driven ventral attentional network, we aim to determine whether listeners’ attentional focus is reoriented whenever they encounter talker switches in the mixed-talker context. We hypothesize that if mixed-talker speech leads to attentional disruption in a stimulus-driven manner, mixed-talker speech will lead to greater engagement of the bilateral auditory cortices and auditory-based ventral attentional networks. Alternatively, if mixed-talker speech only increases computational demands of configuring acoustic-phonetic mappings, we expect to observe the effect of talker variability only within auditory areas. From this project, we aim to delineate how stimulus-driven attention processes during speech processing operate at different levels of neural computation.
Topic Areas: Speech Perception,