Dissertations, Theses, and Capstone Projects

Date of Degree

2-2020

Document Type

Dissertation

Degree Name

Ph.D.

Program

Speech-Language-Hearing Sciences

Advisor

Loraine K. Obler

Committee Members

Klara Marton

Brett Martin

Teresa S. Pisano

Subject Categories

Psycholinguistics and Neurolinguistics

Keywords

auditory sentence comprehension, L2 listening, bilingual processing, listening in noise, L2 proficiency, interference control

Abstract

Speech perception and comprehension in the presence of interfering auditory stimuli is a challenge for bilingual listeners (e.g., Ezzatian, Avivi-Reich, & Schneider, 2010; Krizman, Bradlow, Lam, & Kraus, 2017). How efficiently and skillfully listeners manage auditory interference may also be closely related to their ability to pay attention to a target and suppress irrelevant information. Based on Friedman and Miyake’s (2004) framework of interference control, this dissertation investigated the underlying mechanisms of late Korean-English bilingual individuals’ auditory interference control in the presence of auditory verbal and nonverbal masking and evaluated the potential interaction between L2 proficiency and interference control.

Two groups of late bilingual listeners with high and mid L2 proficiency participated in three experiments. Experiment 1 investigated the interplay between interference control and L2 proficiency in bilingual listeners. Seventy Korean-English bilingual participants with high- and mid-L2 proficiency levels were recruited and tested with an L2 auditory sentence comprehension test. In this task, participants listened to English target sentences with and without masking (i.e., an auditory distractor). For each sentence, they judged semantic plausibility. Three masking conditions were presented: nonverbal speech-modulated noise, L1 verbal, and L2 verbal masking, to see the effect of different types of auditory interference during L2 listening. The results of the plausibility task indicated that the effect of the verbal masking was dependent on the listeners’ L2 proficiency. The effects of the L1 and L2 masking did not differ significantly for the high-proficiency L2 listeners when they listened in L2. However, the L1 masking had a significantly greater interference effect than the L2 masking on L2 listening among the mid-proficiency listeners. This suggests an interaction between L2 proficiency and interference control.

Experiment 2 examined to see whether there is interference effect beyond L2 proficiency effect found in Experiment 1. The participants from Experiment 1 engaged in an auditory sentence comprehension task, called word selection task, whereby they listened to English target sentences presented with L1 or L2 verbal masking. The procedure and the type of stimuli of this task were exactly the same as those in Experiment 1. In particular, the participants were still asked to pay attention to the target sentence based on a given picture cue. However, instead of making plausibility judgments, they were asked to select all the words that they had heard. . From the list, they were asked to select all the words that they had heard. The word list included two words extracted from the target sentence, two words from the masking sentence, and four words that were not presented but had a semantic or phonological association with the other four words from the two presented sentences. The results showed an interaction between proficiency and interference effect. The high-proficiency group identified a similar number of target words in both L1 and L2 masking conditions whereas the mid-proficiency group identified more target words in the L2 masking condition, suggesting a proficiency effect. On the other hand, both groups identified more content words from the non-target sentence in the L1 masking sentence than from the L2 masking sentence, suggesting a greater interference effect from L1 masking than L2 masking. Interestingly, both groups identified a similar number of non-target words in the L1 masking condition. Nonetheless, the high-proficiency group identified more content words from the target sentence than the mid-proficiency group. These findings suggest that the high-proficiency L2 listeners have better attentional control on the target stimuli than the mid-proficiency group. The results that L2 listeners identified content words in both the target and masking sentences, particularly in L1, suggest that the major difference between the groups lies in their ability to divide their attention and orient it to the target signal.

Experiment 3 employed a nonverbal auditory interference control task to investigate whether the group difference in the verbal task reflects primarily language-specific or domain-general cognitive-control systems. The same participant groups listened to and simultaneously counted two types of target animal sounds masked by various other animal sounds. The results showed that the high and mid-proficiency groups performed similarly on this nonverbal task, unlike the verbal task in Experiment 2. This suggests that the suppression of nonverbal interference involves a domain-general interference control system whereas verbal interference may require an additional domain-specific control ability above and beyond domain-general cognitive control.

This study provides novel evidence that the effects of auditory interference in a bilingual’s two languages on L2 listening differ according to the listeners’ L2 proficiency. Second language listening with accompanying auditory interference requires interference control. This control ability is subserved – at least in part - by a different control system from the one for nonverbal materials. Sentence comprehension in L2 was more adversely affected by L1 interference than by L2 interference, particularly in bilingual individuals with mid-proficiency in L2. However, L2 listeners with high L2 proficiency exhibited a better control ability in suppressing the L1 interference than the mid-proficiency listeners. These findings highlight the importance of considering both language proficiency and interference control abilities combined regarding L2 listening comprehension for individuals who listen in an L2.

Download

Share

COinS