English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
  Prosodic entrainment influences syntactic phrase generation

Lamekina, Y., & Meyer, L. (2020). Prosodic entrainment influences syntactic phrase generation. Poster presented at 26th Architectures and Mechanisms for Language Processing Conference (AMLaP), Virtual.

Item is

Files

show Files
hide Files
:
lamekina_amlap_080920.pdf (Any fulltext), 469KB
Name:
lamekina_amlap_080920.pdf
Description:
-
OA-Status:
Not specified
Visibility:
Public
MIME-Type / Checksum:
application/pdf / [MD5]
Technical Metadata:
Copyright Date:
-
Copyright Info:
-
License:
-

Locators

show

Creators

show
hide
 Creators:
Lamekina, Yulia1, Author           
Meyer, Lars1, Author                 
Affiliations:
1Max Planck Research Group Language Cycles, MPI for Human Cognitive and Brain Sciences, Max Planck Society, ou_3025666              

Content

show
hide
Free keywords: Prosody, Entrainment, Sentence segmentation
 Abstract: Speech processing is thought to be subserved by cyclic electrophysiological activity, so-called neural oscillations. Oscillations are thought to facilitate speech perception by synchronizing or entraining to rhythmic acoustic cues in speech (Luo & Poeppel, 2007; for review, see Meyer, 2018). In addition to speech acoustics, oscillations within the delta band (i.e., < 4 Hz) were also found to synchronize to abstract linguistic information—in particular, syntactic phrases (Ding et al., 2015). A recent study on attachment ambiguities found delta-band oscillations to subserve the internal segmentation of speech into syntactic phrases, independent of prosodic boundary cues (Meyer et al., 2017).
Our study aims at further supporting the role of cyclic electrophysiological activity in the generation of syntactic phrases. Specifically, we will test whether the generation of phrases can be entrained, which would be evidence that the generation of phrases is achieved through an electrophysiological oscillator. The experiment builds on prior evidence that oscillations inherit the rhythm of an acoustic stimulation to persist even after stimulation offset, influencing the perception of upcoming stimuli. For instance, Kösem et al. (2018) showed that oscillatory entrainment to a syllable rhythm can affect the comprehension of a subsequent vowel. Furthermore, priming studies have observed that the prosodic contour of a preceding stimulus can affect phrase generation during a subsequent sentence (Steinhauer & Friederici, 2001). We will will thus test whether oscillatory entrainment to speech prosody can trigger the perception of a prosodic boundary in an upcoming stimulus, leading to the internal generation of a syntactic phrase boundary.
Participants will listen to a prosodic contour, repeated three times in order to induce entrainment; contours are delexicalized, but include pitch modulations characteristic of a prosodic boundary. Two conditions—long and short duration of the contour, and thus a slow and a fast sequence of prosodic boundaries—will be presented. Both prosodic contours will be followed by rapid serial visual presentation of an identical sentence involving a coordination ambiguity such as Jim saw Tom and John laughed. (Hoeks, Vonk & Schriefers, 2002; see Figure 1). Critically, the long contour matches the duration of Jim saw Tom and John, aiming to drive participants up the garden-path; in contrast, the short contour matches the duration of Jim saw Tom, aiming to avoid the garden-path (see lower panel of Figure 1). Performance, and thus phrase generation, will be assessed by self-paced reading of the final words of the sentence, as well as by comprehension questions. Relying on online testing with the Prolific and Gorilla infrastructures, we are currently acquiring behavioural data from 40 participants. Once the lockdown of our testing facilities ends, we will record magnetoencephalography (MEG) data from 40 participants in order to assess oscillatory entrainment.
We predict participants to exhibit more errors and longer reaction times on comprehension questions in the long condition, as well as longer self-paced reading times. Both would reflect a garden-path effect. For the MEG data, we predict a difference between the long and short conditions in the phase of delta-band oscillation at the offset of Tom (see Figure 1): Under long entrainment, the delta-band cycle would only reach its end at the offset of John, whereas under short entrainment, the cycle would end at the offset of Tom, leading to a phase difference at Tom. In further support of oscillatory phrase generation, we expect to find peaks at the respective long and short entrainment frequencies in the MEG power spectra to the visual sentence. Finally, we expect an increased P600 in the event-related brain potential to the sentence-final verb (i.e., laughed) in the long compared to the short condition, consistent with syntactic re-analysis (Osterhout & Holcomb, 1992). P600 amplitude should correlate with delta-band phase at the critical point in the sentence: When the end of an oscillatory cycle induces the correct syntactic boundary at the offset of Tom, there should be no P600 at laughed; in contrast, when the cycle is continuing into the garden-path, P600 amplitude should increase. Taken together, this would be evidence that the internal generation of syntactic phrases is subserved by electrophysiological cycles at delta-band frequency.

Details

show
hide
Language(s):
 Dates: 2020-09
 Publication Status: Not specified
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: -
 Degree: -

Event

show
hide
Title: 26th Architectures and Mechanisms for Language Processing Conference (AMLaP)
Place of Event: Virtual
Start-/End Date: 2020-09-03 - 2020-09-05

Legal Case

show

Project information

show

Source

show