AAAL 2023: Invited Colloquium

Broadening the Base and Charting New Courses in Second Language Speech Comprehensibility Research


Dustin Crowther, University of Hawaiʻi at Mānoa 
Daniel Isbell, University of Hawaiʻi at Mānoa


Pavel Trofimovich, Concordia University

Colloquium Abstract

In an increasingly globalized world where second language (L2) users find themselves in contact with fellow users from a range of linguistic and cultural backgrounds, it is unsurprising that L2 pronunciation research has continually advocated for a shift in scholarly and pedagogical emphasis from nativelikeness to intelligibility. In line with this shift, the global speech dimension of comprehensibility, most frequently defined as listeners’ perceived ease/difficulty of understanding L2 speech, has received increased scholarly focus over the past 25+ years. Comprehensibility has been linked to the processing effort required by a listener, which in turn can impact listeners’ social judgments of a given speaker. The implications of this line of inquiry have been significant in reference to language pedagogy, frameworks for describing language proficiency, and speaking assessment. Though such research has been productive and influential, in some ways the territory it covers has become well-tread. Recent advancements have aimed to push beyond a frequent focus on adult L2 English speakers, adult native listeners’ judgments, linguistic influences on listeners’ judgments, and evaluations of L2 speaking instruction effectiveness. Though these foci are worthy and associated methods are sound, more diverse approaches have begun to emerge with aims to learn more about comprehensibility in non-English languages, among different types of speakers and listeners, and its role in diverse social contexts.

This two-hour colloquium brings together researchers who are broadening the knowledge base of L2 comprehensibility and pushing it into new methodological and interdisciplinary directions. The session will begin with a 10-minute introduction followed by four 15-minute presentations that enrich our understanding of L2 comprehensibility (including discussions of theoretical, applied, and methodological aspects). The colloquium will conclude with thoughts and insight from discussant Dr. Pavel Trofimovich. Each presentation will include a brief, presenter-specific Q&A, followed by a 20-minute general discussion at the conclusion of the colloquium.

Task engagement and comprehensibility development: A longitudinal study of video-mediated intercultural exchange
Yuka Akiyama, University of Tokyo,
Sachiko Nakamura, Tamagawa University,
Takumi Uchihara, Waseda University,
Kazuya Saito, University College London, 

One of the most extensively researched topics in SLA is the role of conversational interaction for language development. In a series of precursor studies that examined how video-mediated intercultural communication with native speakers over a semester impacted L2 learners’ speech development (e.g., Akiyama & Saito, 2016; Saito & Akiyama, 2017), L2 learners as a whole demonstrated improvement in comprehensibility (listeners’ perceived ease of understanding), but the quality of interaction and its effect on L2 learning greatly varied across individuals. As such, we may examine more closely how dynamic, contextual, and highly individual L2 interactions affect comprehensibility development by, for example, looking at how intercultural dyads engaged in conversations. 

Engagement is a multidimensional construct that has been shown to play a key role for L2 learning (Dörnyei, 2001; Nakamura et al., 2021; Philp & Duchesne, 2016; Trofimovich et al. 2021). Although the construct seems suitable for revealing the dynamicity and individuality of L2 interactional processes over time, there is a lack of studies that examine this construct longitudinally in relation to L2 speech development (Hiver et al., 2020), let alone in a technology-mediated foreign language learning context (Gijsen, 2021). 

Thus, this study examined the relationship between task engagement and comprehensibility development in a semester-long virtual exchange project. The participants were L2 English learners in Japan (N = 10) and L2 Japanese learners in the U.K. (N = 10) who formed dyads to learn each other’s languages and cultures via Zoom. Their L2 speech samples were collected using a picture description task before and after the project and were evaluated by two trained bilingual raters for comprehensibility. The 10 dyads’ video interaction data were analyzed focusing on four dimensions of engagement (behavioral, cognitive, social, emotional) over time. The results found a complex relationship between various measures of engagement and comprehensibility development.

The Effects of Social Attitudes and Contexts on Listeners’ Judgements of L2 Comprehensibility

Okim Kang, Northern Arizona University,

It is well known that attitudes affect our judgments and perceptions of others (Kang & Rubin, 2009; Subtirelu & Lindemann, 2016). Much research has demonstrated that people make biased judgements based on speakers’ language choice and accent (Kang & Yaw, 2021; Lambert et al., 1960). However, the influence of social attitudes and situational contexts on listeners’ comprehensibility of second language (L2) speech is largely unknown. Therefore, the current presentation briefly reports two separate studies: (1) comprehensibility judgements and L2 accented speech stereotyping, and (2) roles of different situational contexts on L2 comprehensibility. Based on the concept of reverse linguistic stereotyping (RLS) (Rubin, 1992), the first study operationalized listener aberration as proclivity toward RLS along with a dimension of speaker social attractiveness, superiority, and dynamism. It examined to what extent listeners’ RLS propensity affected their comprehensibility judgements of L2 accented speech. Seventy-six listeners listened to 11 international teaching assistants’ speech files. Results showed that listeners who held negatively stereotyped expectations about L2 accents found accented speech less comprehensible. Using a matched-guise technique, the second study examined how subject-related academic contexts and workforce-related professional contexts affected listeners’ comprehensibility judgements of accented speech. Sixty-four undergraduate students listened to 24 sound files (18 target stimuli plus six distractors) and each stimulus appeared with a written description introducing the speaker’s profession and the context of the speech. Results revealed that simulated contexts made a significant difference in listeners’ comprehensibility judgements, with speakers perceived as significantly more comprehensible and acceptable in service-occupational roles than in academic contexts. Findings suggest that listeners’ stereotyping factors affect their judgments on speakers’ oral performances and their judgements are heavily influenced by speakers’ situational contexts. The studies offer implications to language education and workforce-related communication in global contexts.

Comprehensibility and Client Interactions

Mary Grantham O’Brien, University of Calgary,

Although Canada protects individuals from discrimination based on race, national or ethnic origin, color, and language, linguistic bias is a reality in work settings. This bias, in turn, can have a negative impact on individuals’ ability to obtain gainful employment. The goal of this research is to investigate the extent to which listener perceptions of second language (L2) speakers are differentially affected by speakers’ first language (English or Tagalog), the relative prestige of a given job, and the relative skill level demonstrated on the job. Listeners heard recordings of speakers portraying workers responding to their clients’ needs on the job. The recordings depicted situations in which speakers in higher (doctor, lawyer, professor) or lower (cleaner, server, retail worker) prestige jobs performed their jobs at higher or lower skill levels. Listeners rated speakers along three dimensions: one linguistic (comprehensibility), one professional (competence), and one behavioral (a listener’s preference for treatment). A total of 192 English-speaking listeners living in Calgary completed the study. Half of the listeners completed the task in the absence of priming, and the other half completed the task after hearing the story of an immigrant’s struggles finding work. Listeners rated depictions of the high-skill jobs more favorably. Tagalog speakers in low-prestige, as opposed to high-prestige, jobs were rated as more comprehensible, and listeners indicated that they would like to be treated more like these same speakers. Listeners who were exposed to priming indicated that they would like to be treated more like English speakers, whereas those not exposed to the priming indicated that they want to be treated more like Tagalog speakers. Overall, the results support findings that listeners may have lower expectations for L2 speakers with lower-prestige jobs and that the effects of listener priming, designed to mitigate listener perceptions, may not function as expected.

You’re comprehensible, so what? Examining comprehensibility as a predictor of communicative outcomes

Charlie Nagle, Iowa State University,

Over the past ten years, comprehensibility has risen to the forefront of second language speech research. One of the most productive lines of work has examined linguistic predictors of comprehensibility in several target languages and on a variety of speaking tasks. Recent work, however, has begun to unpack comprehensibility to understand how it fluctuates in real time as listeners and speakers interact with and adapt to one another. Studies in this area have shown that comprehensibility is dynamic, co-constructed, and related to a range of behavioral and affective correlates. In all this work, comprehensibility has been viewed as the outcome measure, to which other variables are mapped, which begs the question: What does comprehensibility predict? Or, to put it more bluntly, you’re comprehensible, so what? 

In this presentation, I discuss the past, present, and future of comprehensibility research. I set the stage by briefly tracing the genesis of the current wave of comprehensibility studies. I then turn to work on the internal dynamics of comprehensibility, including interactive approaches in which comprehensibility measures must be coordinated between both interlocutors. I end by discussing emerging research using comprehensibility to predict other interlocutor-based outcome measures that may regulate willingness to communicate again, such as the interlocutors’ impression of their interactive experience, communicative success, and comfort interacting with their partner. Ultimately, I make the case that we do not need to close the book on comprehensibility, but we do need to turn the page, considering new research questions and methodologies that situate comprehensibility within a more comprehensive communicative framework.

>>>Back to AAAL 2023 Invited Colloquia