You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

Poster D65, Thursday, November 9, 6:15 – 7:30 pm, Harborview and Loch Raven Ballrooms

Processing of English focal stress by L1-English and L1-Mandarin/L2-English speakers: An auditory ERP study

Ellen Guigelaar1,2, John Drury1;1Stony Brook University, 2East Tennessee State University

[INTRODUCTION] The present auditory ERP study addresses two issues facing the literature on the neural basis of prosody: (i) previous work on question-induced focus and prosodic marking of accent in question/answer pairs has yielded inconsistent findings, and (ii) the nature of the neural basis of non-native (L2) prosodic processing has been under-investigated (though see Nickels & Steinhauer 2016). Concerning (i): Part of the source of the inconsistency of previous work is methodological, as argued in Dimitrova et al. (2012). To that list of concerns we add an additional worry: several previous studies appear likely to have serious baseline confounds, in particular concerning ERP effect estimates for the second of two target time-locking points in auditorily presented sentences. Here we take steps to avoid these problems, and also begin closing the research gap in the ERP literature concerning non-native/L2 prosodic processing. [PRESENT STUDY/METHODS] Target sentences involved prosodic accent marked on either the first or second object of distransitives (e.g., “Steve only gave [BORIS] [the bulldog] last week” / “Steve only gave [Boris] [the BULLDOG] last week”). In addition to these double object constructions, dative constructions were also employed, so that the direct/indirect object NPs occurred equally often in both the first and second positions in the verb phrase (e.g., “Steve only gave [the BULLDOG] to [Boris] last week” / “Steve only gave [the bulldog] to [BORIS] last week). Collapsing over the construction types, here we examine the contrast between prosodic Accent on first object (OBJ1) versus the second (OBJ2). Accent was crossed in a 2x2 design with expected semantic Focus on OBJ1 versus OBJ2, determined by lead-in questions rendering the answer felicitous (e.g., Q: “Who did Steve give the bulldog to last week?” A: “Steve only gave [BORIS] [the bulldog] last week”), or not (e.g., Q: “What did Steve give to Boris last week?” A: “Steve only gave [#BORIS] [the bulldog] last week”). Native English speaker participants (N=16) and native Mandarin (English-L2) participants (N=27) listened to these question/answer pairs and performed match/mismatch judgments. Potential baseline confounds which we argue have affected previous studies were avoided in analyses of OBJ2 effects by time-locking to the onset of OBJ2 and using a distant (-900 to -700 ms) baseline. [RESULTS & DISCUSSION] Both native and L2 groups demonstrated robust main effects of Focus (N400 effects for new>old/repeated information) and Accent (larger P200 for Accented>Unaccented) for both OBJ1/OBJ2 positions. Both groups showed an N400-like effect for missing accent on focused OBJ1, with no comparable effect on OBJ2. However, only native speakers demonstrated Focus x Accent interactions tied to an early positivity preceding the Accent P200 effect where the marking of Accent was superfluous, and this pattern manifested for both OBJ1 and OBJ2. However, closer investigation of the L2 group demonstrates that native-like Focus x Accent early interaction for superfluous accent did manifest in earlier, but not in later learners. We relate these patterns to earlier behavioral findings by Akker & Cutler (2003) where similar presence/absence of Focus x Accent interactions were reported for L1/L2 processing.

Topic Area: Perception: Auditory

Back to Poster Schedule