Linguistics Department

Colloquium - Fernando Llanos (University of Pittsburgh)

The effects of language experience and adverse listening in the neural processing of speech

Mon, February 10, 2020 | RLP 1.302E

3:00 PM - 5:00 PM

Listening to speech gives rise to a cascade of computations that result in speech comprehension. These computations involve the extraction of segmental and suprasegmental patterns from a continuous and acoustically complex signal and the subsequent mapping of these patterns into lexical representations that convey meaning. While speech processing does not demand much effort when speech is native, it can be very demanding when speech is not native or under adverse listening (e.g., in the presence of a competing auditory signal). In this talk, I will present a series of electrophysiological (EEG) studies that examined the neural and perceptual processing of segmental and suprasegmental patterns across multiple conditions, including non-native speech and adverse listening. First, I will discuss the effects of different native and non-native experiences in the processing of suprasegmental patterns. Then, I will focus on the processing of segmental features while listening to continuous speech under adverse listening. Throughout the talk, I will also discuss how machine learning can be used to overcome experimental limitations regarding the analysis of subcortical EEG signals and the interpretation of cortical EEG responses to continuous speech. Using these computational techniques, I will show that the bottom-up cascade of computations underlying speech sound processing is top-down modulated by cognitive and experience-dependent processes that are not predictable from the speech signal. I will conclude by discussing the implications of these processes for current models of speech processing and second language acquisition.

Bookmark and Share

  • Department of Linguistics

    University of Texas at Austin
    305 E. 23rd Street STOP B5100
    Robert L. Patton Hall (RLP) 4.304
    Austin, TX 78712