Opportunities associated with emerging technologies and oral language assessment

Gary J. Ockey & Fatimah M. Aseeri
Iowa State University | Najran University

In recent years, there has been a rapid increase in the use of technology to assist in the delivery of second language oral communication (OC) assessments. In particular, video mediation, virtual reality, and spoken dialogue systems (SDSs) have been employed for this purpose. These emerging technologies have helped in assessing aspects of the OC construct that have been previously neglected, such as interactional competence. Thus, the purpose of this paper is to showcase how these technologies have been incorporated into OC assessments, what aspects of the construct they can help to assess, areas where they are inadequate, and what new challenges they introduce. The paper concludes with directions for future research.

Publication history

Date received: 17 October 2025

Date accepted: 21 August 2026

Published online: 20 October 2026

Table of contents

Opportunities associated with emerging technologies and oral language assessment

Oral communication (OC) refers to the use of spoken language to engage, share, and exchange knowledge for direct communication with others (Jamieson et al., 2008Jamieson, J., Eignor, D. R., Grabe, W. and Kunnan, A. J. (2008) “Frameworks for a new TOEFL”. In C. A. Chapelle, J. Jamieson, and M. K. Enright (Eds.), Building a validity argument for the test of English as a foreign language (pp. 55–95). Routledge.Google Scholar logo with link to Google ScholarJamieson, J., Eignor, D. R., Grabe, W. and Kunnan, A. J. (2008) “Frameworks for a new TOEFL”. In C. A. Chapelle, J. Jamieson, and M. K. Enright (Eds.), Building a validity argument for the test of English as a foreign language (pp. 55–95). Routledge.Google Scholar logo with link to Google Scholar). This definition implies that oral communication requires a real-time interactive conversation between two or more speakers. Thus, many language testers have encouraged test developers to include paired or group discussion tasks in OC assessments (Brooks, 2009Brooks, L. (2009) Interacting in pairs in a test of oral proficiency: Co-constructing a better performance. Language Testing, 26 , 342–366. Google Scholar logo with link to Google ScholarBrooks, L. (2009) Interacting in pairs in a test of oral proficiency: Co-constructing a better performance. Language Testing, 26, 342–366. Google Scholar logo with link to Google Scholar; Ducass & Brown, 2009Ducasse, A. M., & Brown, A. (2009) Assessing paired orals: Raters’ orientation to interaction. Language Testing, 26 , 423–443. Google Scholar logo with link to Google ScholarDucasse, A. M., & Brown, A. (2009) Assessing paired orals: Raters’ orientation to interaction. Language Testing, 26, 423–443. Google Scholar logo with link to Google Scholar; Ockey, 2014 (2014) The potential of the L2 group oral to elicit discourse with a mutual contingency pattern and afford equal speaking rights in an ESP context. English for Specific Purposes, 35 , 17–29. Google Scholar logo with link to Google Scholar (2014) The potential of the L2 group oral to elicit discourse with a mutual contingency pattern and afford equal speaking rights in an ESP context. English for Specific Purposes, 35, 17–29. Google Scholar logo with link to Google Scholar), in order to enhance these assessments’ authenticity. Such tasks, which are interactive (requiring immediate two-way communication between two or more speakers), and integrated (assessing more than one skill at a time: listening and speaking), replicate how language is used in most real-life contexts.

In an effort to ensure that OC assessments are both integrated and interactive, researchers and test developers have considered various approaches to guide them, including treating English as a lingua franca (ELF) and considering language through the lens of interactional competence features. Based on ELF principles, individuals must be able to communicate with speakers of different speech varieties, and nativism should not be used as a standard of second language oral ability (Levis, 2020Levis, J. (2020) Revisiting the intelligibility and nativeness principles. Journal of Second Language Pronunciation, 6 (3), 310–328. Google Scholar logo with link to Google ScholarLevis, J. (2020) Revisiting the intelligibility and nativeness principles. Journal of Second Language Pronunciation, 6(3), 310–328. Google Scholar logo with link to Google Scholar). This argument arises from the fact that in real-life contexts people often do not communicate with (only) native English speakers, as the number of nonnative speakers has exceeded that of native speakers in many parts of the world. Thus, test takers need to be able to communicate with speakers from diverse English use contexts (Jenkins, 2016Jenkins, J. (2016) International tests of English: Are they fit for purpose? In H. Liao (Ed.), Critical reflections on foreign language education: Globalization and local interventions. The Language Training and Testing Center.Google Scholar logo with link to Google ScholarJenkins, J. (2016) International tests of English: Are they fit for purpose? In H. Liao (Ed.), Critical reflections on foreign language education: Globalization and local interventions. The Language Training and Testing Center.Google Scholar logo with link to Google Scholar; McNamara, 2018McNamara, T. (2018) A challenge for language testing: The assessment of English as a Lingua Franca. Language Assessment Research Conference, Ames, Iowa, March 21–23.McNamara, T. (2018) A challenge for language testing: The assessment of English as a Lingua Franca. Language Assessment Research Conference, Ames, Iowa, March 21–23.; Ockey & Hirch, 2020Ockey, G. J., & Hirch, R. R. (2020) A step toward the assessment of English as a lingua franca. In G. J. Ockey & B. A. Green (Eds.), Another generation of fundamental considerations in language testing (pp. 9–28). Springer. Google Scholar logo with link to Google ScholarOckey, G. J., & Hirch, R. R. (2020) A step toward the assessment of English as a lingua franca. In G. J. Ockey & B. A. Green (Eds.), Another generation of fundamental considerations in language testing (pp. 9–28). Springer. Google Scholar logo with link to Google Scholar), and this may be the case for other languages as well.

Interactional competence principles have also been used to guide the design and delivery of OC assessments, and several researchers have indicated the importance of their inclusion in the rubrics to evaluate test takers’ OC abilities (Galaczi & Taylor, 2018Galaczi, E., & Taylor, L. (2018) Interactional competence: Conceptualisations, operationalisations, and outstanding questions. Language Assessment Quarterly, 15 (3), 219–236. Google Scholar logo with link to Google ScholarGalaczi, E., & Taylor, L. (2018) Interactional competence: Conceptualisations, operationalisations, and outstanding questions. Language Assessment Quarterly, 15(3), 219–236. Google Scholar logo with link to Google Scholar; Ockey & Li, 2015Ockey, G. J., & Li, Z. (2015) New and not so new methods for assessing oral communication. Language Value, 7 (1), 1–21. Google Scholar logo with link to Google ScholarOckey, G. J., & Li, Z. (2015) New and not so new methods for assessing oral communication. Language Value, 7(1), 1–21. Google Scholar logo with link to Google Scholar; Vo, 2019Vo, S. (2019) Effects of task types on interactional competence in oral communication assessment (Publication No. 17345). Doctoral dissertation, Iowa State University. ProQuest Dissertation and Theses Global.Vo, S. (2019) Effects of task types on interactional competence in oral communication assessment (Publication No. 17345). Doctoral dissertation, Iowa State University. ProQuest Dissertation and Theses Global.), in addition to other subconstructs such as fluency, pronunciation, intelligibility, grammar, and vocabulary. According to Galaczi and Taylor (2018)Galaczi, E., & Taylor, L. (2018) Interactional competence: Conceptualisations, operationalisations, and outstanding questions. Language Assessment Quarterly, 15 (3), 219–236. Google Scholar logo with link to Google ScholarGalaczi, E., & Taylor, L. (2018) Interactional competence: Conceptualisations, operationalisations, and outstanding questions. Language Assessment Quarterly, 15(3), 219–236. Google Scholar logo with link to Google Scholar, interactional competence features could be verbal and/or nonverbal, and both are important for engaging in successful oral communication. Verbal features include topic management, turn management, and breakdown repair; while non-verbal features include facial expressions, posture, and eye contact.

According to Mehan (1979)Mehan, H. (1979) Learning lessons: Social organization in the classroom. Harvard UP. Google Scholar logo with link to Google ScholarMehan, H. (1979) Learning lessons: Social organization in the classroom. Harvard UP. Google Scholar logo with link to Google Scholar, there are two understandings of the concept of interactional competence: the first refers to the “competence necessary for effective interaction,” and the second denotes “the competence that is available in the interaction between participants” (p. 130). The first definition implies that interactional competence is an individual ability, while the latter refers to the competence existing in the interaction between interlocutors to construct a shared meaning. Dai (2024)Dai, D. W. (2024) Assessing interactional competence: Principles, test development and validation through an L2 Chinese IC test (p. 446). Peter Lang International Academic Publishers. Google Scholar logo with link to Google ScholarDai, D. W. (2024) Assessing interactional competence: Principles, test development and validation through an L2 Chinese IC test (p. 446). Peter Lang International Academic Publishers. Google Scholar logo with link to Google Scholar extends Mehan’s first conceptualization by defining it to mean that competence necessary for effective interaction includes emotional, logical, moral, and categorial dimensions.

Mehan’s second conceptualization makes it difficult, if not impossible, to effectively estimate an individual’s ability, which is the goal of most OC assessments. As a result, most language assessment researchers adhere to the first conceptualization, while recognizing that interlocutors may influence a test taker’s score depending on the task type that is used, and according to the personal characteristics of the test takers and the interlocutor, such as shyness, proficiency level, and speech variety (e.g., Neiriz, 2023Neiriz, R. (2023) Eliciting interactive oral communication samples through a spoken dialogue system to measure interactional competence. Unpublished doctoral dissertation, Iowa State University. Neiriz, R. (2023) Eliciting interactive oral communication samples through a spoken dialogue system to measure interactional competence. Unpublished doctoral dissertation, Iowa State University. ; Ockey et al., 2023Ockey, G. J., Chukharev-Hudilainen, E., & Hirch, R. (2023) Assessing interactional competence: Interactional Competence Elicitor (ICE) vs a human partner. Language Assessment Quarterly, 20 (4–5), 377–398. Google Scholar logo with link to Google ScholarOckey, G. J., Chukharev-Hudilainen, E., & Hirch, R. (2023) Assessing interactional competence: Interactional Competence Elicitor (ICE) vs a human partner. Language Assessment Quarterly, 20(4–5), 377–398. Google Scholar logo with link to Google Scholar, Vo, 2019Vo, S. (2019) Effects of task types on interactional competence in oral communication assessment (Publication No. 17345). Doctoral dissertation, Iowa State University. ProQuest Dissertation and Theses Global.Vo, S. (2019) Effects of task types on interactional competence in oral communication assessment (Publication No. 17345). Doctoral dissertation, Iowa State University. ProQuest Dissertation and Theses Global.).

Within the call for including interactive and integrated tasks in assessment of OC ability (Brooks, 2009Brooks, L. (2009) Interacting in pairs in a test of oral proficiency: Co-constructing a better performance. Language Testing, 26 , 342–366. Google Scholar logo with link to Google ScholarBrooks, L. (2009) Interacting in pairs in a test of oral proficiency: Co-constructing a better performance. Language Testing, 26, 342–366. Google Scholar logo with link to Google Scholar; Ducass & Brown, 2009Ducasse, A. M., & Brown, A. (2009) Assessing paired orals: Raters’ orientation to interaction. Language Testing, 26 , 423–443. Google Scholar logo with link to Google ScholarDucasse, A. M., & Brown, A. (2009) Assessing paired orals: Raters’ orientation to interaction. Language Testing, 26, 423–443. Google Scholar logo with link to Google Scholar), language testers have become more hesitant to allow practical concerns associated with the exam settings, such as the need for an interlocutor, to impact assessments (e.g., Na, 2025Na, I. (2025, April). GenAI in assessing interactional competence: Examining differences between human-human and human-ai interactions. Language Assessment Research Conference. Fullerton, California, USA.Na, I. (2025, April). GenAI in assessing interactional competence: Examining differences between human-human and human-ai interactions. Language Assessment Research Conference. Fullerton, California, USA.; Neiriz, 2023Neiriz, R. (2023) Eliciting interactive oral communication samples through a spoken dialogue system to measure interactional competence. Unpublished doctoral dissertation, Iowa State University. Neiriz, R. (2023) Eliciting interactive oral communication samples through a spoken dialogue system to measure interactional competence. Unpublished doctoral dissertation, Iowa State University. ). Fortunately, technology has helped to minimize concerns about practical issues. For example, video-mediated tools have negated the need for test takers’ in-person attendance at an exam, to allow the process of administering exams to test takers from around the world. In addition, the test takers’ performances can be video recorded; this allows raters to evaluate them anytime at their convenience, and prevents the rating process from becoming exhausting. Also, technological innovations such as a spoken dialogue system (SDS) — “A computational system that can orally converse with individuals to create interactive speech samples” (Ockey et al., 2023Ockey, G. J., Chukharev-Hudilainen, E., & Hirch, R. (2023) Assessing interactional competence: Interactional Competence Elicitor (ICE) vs a human partner. Language Assessment Quarterly, 20 (4–5), 377–398. Google Scholar logo with link to Google ScholarOckey, G. J., Chukharev-Hudilainen, E., & Hirch, R. (2023) Assessing interactional competence: Interactional Competence Elicitor (ICE) vs a human partner. Language Assessment Quarterly, 20(4–5), 377–398. Google Scholar logo with link to Google Scholar p. 1) — may have reduced the need for a human interlocutor (e.g., an interviewer or peer partner). Human interlocutors are a costly and impractical part of oral assessments that aim to measure all aspects of oral communication, including interactional competence. Thus, the inclusion of technology as part of OC assessments has provided an alternative to face-to-face (F2F) human task delivery.

This paper discusses how various types of technology, including Video-mediated tools and spoken dialogue systems have been applied in OC assessment. It explores how they have both helped and introduced new challenges to this process. The focus is on the assessment of interactional competence, given that compared to other aspects of OC (e.g., pronunciation, spoken grammar/vocabulary, and fluency), it has been shown to be the most challenging to assess, and can be the aspect most often neglected (Ockey et al., 2015Ockey, G. J., Koyama, D., Setoguchi, E., & Sun, A. (2015) The extent to which TOEFL iBT speaking scores are associated with performance on oral language tasks and oral ability components for Japanese university students. Language Testing, 32 (1), 39–62. Google Scholar logo with link to Google ScholarOckey, G. J., Koyama, D., Setoguchi, E., & Sun, A. (2015) The extent to which TOEFL iBT speaking scores are associated with performance on oral language tasks and oral ability components for Japanese university students. Language Testing, 32(1), 39–62. Google Scholar logo with link to Google Scholar; Ockey & Chukharev-Hudilainen, 2021Ockey, G. J., & Chukharev-Hudilainen, E. (2021) Human versus computer partner in the paired oral discussion test. Applied Linguistics, 42 (5), 924–944. Google Scholar logo with link to Google ScholarOckey, G. J., & Chukharev-Hudilainen, E. (2021) Human versus computer partner in the paired oral discussion test. Applied Linguistics, 42(5), 924–944. Google Scholar logo with link to Google Scholar). The paper concludes with suggestions for future research.

Video-mediated OC assessments

Video-mediated platforms (e.g., Skype, Zoom, and Teams) allow test takers to interact orally and to show their physical appearance (actual faces) while interacting. Early application of video-mediated OC assessment started by determining whether or not it measures a different construct from that elicited in a F2F format (Kim & Craig, 2012Kim, J., & Craig, D. A. (2012) Validation of a videoconferenced speaking test. Computer Assisted Language Learning, 25(3), 257–275. Google Scholar logo with link to Google ScholarKim, J., & Craig, D. A. (2012) Validation of a videoconferenced speaking test. Computer Assisted Language Learning, 25(3), 257–275. Google Scholar logo with link to Google Scholar; Lee et al., 2024Lee, H. W., Mullooly, A., Devine, A., & Galaczi, E. (2024) Exploring interaction in video-call paired speaking tests: A look at scores, language, and perceptions. Applied Linguistics, 45 (3), 533–552. Google Scholar logo with link to Google ScholarLee, H. W., Mullooly, A., Devine, A., & Galaczi, E. (2024) Exploring interaction in video-call paired speaking tests: A look at scores, language, and perceptions. Applied Linguistics, 45(3), 533–552. Google Scholar logo with link to Google Scholar; Nakatsuhara et al., 2017Nakatsuhara, F., Inoue, C., Berry, V., & Galaczi, E. (2017) Exploring the use of video-conferencing technology in the assessment of spoken language: A mixed-methods study. Language Assessment Quarterly, 14 (1), 1–18. Google Scholar logo with link to Google ScholarNakatsuhara, F., Inoue, C., Berry, V., & Galaczi, E. (2017) Exploring the use of video-conferencing technology in the assessment of spoken language: A mixed-methods study. Language Assessment Quarterly, 14(1), 1–18. Google Scholar logo with link to Google Scholar; Zhang & Jin, 2021Zhang, L., & Jin, Y. (2021) Assessing interactional competence in the computer-based CET-SET: An investigation of the use of communication strategies. Assessment in education: Principles, policy & practice, 28 (4), 389–410. Google Scholar logo with link to Google ScholarZhang, L., & Jin, Y. (2021) Assessing interactional competence in the computer-based CET-SET: An investigation of the use of communication strategies. Assessment in education: Principles, policy & practice, 28(4), 389–410. Google Scholar logo with link to Google Scholar), taking the approach that F2F human interaction is the gold standard for assessing oral communication. These studies generally found that the scores students received in a F2F format did not differ significantly from those in the video-mediated format (Kim & Craig, 2012Kim, J., & Craig, D. A. (2012) Validation of a videoconferenced speaking test. Computer Assisted Language Learning, 25(3), 257–275. Google Scholar logo with link to Google ScholarKim, J., & Craig, D. A. (2012) Validation of a videoconferenced speaking test. Computer Assisted Language Learning, 25(3), 257–275. Google Scholar logo with link to Google Scholar; Lee et al., 2024Lee, H. W., Mullooly, A., Devine, A., & Galaczi, E. (2024) Exploring interaction in video-call paired speaking tests: A look at scores, language, and perceptions. Applied Linguistics, 45 (3), 533–552. Google Scholar logo with link to Google ScholarLee, H. W., Mullooly, A., Devine, A., & Galaczi, E. (2024) Exploring interaction in video-call paired speaking tests: A look at scores, language, and perceptions. Applied Linguistics, 45(3), 533–552. Google Scholar logo with link to Google Scholar; Nakatsuhara et al., 2017Nakatsuhara, F., Inoue, C., Berry, V., & Galaczi, E. (2017) Exploring the use of video-conferencing technology in the assessment of spoken language: A mixed-methods study. Language Assessment Quarterly, 14 (1), 1–18. Google Scholar logo with link to Google ScholarNakatsuhara, F., Inoue, C., Berry, V., & Galaczi, E. (2017) Exploring the use of video-conferencing technology in the assessment of spoken language: A mixed-methods study. Language Assessment Quarterly, 14(1), 1–18. Google Scholar logo with link to Google Scholar; Zhang & Jin, 2021Zhang, L., & Jin, Y. (2021) Assessing interactional competence in the computer-based CET-SET: An investigation of the use of communication strategies. Assessment in education: Principles, policy & practice, 28 (4), 389–410. Google Scholar logo with link to Google ScholarZhang, L., & Jin, Y. (2021) Assessing interactional competence in the computer-based CET-SET: An investigation of the use of communication strategies. Assessment in education: Principles, policy & practice, 28(4), 389–410. Google Scholar logo with link to Google Scholar); these studies provided promising evidence for the usefulness of video-mediated assessment in eliciting real-time conversations that require an interlocutor, such as an interviewer or a partner.

However, when examining the language produced, some differences have been found in the interactional competence features elicited by each format. For example, Nakatsuhara et al. (2017)Nakatsuhara, F., Inoue, C., Berry, V., & Galaczi, E. (2017) Exploring the use of video-conferencing technology in the assessment of spoken language: A mixed-methods study. Language Assessment Quarterly, 14 (1), 1–18. Google Scholar logo with link to Google ScholarNakatsuhara, F., Inoue, C., Berry, V., & Galaczi, E. (2017) Exploring the use of video-conferencing technology in the assessment of spoken language: A mixed-methods study. Language Assessment Quarterly, 14(1), 1–18. Google Scholar logo with link to Google Scholar discovered that in the International English Language Testing System (IELTS) speaking test, candidates tended to ask for clarification more often when completing an oral interview in a video conferencing mode, compared to a F2F mode, while features of comparing and suggesting were more prominent in F2F. In a related study, Lee et al. (2024)Lee, H. W., Mullooly, A., Devine, A., & Galaczi, E. (2024) Exploring interaction in video-call paired speaking tests: A look at scores, language, and perceptions. Applied Linguistics, 45 (3), 533–552. Google Scholar logo with link to Google ScholarLee, H. W., Mullooly, A., Devine, A., & Galaczi, E. (2024) Exploring interaction in video-call paired speaking tests: A look at scores, language, and perceptions. Applied Linguistics, 45(3), 533–552. Google Scholar logo with link to Google Scholar asked 58 test takers to perform a paired discussion task in two modes: F2F versus video conferencing. In the F2F mode, the exam was conducted in person, while for the video conferencing mode, test takers performed the exam from different locations via Zoom. The analysis showed that explicit turn taking appeared to be more common in the video conferencing mode, while instances of reaching a joint decision and modifying the partner’s speech were more common in F2F; nevertheless, the difference was not substantial. Zhang and Jin (2021)Zhang, L., & Jin, Y. (2021) Assessing interactional competence in the computer-based CET-SET: An investigation of the use of communication strategies. Assessment in education: Principles, policy & practice, 28 (4), 389–410. Google Scholar logo with link to Google ScholarZhang, L., & Jin, Y. (2021) Assessing interactional competence in the computer-based CET-SET: An investigation of the use of communication strategies. Assessment in education: Principles, policy & practice, 28(4), 389–410. Google Scholar logo with link to Google Scholar found that students tended to use more cooperative and meaning-negotiation strategies during a paired discussion task in the computer-mediated version of the Computer-based College English Test–Spoken English Test (CET-SET) in China, compared to the F2F version. These strategies included asking a question and giving feedback, making/responding to a repetition request, and making/responding to a comprehension check. The findings of these studies suggested that speech via a video-mediated system is somewhat different from that elicited from a F2F format.

Although several studies aimed to compare the two modes, and seemingly hoped for similar results, Ockey et al. (2019)Ockey, G. J., Timpe-Laughlin, V., Davis, L., & Gu, L. (2019) Exploring the potential of a video-mediated interactive speaking assessment (Report No. ETS RR-19-05). Princeton, NJ: Educational Testing Service. Google Scholar logo with link to Google ScholarOckey, G. J., Timpe-Laughlin, V., Davis, L., & Gu, L. (2019) Exploring the potential of a video-mediated interactive speaking assessment (Report No. ETS RR-19-05). Princeton, NJ: Educational Testing Service. Google Scholar logo with link to Google Scholar believed that comparability between the two modes should not be a goal. In their view, the application of such technology in OC assessment is part of the target language use (TLU) context, and should therefore be implemented in OC assessments, whether it elicits speech similar to the F2F format or not. Their argument has only been made stronger by the rapid increase in the use of video-mediated conversations during and after the COVID-19 pandemic (e.g., hybrid classes, online meetings) within academic contexts.

Virtual environment OC assessments

Some studies have used Web-based virtual environments (VEs) as the mode of delivery for OC assessments. VEs are “a virtual space in which individuals, who are visually represented by avatars, can verbally interact in real time” (Ockey et al., 2017Ockey, G. J., Gu, L., & Keehner, M. (2017) Web-based virtual environments for facilitating assessment of L2 Oral communication ability. Language Assessment Quarterly, 14 (4), 346–359. Google Scholar logo with link to Google ScholarOckey, G. J., Gu, L., & Keehner, M. (2017) Web-based virtual environments for facilitating assessment of L2 Oral communication ability. Language Assessment Quarterly, 14(4), 346–359. Google Scholar logo with link to Google Scholar, p. 346). Through a VE application (e.g., Second Life), the context of the virtual reality replicates a real-life scenario, (e.g., students meeting in a library), an important advantage of this system, and test takers and test administrators are represented by avatars (e.g., Ockey et al., 2017Ockey, G. J., Gu, L., & Keehner, M. (2017) Web-based virtual environments for facilitating assessment of L2 Oral communication ability. Language Assessment Quarterly, 14 (4), 346–359. Google Scholar logo with link to Google ScholarOckey, G. J., Gu, L., & Keehner, M. (2017) Web-based virtual environments for facilitating assessment of L2 Oral communication ability. Language Assessment Quarterly, 14(4), 346–359. Google Scholar logo with link to Google Scholar). The major difference between this technology and video-mediated platforms is the visualization of an avatar versus the interlocutor.

Ockey et al. (2017)Ockey, G. J., Gu, L., & Keehner, M. (2017) Web-based virtual environments for facilitating assessment of L2 Oral communication ability. Language Assessment Quarterly, 14 (4), 346–359. Google Scholar logo with link to Google ScholarOckey, G. J., Gu, L., & Keehner, M. (2017) Web-based virtual environments for facilitating assessment of L2 Oral communication ability. Language Assessment Quarterly, 14(4), 346–359. Google Scholar logo with link to Google Scholar conducted a small-scale exploratory study of a Web-based VE and its role in assessing oral communication. Eighteen international students performed a group oral communication task; through the use of avatars, a group of three test takers were able to communicate with each other in a virtual place that replicated a real-life context (students meeting in the library). After completing the task, test takers filled out a five-point Likert scale questionnaire; findings showed that most students believed that the VE was suitable for conducting an interactive discussion, and that the environment felt authentic enough to replicate a real-life context, despite the presence of some pauses in the speech, which made it sound less natural.

More recently, researchers have attempted to make virtual environments more realistic by including virtual reality (VR). VR refers to a 3D immersive environment that allows its users to interact in a way that is designed to feel real and physical. Muhanna (2015)Muhanna, M. A. (2015) Virtual reality and the CAVE: Taxonomy, interaction challenges and research directions. Journal of King Saud University-Computer and Information Sciences, 27 (3), 344–361. Google Scholar logo with link to Google ScholarMuhanna, M. A. (2015) Virtual reality and the CAVE: Taxonomy, interaction challenges and research directions. Journal of King Saud University-Computer and Information Sciences, 27(3), 344–361. Google Scholar logo with link to Google Scholar explained that the concept “immersive” indicates “a state in which a participant becomes attracted and involved in a virtual space of an activity to an extension that his or her mind is separated from the physical space he or she is being active in” (p. 347). In one VR oral communication assessment study, Kurt (2025)Kurt, S. (2025) Development and evaluation of a virtual reality environment for the Oral English Certification Test. Unpublished doctoral dissertation, Iowa State University. Kurt, S. (2025) Development and evaluation of a virtual reality environment for the Oral English Certification Test. Unpublished doctoral dissertation, Iowa State University. asked international teaching assistants (ITAs) to wear a VR headset and conduct an oral communication exam once in a VR environment and once in a F2F environment. She aimed to investigate whether a VR-based assessment can provide reliable scores, in order to be used as an alternative to a F2F environment. In the VR test format, an ITA was asked to enter a virtual exam room, choose an avatar, and select the role of a student to perform an oral interview in a virtual class. At the same time, raters entered the exam session, chose an avatar, and selected the role of a rater of the ITA’s performance, using the scoring interface that was built within the VR environment. The system allowed test takers and raters in different physical places to attend a shared virtual exam session, and to speak directly using their human voices. The findings showed that the VR was perceived to be a useful testing format that leads to fair scores, by showing high score dependability. Regarding the rating process, there was less variation among the raters and higher agreement in the VR format, compared to that found in the F2F mode. However, when assessing interactional competence, the VR format missed the evaluation of non-verbal communication such as facial expressions.

In another VR study, Song and Hsu (2021)Song, J., & Hsu, W. L. (2021) Design and implementation of a classroom-based virtual reality assessment. In Salaberry, M. & Burch, A. (Eds.), Assessing speaking in context: Expanding the construct and its applications (pp. 265–282). Channel View Publications. Google Scholar logo with link to Google ScholarSong, J., & Hsu, W. L. (2021) Design and implementation of a classroom-based virtual reality assessment. In Salaberry, M. & Burch, A. (Eds.), Assessing speaking in context: Expanding the construct and its applications (pp. 265–282). Channel View Publications. Google Scholar logo with link to Google Scholar asked test takers to wear VR headsets for a classroom speaking assessment. In their study, 24 Korean tertiary-level students were required to accept a friend’s invitation to a movie or dinner on a VR platform. Each test taker sat in front of a computer and read the prompt including the requested act of speech (e.g., accepting or refusing an invitation). Then, they used VR goggles (HTC VIVE) to enter an immersive environment, where they completed an oral communication exam. The VR recorded speech started by greeting the test taker, asking a warmup question, and then inviting the test taker to a movie, while the test taker had to reply to the greeting, answer the warmup question, refuse the invitation, and close the conversation. An avatar was the interlocutor in the VR-mediated mode, and the researcher was the interlocutor in the F2F test. In terms of language, the researchers found that features such as asking for information and negotiating meaning were used to a lesser extent in the VR condition, compared to the F2F condition. The authors explained that such an outcome was expected since the VR speech was pre-recorded, and students’ questioning would not be answered by the VR interlocutor.

Since VR technology does not show test takers’ physical appearance, it limits the ability to consider non-verbal communication as part of the OC construct. Yet, there is the potential that virtual environments may be considered a means of assessing L2 oral communication when non-verbal features are not part of the construct. Specifically, some studies have shown that test takers preferred the VR approach, because they felt less stress when they did not see their interlocutor. These test takers felt this made it possible for them to better demonstrate their OC abilities (Ockey et al., 2017Ockey, G. J., Gu, L., & Keehner, M. (2017) Web-based virtual environments for facilitating assessment of L2 Oral communication ability. Language Assessment Quarterly, 14 (4), 346–359. Google Scholar logo with link to Google ScholarOckey, G. J., Gu, L., & Keehner, M. (2017) Web-based virtual environments for facilitating assessment of L2 Oral communication ability. Language Assessment Quarterly, 14(4), 346–359. Google Scholar logo with link to Google Scholar). Moreover, VEs may provide a more secure assessment environment than video-mediated, since only audio information is delivered (and recorded).

On the other hand, in many real-world contexts, physical appearance is present. Some might argue that to determine how effectively a test taker can converse in such a context, it is necessary to include it in the assessment. For instance, test takers may be able to talk to others when they cannot see them, but they may struggle when they can. If the goal of the assessment is to determine how effectively test takers can orally communicate with others, assessments with avatars may not give an accurate indication of what a test taker can do in a F2F situation.

Importantly, younger generations are more exposed to VEs and the gaming world, where they talk to avatars instead of real faces. If their use of VEs is increasing, including it in OC assessment would replicate how the language is used in real-life contexts. However, research is needed to confirm this hypothesis. A needs analysis which involves the process of collecting information about daily conversational activities in which test takers within a specific context engage, as well as the technologies used to replicate the TLU domain is needed to confirm this hypothesis.

Rating in video-mediated and virtual environments

Another important question that language assessment researchers have considered is whether or not raters should be able to see the test takers when they rate them, and if they do, how this might affect their judgments. Other fields of study have found that physical appearance can have an effect on raters’ judgments of peoples’ abilities. For instance, in marketing research, the attractiveness of women representing a product is thought to impact buying choices, and in some cases, consumers have paid more money when the product was presented by an attractive female (Buunk & Dijkstra, 2011Buunk, A. P., & Dijkstra, P. (2011) Does attractiveness sell? Women’s attitude toward a product as a function of model attractiveness, gender priming, and social comparison orientation. Psychology & Marketing, 28 ( 9 ), 958–973. Google Scholar logo with link to Google ScholarBuunk, A. P., & Dijkstra, P. (2011) Does attractiveness sell? Women’s attitude toward a product as a function of model attractiveness, gender priming, and social comparison orientation. Psychology & Marketing, 28(9), 958–973. Google Scholar logo with link to Google Scholar). Also, research has shown that applicants who were perceived as attractive were at an advantage, as they received better job opportunities and hiring recommendations than their less attractive peers (Hosoda et al., 2003Hosoda, M., Stone-Romero, E. F., & Coats, G. (2003) The effects of physical attractiveness on job-related outcomes: A meta-analysis of experimental studies. Personnel Psychology, 56 ( 2 ), 431–462. Google Scholar logo with link to Google ScholarHosoda, M., Stone-Romero, E. F., & Coats, G. (2003) The effects of physical attractiveness on job-related outcomes: A meta-analysis of experimental studies. Personnel Psychology, 56(2), 431–462. Google Scholar logo with link to Google Scholar).

Research in the field of second language assessment has produced similar findings. For example, Aseeri (2025)Aseeri, F. M. (2025) Judging a book by its cover: Investigating the influence of test takers’ physical attractiveness in synchronous oral communication assessment. Unpublished doctoral dissertation, Iowa State University.Aseeri, F. M. (2025) Judging a book by its cover: Investigating the influence of test takers’ physical attractiveness in synchronous oral communication assessment. Unpublished doctoral dissertation, Iowa State University. investigated to what extent the physical appearance of the test takers impacted raters’ judgment of 30 test takers’ OC performance. Five experienced and trained raters used two evaluation modes. Using the Zoom platform, test takers conducted the exam twice: once where the test takers’ physical appearance was shown to the raters, and once where an avatar instead of their physical appearance was presented. The Zoom platform was selected because the Zoom Avatars are quite realistic: they are designed to mimic head movements and facial expressions, to provide a more natural appearance and more authentic communication. Many-facet Rasch Measurement analysis showed that test takers received slightly higher scores when their physical appearance was shown to the rater, compared to when not shown. Although the difference was not significant, it resulted in four out of the 30 test takers failing the exam when their physical appearance was not shown, while they passed when it was shown; this implies the presence of bias when seeing their appearance.

In addition, Aseeri (2025)Aseeri, F. M. (2025) Judging a book by its cover: Investigating the influence of test takers’ physical attractiveness in synchronous oral communication assessment. Unpublished doctoral dissertation, Iowa State University.Aseeri, F. M. (2025) Judging a book by its cover: Investigating the influence of test takers’ physical attractiveness in synchronous oral communication assessment. Unpublished doctoral dissertation, Iowa State University. investigated whether physical attractiveness is an explanatory variable in OC assessment. The attractiveness of 41 male and female participants was measured using an attractiveness scale; their scores ranged from 59 to 77 for females and from 53 to 71 for males. Simple regression analysis showed that pronunciation, as a subconstruct of OC, was significantly predicted by physical attractiveness (F = 6.04, p = .02). In particular, the most attractive female participant received about one point higher (.88) on pronunciation, compared to a male test taker who received about half a point higher (.58) on a four-point scale; this suggests that females’ scores are impacted by attractiveness more than males.

Aseeri’s (2025)Aseeri, F. M. (2025) Judging a book by its cover: Investigating the influence of test takers’ physical attractiveness in synchronous oral communication assessment. Unpublished doctoral dissertation, Iowa State University.Aseeri, F. M. (2025) Judging a book by its cover: Investigating the influence of test takers’ physical attractiveness in synchronous oral communication assessment. Unpublished doctoral dissertation, Iowa State University. findings suggest that physical appearance can impact raters’ scores despite systematic training and rating experience. Few people, however, would consider an effect of physical appearance on scores as appropriate or fair even though it may be part of the way the real-world functions. The study clearly suggests that in certain contexts, test developers should consider approaches to the assessment of oral communication where raters cannot see the test takers, such as VE, even though it limits the assessment’s authenticity. Stakeholders must make decisions about the type of delivery mode they think is most appropriate based on the construct they desire to assess and the potential biases that may emerge.

Video-mediated and VE technology offer benefits for resolving some issues (e.g., the need for in-person raters or interlocutors) and may lessen test takers’ fear of being looked at or making eye contact with others. However, they introduce new challenges to assessing oral communication that are not present when F2F delivery is used. For example, technological glitches may occur, and sound and video quality may introduce challenges. In addition, they require a large data storage space to save the recorded videos, and a high level of security. Thus, it is recommended to have clear guidelines when applying such types of technology, including the need for a stable and reliable internet connection, and secure storage. In addition, test developers need to ensure that alternative solutions are ready to implement when technological breakdowns, such as low sound quality or video glitches, occur during an exam session.

Although research shows promising results for OC assessments delivered through video-mediated platforms or virtual environments, further research is needed to understand the potential of other technological developments. For example, can VEs such as IMMERSE be implemented effectively in an OC assessment (https://​www​.immerse​.com/)? IMMERSE could allow test administrators to engage test takers in various activities during an exam session, such as watching a video clip, role-playing, and/or delivering a presentation (see Figure 1).

Figure 1.IMMERSE example
Figure 1.

Questions that deserve an answer are: To what extent can IMMERSE VEs be implemented in OC assessment? What advantages do they offer for the assessment of oral communication? What oral communication features could be elicited from this type of technology, compared to a video-mediated or F2F OC exam?

Spoken dialogue systems for assessing oral communication

Spoken dialogue systems (SDSs) are computerized systems that can be used to deliver oral communication assessments. They can serve as a partner for a one-on-one oral interview, paired or group discussion, and many other types of real-time oral interactions (Ockey, 2025 (2025) Introducing second language assessment. Cambridge University Press.Google Scholar logo with link to Google Scholar (2025) Introducing second language assessment. Cambridge University Press.Google Scholar logo with link to Google Scholar). Having a human partner may result in test takers’ oral performance being affected by their partner’s personal characteristics, such as personality (Ockey, 2009Ockey, G. J. (2009) The effects of a test taker’s group members’ personalities on the test taker’s second language group oral discussion test scores. Language Testing, 26 ( 2 ), 161–186. Google Scholar logo with link to Google ScholarOckey, G. J. (2009) The effects of a test taker’s group members’ personalities on the test taker’s second language group oral discussion test scores. Language Testing, 26 (2), 161–186. Google Scholar logo with link to Google Scholar), and a computer partner can eliminate such effects by standardizing the partner for all test takers. Early applications of SDSs restricted the type of speech elicited from the test takers. The test takers did not engage in spontaneous multi-turn conversational speech; instead, they were based on simple predictable dialogic speech (Ramanarayanan et al., 2019Ramanarayanan, V., Evanini, K., & Tsuprun, E. (2019) Beyond monologues. In Zechner, K., & Evanini, K. (Eds.). Automated speaking assessment: Using language technologies to score spontaneous speech (pp. 176–191). Routledge. Google Scholar logo with link to Google ScholarRamanarayanan, V., Evanini, K., & Tsuprun, E. (2019) Beyond monologues. In Zechner, K., & Evanini, K. (Eds.). Automated speaking assessment: Using language technologies to score spontaneous speech (pp. 176–191). Routledge. Google Scholar logo with link to Google Scholar), such as responses to questions about what to put on a pizza (Ockey & Neiriz, 2021Ockey, G. J., & Neiriz, R. (2021) Evaluating technology-mediated second language oral communication assessment delivery models. Assessment in Education: Principles, Policy & Practice, 28 (4), 350–368. Google Scholar logo with link to Google ScholarOckey, G. J., & Neiriz, R. (2021) Evaluating technology-mediated second language oral communication assessment delivery models. Assessment in Education: Principles, Policy & Practice, 28(4), 350–368. Google Scholar logo with link to Google Scholar). Subsequently, SDS systems were further developed to elicit a more naturalistic dialogue experience with unconstrained responses, and to elicit features of interactional competence. In general, SDSs rely on databases to guide their responses. Two common types of SDSs are rule-based and large language model-driven (LLM) systems.

Rule-based SDSs

Simple rule-based systems function by recognizing keywords or phrases in the test taker’s speech. Then these keywords or phrases are matched with language in a language corpus, which is often created for the precise prompts that are used in the language assessment. The SDS can then provide an appropriate response to the test taker which has been created by another human for the same or similar spoken discourse. For example, Ockey and Chukharev-Hudilainen (2021)Ockey, G. J., & Chukharev-Hudilainen, E. (2021) Human versus computer partner in the paired oral discussion test. Applied Linguistics, 42 (5), 924–944. Google Scholar logo with link to Google ScholarOckey, G. J., & Chukharev-Hudilainen, E. (2021) Human versus computer partner in the paired oral discussion test. Applied Linguistics, 42(5), 924–944. Google Scholar logo with link to Google Scholar and Ockey et al. (2023)Ockey, G. J., Chukharev-Hudilainen, E., & Hirch, R. (2023) Assessing interactional competence: Interactional Competence Elicitor (ICE) vs a human partner. Language Assessment Quarterly, 20 (4–5), 377–398. Google Scholar logo with link to Google ScholarOckey, G. J., Chukharev-Hudilainen, E., & Hirch, R. (2023) Assessing interactional competence: Interactional Competence Elicitor (ICE) vs a human partner. Language Assessment Quarterly, 20(4–5), 377–398. Google Scholar logo with link to Google Scholar created a rule-based SDS, called Interactional Competence Elicitor, for a paired speaking assessment, by asking many pairs of proficient English speakers to discuss a topic based on a prompt. The prompt included two 40-second inputs: one in favor of the value of group work for university students, and one against its value. After listening to this input, the pairs were asked to take opposing views on the value of group work. Keywords and phrases that commonly appeared in these discussions were then extracted. Appropriate responses when these keywords or phrases were used were also taken from these discussions and programmed as responses when the test taker used the associated key word or phrase. This small corpus was used to program the SDS to discuss the topic with a test taker. The researchers reported that the system worked reasonably well, based on post-test interviews with test takers, and an analysis of the number of errors in recognizing what the test taker said and giving an appropriate response (Chukharev-Hudilainen & Ockey, 2021Chukharev-Hudilainen, E., & Ockey, G. J. (2021) The development and evaluation of interactional competence elicitor for oral language assessments. ETS Research Report Series, 2021 (1), 1–20. Google Scholar logo with link to Google ScholarChukharev-Hudilainen, E., & Ockey, G. J. (2021) The development and evaluation of interactional competence elicitor for oral language assessments. ETS Research Report Series, 2021(1), 1–20. Google Scholar logo with link to Google Scholar).

The effectiveness of the SDS compared to a human partner was addressed by Ockey and Chukharev-Hudilainen (2021)Ockey, G. J., & Chukharev-Hudilainen, E. (2021) Human versus computer partner in the paired oral discussion test. Applied Linguistics, 42 (5), 924–944. Google Scholar logo with link to Google ScholarOckey, G. J., & Chukharev-Hudilainen, E. (2021) Human versus computer partner in the paired oral discussion test. Applied Linguistics, 42(5), 924–944. Google Scholar logo with link to Google Scholar. They found that the SDS resulted in more reliable scores (Eρ2 = .85 for the SDS partner condition and Eρ2 = .73 for the human partner condition, for four raters). The SDS condition also led to a more uniform delivery system by ensuring that each test taker heard the same speech variety and level of English proficiency and had similar opportunities to demonstrate aspects of interactional competence. The researchers operationalized interactional competence to include the ability to effectively check comprehension, clarify a request, confirm comprehension, ask for an opinion/information, initiate topics, develop topics, connect topics, self-correct mistakes, reply in an appropriate amount of time, respond to clarification requests, elaborate on a point, and recognize a position. The researchers leveraged the SDS by intentionally including conversation breakdowns in the discourse. For example, the SDS would ask the test taker a nonsense question to help determine how effectively the latter could ask for clarification of an unclear phrase. The findings indicated that the SDS provided more opportunities for test takers to demonstrate their interactional competence than the human partner condition. This was in part because some human–human interactions did not include conversation breakdowns, while all SDS-human interactions did, because such breakdowns were programmed into each interaction. The researchers argued that in a human–machine conversation, test takers cannot fake understanding as they may do in a human–human conversation. Interestingly, an equal number of test takers preferred having an SDS partner as those who favored a human partner. Ockey et al. (2023)Ockey, G. J., Chukharev-Hudilainen, E., & Hirch, R. (2023) Assessing interactional competence: Interactional Competence Elicitor (ICE) vs a human partner. Language Assessment Quarterly, 20 (4–5), 377–398. Google Scholar logo with link to Google ScholarOckey, G. J., Chukharev-Hudilainen, E., & Hirch, R. (2023) Assessing interactional competence: Interactional Competence Elicitor (ICE) vs a human partner. Language Assessment Quarterly, 20(4–5), 377–398. Google Scholar logo with link to Google Scholar found that five of the 12 interactional competence features were more ratable when test takers interacted with a computer partner rather than a human partner. These features were clarifying a request, asking for information, self-correcting mistakes, responding to clarification of a request, and prompting. Only one feature, initiating topics, was judged to be more ratable in the human partner condition.

The main disadvantages of the SDS reported in Ockey and Chukharev-Hudilainen’s (2021)Ockey, G. J., & Chukharev-Hudilainen, E. (2021) Human versus computer partner in the paired oral discussion test. Applied Linguistics, 42 (5), 924–944. Google Scholar logo with link to Google ScholarOckey, G. J., & Chukharev-Hudilainen, E. (2021) Human versus computer partner in the paired oral discussion test. Applied Linguistics, 42(5), 924–944. Google Scholar logo with link to Google Scholar research were that it did not always respond appropriately (the conversation was sometimes not smooth), it tended to dominate the conversation, test takers did not feel the need to be pragmatically appropriate when speaking to it, and lack of body language made it difficult for test takers to follow conversational turn-taking conventions. Other researchers expanded on the work of Ockey and Chukharev-Hudilainen. In particular, Gokturk and Chukharev (2024)Gukturk, N., & Chukharev, E. (2024) Exploring the potential of a spoken dialog system-delivered paired discussion task for assessing interactional competence. Language Assessment Quarterly, 21 (1), 60–99. Google Scholar logo with link to Google ScholarGukturk, N., & Chukharev, E. (2024) Exploring the potential of a spoken dialog system-delivered paired discussion task for assessing interactional competence. Language Assessment Quarterly, 21(1), 60–99. Google Scholar logo with link to Google Scholar and Neiriz (2023)Neiriz, R. (2023) Eliciting interactive oral communication samples through a spoken dialogue system to measure interactional competence. Unpublished doctoral dissertation, Iowa State University. Neiriz, R. (2023) Eliciting interactive oral communication samples through a spoken dialogue system to measure interactional competence. Unpublished doctoral dissertation, Iowa State University. also created and evaluated the effectiveness of a rule-based SDS to measure interactional competence in a paired oral discussion test, while Na (2025)Na, I. (2025, April). GenAI in assessing interactional competence: Examining differences between human-human and human-ai interactions. Language Assessment Research Conference. Fullerton, California, USA.Na, I. (2025, April). GenAI in assessing interactional competence: Examining differences between human-human and human-ai interactions. Language Assessment Research Conference. Fullerton, California, USA. applied the principles to an LLM-driven SDS (discussed in the following section). In all three cases, the researchers used the same population of students, and their results supported those of Ockey and Chukharev-Hudilainen. For example, Gokturk and Chukharev (2024)Gukturk, N., & Chukharev, E. (2024) Exploring the potential of a spoken dialog system-delivered paired discussion task for assessing interactional competence. Language Assessment Quarterly, 21 (1), 60–99. Google Scholar logo with link to Google ScholarGukturk, N., & Chukharev, E. (2024) Exploring the potential of a spoken dialog system-delivered paired discussion task for assessing interactional competence. Language Assessment Quarterly, 21(1), 60–99. Google Scholar logo with link to Google Scholar used discourse analysis to compare the measurement of interactional competence with their rule-based SDS across varying levels of intermediate to advanced second-language university English speakers; they found that interactional competence features were mostly measurable and generally differed depending on the test takers’ ability level. Additionally, they reported that 77% of test takers felt that the system “could generally provide meaningful responses throughout the discussion” (p. 75). Interestingly, at least one test taker felt that the SDS’s misunderstandings were realistic, since many people did not understand the test taker in real-world conversations because of limited English proficiency. Neiriz (2023)Neiriz, R. (2023) Eliciting interactive oral communication samples through a spoken dialogue system to measure interactional competence. Unpublished doctoral dissertation, Iowa State University. Neiriz, R. (2023) Eliciting interactive oral communication samples through a spoken dialogue system to measure interactional competence. Unpublished doctoral dissertation, Iowa State University. also created a rule-based SDS with the aim of assessing interactional competence in a paired speaking task. His many-facet Rasch-based psychometric findings indicated that a rule-based SDS could effectively place intermediate to advanced speakers of English into different interactional competence ability levels. Neiriz provided an excellent discussion of test takers’ and raters’ views on the strengths and weaknesses of taking an oral assessment with an SDS, as compared to a human partner.

In addition to their use in paired discussion tasks, rule-based SDSs have also been used to serve as a speaking partner in role-play tasks. For instance, Karatay (2022)Karatay, Y. (2022) Development and validation of spoken dialog system-based oral communication tasks in an ESP context (publication No. 29165842). Doctoral dissertation, Iowa State University ProQuest Dissertations Publishing. Karatay, Y. (2022) Development and validation of spoken dialog system-based oral communication tasks in an ESP context (publication No. 29165842). Doctoral dissertation, Iowa State University ProQuest Dissertations Publishing. used a system that took the role of a dissatisfied hotel guest who complained to a test taker in a hotel receptionist’s role. Test takers were tasked with satisfying the guest without breaching hotel policies that were provided in the role-play guidelines. For instance, the SDS would complain that the assigned hotel room was too cold or smelled smokey. The test taker would be expected to satisfy the guest (played by the SDS) by offering to open the window or decrease the cost of the room by a small amount. Karatay reported that the SDS interactions led to consistent estimates of the test takers’ ability levels; specifically, the human-scored responses were highly generalizable and dependable.

In another study that used an SDS, Dombi et al. (2024)Dombi, J., Sydorenko, T., & Timpe-Laughlin, V. (2024) Openings and closings in human-human versus human-spoken dialogue system conversations. Language Learning & Technology, 28(2), 32–61. https://​hdl​.handle​.net​/10125​/73571. Dombi, J., Sydorenko, T., & Timpe-Laughlin, V. (2024) Openings and closings in human-human versus human-spoken dialogue system conversations. Language Learning & Technology, 28(2), 32–61. https://​hdl​.handle​.net​/10125​/73571. implemented a role play task by pairing 47 test takers with an SDS. Their SDS system used HALEF, an open-source cloud-based dialogue system that does not require programming knowledge, compared to other systems (Ramanarayanan et al., 2019Ramanarayanan, V., Evanini, K., & Tsuprun, E. (2019) Beyond monologues. In Zechner, K., & Evanini, K. (Eds.). Automated speaking assessment: Using language technologies to score spontaneous speech (pp. 176–191). Routledge. Google Scholar logo with link to Google ScholarRamanarayanan, V., Evanini, K., & Tsuprun, E. (2019) Beyond monologues. In Zechner, K., & Evanini, K. (Eds.). Automated speaking assessment: Using language technologies to score spontaneous speech (pp. 176–191). Routledge. Google Scholar logo with link to Google Scholar). The test takers were asked to call their boss, ask for a meeting, and for the possibility for their boss to review some presentation slides; once with a human partner (F2F) and once with an SDS. The authors investigated the use of specific interactional competence features, in particular, opening and closing a conversation in the SDS system, and compared it to how such features were used in F2F interaction. They found that test takers were mostly trying to finish the task at hand when talking with the automated agent, but their speech was longer and more engaging when a human interlocutor was the partner. A possible explanation could be that their SDS system was designed to search for keywords instead of following up the open and closing sequences, resulting in more of a structured dialogue flow.

Large language model-driven SDSs

More recently, researchers have attempted to use LLMs in addition to or in place of rule-based systems to power SDSs. LLMs are driven by large databases, including a large amount of texts, available on the internet to respond to human language. For instance, Karatay and Xu (2025)Karatay, Y., & Xu, J. (2025) Exploring the potential of conversational AI for assessing second language oral proficiency. TESOL Quarterly (first view). Google Scholar logo with link to Google ScholarKaratay, Y., & Xu, J. (2025) Exploring the potential of conversational AI for assessing second language oral proficiency. TESOL Quarterly (first view). Google Scholar logo with link to Google Scholar used GPT 4o (Open-AI, n.d.) to power their SDS. Their aim was to use an LLM-driven SDS to replace a human in the IELTS Speaking Test, Part 3. This part of the test, “Two-way discussion,” allows a test taker to demonstrate their ability to discuss a complex topic based on previous test input, including answering questions, expanding on their views, and providing justifications for them; these features of interactional competence have been difficult to assess with computer-delivered oral assessments. The researchers recorded and transcribed the discussion between the SDS and 30 adult learners of English as a second language, from a variety of first language backgrounds. The discourse analysis indicated that both the test taker and the SDS had opportunities to demonstrate topic development, turn-taking, listener support, interruption, and repair and that more proficient test takers were able to demonstrate these abilities more effectively than less able test takers. Both test takers and trained raters believed that the SDS was a “competent and reliable interlocutor” (p. 25). However, similar to the rule-based system that Ockey and Chukharev-Hudilainen (2021)Ockey, G. J., & Chukharev-Hudilainen, E. (2021) Human versus computer partner in the paired oral discussion test. Applied Linguistics, 42 (5), 924–944. Google Scholar logo with link to Google ScholarOckey, G. J., & Chukharev-Hudilainen, E. (2021) Human versus computer partner in the paired oral discussion test. Applied Linguistics, 42(5), 924–944. Google Scholar logo with link to Google Scholar and Ockey et al. (2023)Ockey, G. J., Chukharev-Hudilainen, E., & Hirch, R. (2023) Assessing interactional competence: Interactional Competence Elicitor (ICE) vs a human partner. Language Assessment Quarterly, 20 (4–5), 377–398. Google Scholar logo with link to Google ScholarOckey, G. J., Chukharev-Hudilainen, E., & Hirch, R. (2023) Assessing interactional competence: Interactional Competence Elicitor (ICE) vs a human partner. Language Assessment Quarterly, 20(4–5), 377–398. Google Scholar logo with link to Google Scholar used, both test takers and raters pointed out that the SDS sometimes interrupted inappropriately, and a lack of verbal cues (e.g., no eye contact or body language such as nodding) led to unnatural interaction.

An LLM-powered SDS system was also used by Na (2025)Na, I. (2025, April). GenAI in assessing interactional competence: Examining differences between human-human and human-ai interactions. Language Assessment Research Conference. Fullerton, California, USA.Na, I. (2025, April). GenAI in assessing interactional competence: Examining differences between human-human and human-ai interactions. Language Assessment Research Conference. Fullerton, California, USA. to assess interactional competence, which she referred to as GICE. She used GPT-4o to serve as an artificial intelligence examiner and Claude 3.5 Sonnet as an artificial intelligence speech partner. Her system was able to give test takers opportunities to demonstrate their abilities for six of the 12 interactional competence features that Ockey and Chukharev-Hudilainen (2021)Ockey, G. J., & Chukharev-Hudilainen, E. (2021) Human versus computer partner in the paired oral discussion test. Applied Linguistics, 42 (5), 924–944. Google Scholar logo with link to Google ScholarOckey, G. J., & Chukharev-Hudilainen, E. (2021) Human versus computer partner in the paired oral discussion test. Applied Linguistics, 42(5), 924–944. Google Scholar logo with link to Google Scholar and Ockey et al. (2023)Ockey, G. J., Chukharev-Hudilainen, E., & Hirch, R. (2023) Assessing interactional competence: Interactional Competence Elicitor (ICE) vs a human partner. Language Assessment Quarterly, 20 (4–5), 377–398. Google Scholar logo with link to Google ScholarOckey, G. J., Chukharev-Hudilainen, E., & Hirch, R. (2023) Assessing interactional competence: Interactional Competence Elicitor (ICE) vs a human partner. Language Assessment Quarterly, 20(4–5), 377–398. Google Scholar logo with link to Google Scholar explored. These features included responding to clarification requests, initiating topics, developing topics, connecting topics, replying in an appropriate amount of time, and recognizing a position. Critical to their success in enabling test takers to demonstrate their knowledge of these features was the use of several breakdown strategies that they programmed into the system. For example, the system was programmed to include nonsense words which would require the test taker to demonstrate the ability to ask for clarification. While the LLM-driven SDS had the potential to elicit eight of the 12 interactional competence features it targeted, Na (2025)Na, I. (2025, April). GenAI in assessing interactional competence: Examining differences between human-human and human-ai interactions. Language Assessment Research Conference. Fullerton, California, USA.Na, I. (2025, April). GenAI in assessing interactional competence: Examining differences between human-human and human-ai interactions. Language Assessment Research Conference. Fullerton, California, USA. lamented that the system tended to help the test takers too much by repairing communication breakdowns and filling in information for the test taker. This made it difficult to assess how much the test taker could do. In essence, the LLM-driven SDS made it difficult to judge the test takers’ abilities because it provided too much support to them in the conversation. Interestingly, this is the same challenge that other researchers have noted when assessing a test taker with a human interlocutor and what Mehan (1979)Mehan, H. (1979) Learning lessons: Social organization in the classroom. Harvard UP. Google Scholar logo with link to Google ScholarMehan, H. (1979) Learning lessons: Social organization in the classroom. Harvard UP. Google Scholar logo with link to Google Scholar noted decades ago when he created two alternative definitions for the concept of interactional competence.

Other researchers have embedded LLM-powered SDS studies in a dynamic assessment (DA) framework. The aims of DA include to determine the degree to which learners can identify and correct their errors, and the type and amount of support learners need to complete an assessment. DA involves using prompts, suggestive questions, feedback, and examples to support learners when they complete language use activities (Poehner & Lantolf, 2023Poehner, M., & Lantolf, J. (2023) Advancing L2 Dynamic Assessment: Innovations in Chinese contexts. Language Assessment Quarterly, 20 (1), 1–19. Google Scholar logo with link to Google ScholarPoehner, M., & Lantolf, J. (2023) Advancing L2 Dynamic Assessment: Innovations in Chinese contexts. Language Assessment Quarterly, 20(1), 1–19. Google Scholar logo with link to Google Scholar), such as talking with a teacher about a missed homework assignment. This approach makes it possible to provide oral communication assessment learning and assessment activities that are targeted at the level of test takers; it can provide a fairly precise estimate of their oral abilities, while helping them to learn.

Kim (2025)Kim, G. (2025) Developing a GPT-powered dynamic pragmatics assessment using the expanded evidence centered design. Language Assessment Research Conference. Fullerton, California, USA.Kim, G. (2025) Developing a GPT-powered dynamic pragmatics assessment using the expanded evidence centered design. Language Assessment Research Conference. Fullerton, California, USA. examined the effectiveness of a role-play task between a test taker and an LLM-driven SDS. Using a DA framework, his specific aims were to determine test takers’ language ability, how much support they needed to complete the task, and how much they learned when completing the task. In the role-play scenario, an LLM (ChatGPT) took the role of a new middle school student, and the test taker played the part of a more senior middle school student. The SDS began by describing some of its role-play character’s interests, and by asking for a suggestion on what school club would be appropriate to join. If the test taker was unable to make a suggestion, the SDS would provide a series of hints to help them. It began with a phrase such as, “Excuse me. I don’t understand. Could you suggest a school club” (Clarification request phase). If the test taker could not respond appropriately, the SDS would say something like “Please use (insert appropriate supporting language)” (Elicitation phase). If the test taker was still unable to provide appropriate advice for which club to join, the SDS would provide explicit correction by saying, “Did you want to say (insert appropriate phrasing).” The discussion would continue with the SDS asking for a different recommendation. Grading was based on the effectiveness of the test taker’s response, how much support was needed, and how many recommendations were required for the test taker to successfully complete the task. Kim’s research underscores the value of using a machine to aid in learning and assessment. The machine can provide repeated opportunities, along with carefully graded feedback at the learner’s level, to facilitate the assessment and learning process.

Overall, both rule-based and LLM-driven SDSs have strong potential for assisting in the delivery of effective oral communication assessments. They have been successful in eliciting various interactional competence features, and can also be used to help connect assessment to learning with approaches, such as DA.

Conclusions and future research

This paper has discussed the use of technology to aid in assessing oral communication. The particular focus has been on using a carefully considered oral communication construct to guide assessment design and delivery practices, with particular focus on assessing interactional features of oral assessment which have proven more challenging than assessing pronunciation, grammar, vocabulary, and fluency. The paper concentrated on OC assessments via video-mediated contexts, virtual environments, and SDSs where a test taker talks with a computer chatbot.

The research generally indicates that video-mediated approaches, virtual environments, and SDSs can be effective delivery models for OC assessment. While substantial research has explored video-mediated platforms and SDSs in assessing oral communication, studies on VR remain comparatively scarce. As discussed previously, conducting a needs analysis to examine its use in real-life contexts among younger generations is recommended, and further investigation into its role in assessing oral communication is encouraged.

The choice of which technology to use and how to use it should depend primarily on the aim of the test and the construct being tested. Importantly, stakeholders in a particular assessment context must come to agreement about the particular construct that they desire to measure. For example, when deciding whether to include or exclude visual appearance of test takers and/or interlocutors to each other and/or the performance judges in the exam, it is important to consider that in many real-world contexts, visual information is present and helps people communicate orally. However, in practice, OC assessments have not explicitly included non-verbal communication in the scoring process. This may be due to the cultural differences associated with non-verbal communication, and the perceptions of unfairness when judging test takers based on their cultural backgrounds. Moreover, as Aseeri (2025)Aseeri, F. M. (2025) Judging a book by its cover: Investigating the influence of test takers’ physical attractiveness in synchronous oral communication assessment. Unpublished doctoral dissertation, Iowa State University.Aseeri, F. M. (2025) Judging a book by its cover: Investigating the influence of test takers’ physical attractiveness in synchronous oral communication assessment. Unpublished doctoral dissertation, Iowa State University. found, visual features such as physical appearance may impact scores inappropriately which makes it difficult to justify the use of visual information in an oral assessment context, at least from the judges vantage point. Such a finding suggests that at least in high-stakes exams test developers should consider solutions such as the use of avatars or other audio-only presentation of test takers’ oral performances to minimize this potential effect.

In addition, choosing whether a computer partner or a human partner should take part also depends on the construct of the OC exam. Ockey et al. (2023)Ockey, G. J., Chukharev-Hudilainen, E., & Hirch, R. (2023) Assessing interactional competence: Interactional Competence Elicitor (ICE) vs a human partner. Language Assessment Quarterly, 20 (4–5), 377–398. Google Scholar logo with link to Google ScholarOckey, G. J., Chukharev-Hudilainen, E., & Hirch, R. (2023) Assessing interactional competence: Interactional Competence Elicitor (ICE) vs a human partner. Language Assessment Quarterly, 20(4–5), 377–398. Google Scholar logo with link to Google Scholar argue that an SDS may be preferable when interactional competence is defined as an individual’s “competence necessary for effective interaction” (Mehan, 1979Mehan, H. (1979) Learning lessons: Social organization in the classroom. Harvard UP. Google Scholar logo with link to Google ScholarMehan, H. (1979) Learning lessons: Social organization in the classroom. Harvard UP. Google Scholar logo with link to Google Scholar, p. 130). On the other hand, a human partner may be preferred when interactional competence is defined as “the competence that is available in the interaction between participants” (Mehan, 1979Mehan, H. (1979) Learning lessons: Social organization in the classroom. Harvard UP. Google Scholar logo with link to Google ScholarMehan, H. (1979) Learning lessons: Social organization in the classroom. Harvard UP. Google Scholar logo with link to Google Scholar, p. 130). With regard to Dai’s (2024)Dai, D. W. (2024) Assessing interactional competence: Principles, test development and validation through an L2 Chinese IC test (p. 446). Peter Lang International Academic Publishers. Google Scholar logo with link to Google ScholarDai, D. W. (2024) Assessing interactional competence: Principles, test development and validation through an L2 Chinese IC test (p. 446). Peter Lang International Academic Publishers. Google Scholar logo with link to Google Scholar conceptualization of interactional competence to include emotional, logical, moral, and categorial dimensions, further research is needed to identify whether the inclusion of such features is feasible when using technology and if so if it makes the interaction similar to that of F2F interaction. If not, how is it different and how might it impact the construct of oral communication if it is used.

With the increasing access to LLMs, such as ChatGPT, CoPilot, Gemini, and others freely available on the internet in many parts of the world, language assessment researchers are developing their own LLM-driven SDSs for assessing oral communication. This is exciting because it increases the likelihood of oral communication assessments that include interactional features. As previously discussed, such systems do tend to help test takers when they are unable to effectively repair a conversation break-down or the test taker is struggling to communicate. A potential solution to this challenge is to use a hybrid SDS which is composed of both a rule-based system and an LLM-driven system. Hybrid systems that use a rule-based system to provide test takers with opportunities to demonstrate their abilities to repair meaning, interactively listen, etc. combined with LLM-driven systems that make the conversation more natural are likely to gain popularity in the coming years. Such systems may be programmed to alternate between two LLM-driven turns and one rule-based turn.

Another interesting area for research with LLM-driven systems relates to the use of diverse speech varieties. When an LLM-driven SDS is used, it could be programmed to use various types of speech with varying strengths of accent. It is increasingly believed that the ability to accommodate to an unknown speech variety is an important part of the OC construct. However, finding speech varieties that are appropriately different from ones that are familiar to test takers depends on many factors, including the test takers’ background and personal characteristics. Using LLMs to create various speech varieties with different accent strengths may make it possible to assess this ability to accommodate, which has been extremely difficult to measure in the past (Ockey & Qi, in pressOckey, G. J. & Qi, Q. in press). Why multiple speech varieties when assessing L2 Chinese and English listening. In Wagner, E., & In’nami, Y. Eds. The Theory and Practice of Second Language Listening Assessment Ockey, G. J. & Qi, Q. in press). Why multiple speech varieties when assessing L2 Chinese and English listening. In Wagner, E., & In’nami, Y. Eds. The Theory and Practice of Second Language Listening Assessment ).

In conclusion, advances in technology have made new approaches to assessing oral communication possible. Its use has also introduced new challenges for language assessment stakeholders, prompting the need for better language assessment literacy and decision making for which tool to use for a particular context. It is critical that test developers and other stakeholders carefully consider what they desire to measure and then take into account the strengths and weaknesses of the technology (or non-technology) that they decide to use when they create or select an assessment.

Funding

Open Access publication of this article was funded through a Transformative Agreement with Iowa State University.

References

Aseeri, F. M.
(2025) Judging a book by its cover: Investigating the influence of test takers’ physical attractiveness in synchronous oral communication assessment. Unpublished doctoral dissertation, Iowa State University.
① ② ③ ④
Brooks, L.
(2009) Interacting in pairs in a test of oral proficiency: Co-constructing a better performance. Language Testing, 26 , 342–366. Google Scholar logo with link to Google Scholar
① ②
Buunk, A. P., & Dijkstra, P.
(2011) Does attractiveness sell? Women’s attitude toward a product as a function of model attractiveness, gender priming, and social comparison orientation. Psychology & Marketing, 28 ( 9 ), 958–973. Google Scholar logo with link to Google Scholar
①
Chukharev-Hudilainen, E., & Ockey, G. J.
(2021) The development and evaluation of interactional competence elicitor for oral language assessments. ETS Research Report Series, 2021 (1), 1–20. Google Scholar logo with link to Google Scholar
①
Dai, D. W.
(2024) Assessing interactional competence: Principles, test development and validation through an L2 Chinese IC test (p. 446). Peter Lang International Academic Publishers. Google Scholar logo with link to Google Scholar
① ②
Dombi, J., Sydorenko, T., & Timpe-Laughlin, V.
(2024) Openings and closings in human-human versus human-spoken dialogue system conversations. Language Learning & Technology, 28(2), 32–61. https://​hdl​.handle​.net​/10125​/73571.
①
Ducasse, A. M., & Brown, A.
(2009) Assessing paired orals: Raters’ orientation to interaction. Language Testing, 26 , 423–443. Google Scholar logo with link to Google Scholar
① ②
Galaczi, E., & Taylor, L.
(2018) Interactional competence: Conceptualisations, operationalisations, and outstanding questions. Language Assessment Quarterly, 15 (3), 219–236. Google Scholar logo with link to Google Scholar
① ②
Gukturk, N., & Chukharev, E.
(2024) Exploring the potential of a spoken dialog system-delivered paired discussion task for assessing interactional competence. Language Assessment Quarterly, 21 (1), 60–99. Google Scholar logo with link to Google Scholar
① ②
Hosoda, M., Stone-Romero, E. F., & Coats, G.
(2003) The effects of physical attractiveness on job-related outcomes: A meta-analysis of experimental studies. Personnel Psychology, 56 ( 2 ), 431–462. Google Scholar logo with link to Google Scholar
①
Jamieson, J., Eignor, D. R., Grabe, W. and Kunnan, A. J.
(2008) “Frameworks for a new TOEFL”. In C. A. Chapelle, J. Jamieson, and M. K. Enright (Eds.), Building a validity argument for the test of English as a foreign language (pp. 55–95). Routledge.Google Scholar logo with link to Google Scholar
①
Jenkins, J.
(2016) International tests of English: Are they fit for purpose? In H. Liao (Ed.), Critical reflections on foreign language education: Globalization and local interventions. The Language Training and Testing Center.Google Scholar logo with link to Google Scholar
①
Karatay, Y.
(2022) Development and validation of spoken dialog system-based oral communication tasks in an ESP context (publication No. 29165842). Doctoral dissertation, Iowa State University ProQuest Dissertations Publishing.
①
Karatay, Y., & Xu, J.
(2025) Exploring the potential of conversational AI for assessing second language oral proficiency. TESOL Quarterly (first view). Google Scholar logo with link to Google Scholar
①
Kim, G.
(2025) Developing a GPT-powered dynamic pragmatics assessment using the expanded evidence centered design. Language Assessment Research Conference. Fullerton, California, USA.
①
Kim, J., & Craig, D. A.
(2012) Validation of a videoconferenced speaking test. Computer Assisted Language Learning, 25(3), 257–275. Google Scholar logo with link to Google Scholar
① ②
Kurt, S.
(2025) Development and evaluation of a virtual reality environment for the Oral English Certification Test. Unpublished doctoral dissertation, Iowa State University.
①
Lee, H. W., Mullooly, A., Devine, A., & Galaczi, E.
(2024) Exploring interaction in video-call paired speaking tests: A look at scores, language, and perceptions. Applied Linguistics, 45 (3), 533–552. Google Scholar logo with link to Google Scholar
① ② ③
Levis, J.
(2020) Revisiting the intelligibility and nativeness principles. Journal of Second Language Pronunciation, 6 (3), 310–328. Google Scholar logo with link to Google Scholar
①
McNamara, T.
(2018) A challenge for language testing: The assessment of English as a Lingua Franca. Language Assessment Research Conference, Ames, Iowa, March 21–23.
①
Mehan, H.
(1979) Learning lessons: Social organization in the classroom. Harvard UP. Google Scholar logo with link to Google Scholar
① ② ③ ④
Muhanna, M. A.
(2015) Virtual reality and the CAVE: Taxonomy, interaction challenges and research directions. Journal of King Saud University-Computer and Information Sciences, 27 (3), 344–361. Google Scholar logo with link to Google Scholar
①
Na, I.
(2025, April). GenAI in assessing interactional competence: Examining differences between human-human and human-ai interactions. Language Assessment Research Conference. Fullerton, California, USA.
① ② ③ ④
Nakatsuhara, F., Inoue, C., Berry, V., & Galaczi, E.
(2017) Exploring the use of video-conferencing technology in the assessment of spoken language: A mixed-methods study. Language Assessment Quarterly, 14 (1), 1–18. Google Scholar logo with link to Google Scholar
① ② ③
Neiriz, R.
(2023) Eliciting interactive oral communication samples through a spoken dialogue system to measure interactional competence. Unpublished doctoral dissertation, Iowa State University.
① ② ③ ④
Ockey, G. J.
(2009) The effects of a test taker’s group members’ personalities on the test taker’s second language group oral discussion test scores. Language Testing, 26 ( 2 ), 161–186. Google Scholar logo with link to Google Scholar
①
(2014) The potential of the L2 group oral to elicit discourse with a mutual contingency pattern and afford equal speaking rights in an ESP context. English for Specific Purposes, 35 , 17–29. Google Scholar logo with link to Google Scholar
①
(2025) Introducing second language assessment. Cambridge University Press.Google Scholar logo with link to Google Scholar
①
Ockey, G. J., & Chukharev-Hudilainen, E.
(2021) Human versus computer partner in the paired oral discussion test. Applied Linguistics, 42 (5), 924–944. Google Scholar logo with link to Google Scholar
① ② ③ ④ ⑤ ⑥
Ockey, G. J., Chukharev-Hudilainen, E., & Hirch, R.
(2023) Assessing interactional competence: Interactional Competence Elicitor (ICE) vs a human partner. Language Assessment Quarterly, 20 (4–5), 377–398. Google Scholar logo with link to Google Scholar
① ② ③ ④ ⑤ ⑥ ⑦
Ockey, G. J., Gu, L., & Keehner, M.
(2017) Web-based virtual environments for facilitating assessment of L2 Oral communication ability. Language Assessment Quarterly, 14 (4), 346–359. Google Scholar logo with link to Google Scholar
① ② ③ ④
Ockey, G. J., & Hirch, R. R.
(2020) A step toward the assessment of English as a lingua franca. In G. J. Ockey & B. A. Green (Eds.), Another generation of fundamental considerations in language testing (pp. 9–28). Springer. Google Scholar logo with link to Google Scholar
①
Ockey, G. J., Koyama, D., Setoguchi, E., & Sun, A.
(2015) The extent to which TOEFL iBT speaking scores are associated with performance on oral language tasks and oral ability components for Japanese university students. Language Testing, 32 (1), 39–62. Google Scholar logo with link to Google Scholar
①
Ockey, G. J., & Li, Z.
(2015) New and not so new methods for assessing oral communication. Language Value, 7 (1), 1–21. Google Scholar logo with link to Google Scholar
①
Ockey, G. J., & Neiriz, R.
(2021) Evaluating technology-mediated second language oral communication assessment delivery models. Assessment in Education: Principles, Policy & Practice, 28 (4), 350–368. Google Scholar logo with link to Google Scholar
①
Ockey, G. J., Timpe-Laughlin, V., Davis, L., & Gu, L.
(2019) Exploring the potential of a video-mediated interactive speaking assessment (Report No. ETS RR-19-05). Princeton, NJ: Educational Testing Service. Google Scholar logo with link to Google Scholar
①
Ockey, G. J. & Qi, Q.
in press). Why multiple speech varieties when assessing L2 Chinese and English listening. In Wagner, E., & In’nami, Y. Eds. The Theory and Practice of Second Language Listening Assessment
①
Poehner, M., & Lantolf, J.
(2023) Advancing L2 Dynamic Assessment: Innovations in Chinese contexts. Language Assessment Quarterly, 20 (1), 1–19. Google Scholar logo with link to Google Scholar
①
Ramanarayanan, V., Evanini, K., & Tsuprun, E.
(2019) Beyond monologues. In Zechner, K., & Evanini, K. (Eds.). Automated speaking assessment: Using language technologies to score spontaneous speech (pp. 176–191). Routledge. Google Scholar logo with link to Google Scholar
① ②
Song, J., & Hsu, W. L.
(2021) Design and implementation of a classroom-based virtual reality assessment. In Salaberry, M. & Burch, A. (Eds.), Assessing speaking in context: Expanding the construct and its applications (pp. 265–282). Channel View Publications. Google Scholar logo with link to Google Scholar
①
Vo, S.
(2019) Effects of task types on interactional competence in oral communication assessment (Publication No. 17345). Doctoral dissertation, Iowa State University. ProQuest Dissertation and Theses Global.
① ②
Zhang, L., & Jin, Y.
(2021) Assessing interactional competence in the computer-based CET-SET: An investigation of the use of communication strategies. Assessment in education: Principles, policy & practice, 28 (4), 389–410. Google Scholar logo with link to Google Scholar
① ② ③

Address for correspondence

Gary J. Ockey

Iowa State University

United States

[email protected]

Biographical notes

Gary J. Ockey is a professor in the Applied Linguistics and Technology Program at Iowa State University in the United States. His research interests evolve around the assessment of second language oral communication. His most recent book (2025) is, “Introducing second language assessment”, Cambridge University Press.

Fatimah M. Aseeri is an Associate Professor in the English Department at Najran University, Saudi Arabia. Her research interests focus on oral communication assessment. Her doctoral dissertation investigated the impact of physical attractiveness on oral communication assessment, contributing to research on the role of non-linguistic factors in language assessment and evaluation.

 
Authentication
Mobile Menu Logo with link to supplementary files background Layer 1 jbe-platform.quadraat-sans-bold jbe-platform.quadraat-sans-bold-dark jbe-platform.quadraat-sans-bold-light jbe-platform.quadraat-sans-bold prag Twitter_Logo_Blue