Dmitri Leontjev

DD-Lang team's paper in Kielikukko - the Journal of the Finnish Reading Association

This was such a productive year for DD-LANG, with numerous meetings with #teachers and #learners, many #conference #presentations, and notable developments incorporating #AI into DD-LANG.

We also wrote a paper with teachers for the #Kielikukko journal.

Thank you to everyone who made this happen!

Simplified PRISMA chart (adapted from Page et al., 2021)

Stefan O‘Grady, Nazlınur Göktürk & Olena Rossi presented a systematic review of AI use in language testing between 2019 and 2024 at #EALTA2025

Ari Huhta, Sylvia Opriessnig, Josep Maria Montcada Escbairo, Doris MoserFrötscher, Olivia Rütti Joy, Gyula Tánko, & Karin Vogt
Gave a panel discussion at #EALTA2025 focusing on assessment impact
- who and how much have agency at the steering wheel?
- can we make it into a discussion?

We discussed how learner-chatbot interactions created opportunities for students to gain further understanding of the texts and the tasks and provide extra information about students' performance and areas of struggle.

Matt Poehner, Ari Huhta, and I presented a part of the DD-Lang project's work focusing on AI as a part of mediational sequences.

Recommendations for Assessment Literacy courses • Course content • Glossary with a related language-focussed task • (Simple) quantitative and qualitative test analysis • Facilitating assessment design and marking • Course delivery • Practical test development activities & peer feedback • Topics seen from different viewpoints to raise awareness of function of assessment • Carefully staged content • Contextualisation • Frame within a particular context • Emphasise learner-controd aspects

Carolyn Westbrook, Richard Spiby & Jordan Weide on corpus linguistics-based approach to investigate teachers’ challenges in language assessment at #EALTA2025

Breakthroughs: Al & LLM Application to language testing: Scoring essays for a high-stakes test Scoring accuracy: A fine-tuned BERT model was most accurate Detecting anomalous responses: Prompt engineering with an LLM was best Providing qualitative feedback: Prompt engineering with an LLM was most informative Solution: One Al agent retrieves the scores from the fine-tuned BERT model Another Al agent checks for anomalous essays and flags them Another Al agent provides qualitative feedback by prompting with an LLM

Jamie Dunlea Lecture #EALTA2025: Alistair Van Moere's giving a keynote on AI innovation transforming language assessment

Title slide of the presentation "I think the most interesting thing iscthat this would help them"

Very interesting CBLA and MLA SIG today at #EALTA2025 with a large variety of foci, contexts, and methodologies. Vera Vesala and I presented our paper on a part of @ddlangproject.bsky.social project, focusing on our collaboration with teachers, namely, on how we together design classroom activities

Thu, May 28, Matt Poehner, @dima403.bsky.social, and Ari Huhta present a paper discussing #GenAI, namely #chatbot interactions as part of #mediation #sequences for #assessing and promoting #learners' L2 #English #reading abilities

Wed, May 28, Vera Vesala and @dima403.bsky.social give a paper “I think the most interesting thing is that this would help them” focusing on #praxis with DD-Lang #teacher collaboratorsat Classroom-Based Assessment & Multilingual Language Assessment SIG meeting

DD-Lang team is travelling to #EALTA2025, Saltzurg. We will be presenting two papers focusing on different aspects of the project: ⏭️

a view of city centre, Tampere, Tammerkoski river, and red-brick industrial buildings.

DD-Lang team has been out and about again, collecting data. This time in #Tampere.
We owe this all to our wonderful #teacher #collaborators and enthusiastic #students 🙌
Kiitos kaikille for making this happen!

Theoretical framework of the study Bronfenbrenner and Farrel

Maria Matilde Olivero presented a paper on #self-exploration of an #LTE's emotions at #AAAL2025
#Self-inquiry into own #emotions as key to understanding own #identity as a #peacebuilder in teacher education

Tu / usted SCOBA

Adam van Compernolle gave an excellent paper on developing #L2 #indexing #capacity at #AAAL2025
#Dialectics of #language and #culture and whst it means for learner development

Title slide of the presentation: CHILEAN EFL TEACHERS' ASSESSMENT LITERACY: BRIDGING THE GAP BETWEEN NATIONAL STANDARDS AND TEACHING PRACTICES

Evelyn Dlaz-lturriaga on Chilean Teachers' #assessment #literacy
Teacher autonomy
Accountability and
Support

Recommendations for applied linguistics authors 1)Know GenAl capacities and limitations in research practices 2) Evaluate and fact-check GenAl output - Disclose GenAl use 3) Search and review GenAl guidelines provided for authors - Contact journal editor for clarification whenever uncertainty exists - Adhere to journal's instructions and ensure responsible GenAl use

Shuhui Yin and Carol Chapelle present their study of academic journals' author #guidelines regarding the #ethical use of #GenAI at #AAAL2025

Title slide of the presentation: Dynamic Assessment of Integrated Argumentative Writing: A Comparative Case Study of Two Multilingual Writers' Development

Matt Poehner and Lu Yu are presenting a paper on #dynamic #assessment of #argumentative #writing, focusing on developmental trajectories of two multilingual learners at #AAAL2025

Title slide of the presentation: Teacher and researcher positionality in praxis: Adapting dynamic assessment to increase learner agency in classrooms

DD-Lang project's @dima403.bsky.social and Matt Poehner presented the paper:
#Teacher and #researcher
#positionality in #praxis:
Adapting dynamic #assessment to increase #learner #agency in classrooms
at #AAAL2025

Title slide of the plenary

Linguistic Racism: Origins 1. Sociology of nationalism (1960s onward) (Anderson, Bauman & Briggs, May, Mignolo) • Linked Intrinsically - although not always explicitly - to racialization, (internal and external) colonization, and modern nation-building (nation-state congruence) • Emphasized linguistic homogeneity via 'national' (invariably, European) languages-perceived as unifying and modern • Indigenous and other local languages were dismissed as primitive / limiting • A socially and politically constructed process of linguistic hierarchies of prestige and an ideology of contempt toward other languages, and a recent historical development • Achieved via legitimation and institutionalization (esp, via education) (text automatically recognised)

Stephen May giving a plenary on #linguistic #racism, #(post)colonialism, and #indigenous #languages at #AAAL2025

Slide from the presentation focusing on pedagogical approach

Lia Plakans, Kwangmin Lee, and Marina Durinova on developing Students' #critical language #assessment #literacy: focusing on #fairness and #justice at #AAAL2025

Screenshot from the presentation, focusing on how conversations with chatbot are not the same as talking with a human conversation partner

Erik Voss and Hansun Waring on exploring #naturalness of #AI conversational partners using #CA at #AAAL2025

General workflow for ChatGPT-enhsnced DA

Xiaozgeng Dai, Lu Yu, Zhingyuan Zhuang, Xiaofei Lu, and Matt Poehner
On potential of #ChatGPT to enhance #DA of #L2 #argumentative #writing at #AAAL2025

refusal as another R Refusal understands the wisdom in a story, as well as the wisdom in not passing that story on. Refusal in research makes way for other r-words- for resistance, reclaiming, recovery, reciprocity. repatriation, regeneration -Tuck and Yang 2014, p. 244

Brilliant plenary by Wesley Y. Leonard on relational accountability
A framework for language work and beyond
At #AAAL2025

Why Integrate Al with Mediation in Language Instruction • Al tools provide immediate, targeted feedback enhancing learner engagement. • Teacher mediation helps learners critically assess Al-generated feedback. • Integrating Al with mediation promotes deeper metalinguistic awareness • Encourages autonomous learning by guiding informed use of language tools Significance: • Combines strengths of human teachers and Al technology • Addresses Al limitations through human judgment. • Supports learner development within the Zone of Proximal Development (ZPD).

Yuwei Xia and Qi Xu on #AI #integration with #Vygotskian #mediation fostering the #development of #lexical #sophistication awareness at #AAAL2025

Slide: Best practices are situated practices

Catherine DeFelice-Box and Anne Pomerantz on listening to your #participants in #EMI #research at #AAAL2025
Best #practices are #situated practices.
EMI is different for different stakeholders in different situations.

Teachers and researchers in dialogue

An interesting special session on language #teachers, teacher #educators, and #researchers #collaboration and #networking at #AAAL2025

AAAL Denver 2025

DD-Lang project's @dima403.bsky.social is participating in #AAAL2025 conference in #Denver.
Matt Poehner and Dmitri will be presenting a part of the project, focusing on our praxis with teachers on March 24 at 10.45. If you are at AAAL, consider joining us

A view of Aurajoki in Turku

DD-LANG team had great two days in #Turku last week, meeting our #teacher #collaborators and presenting the project to several groups of #students. It was an excellent opportunity to develop our ideas together, including using #AI as #pedagogical tool in the #classroom!

In dialectical collaboration - #praxis - with #language #teachers, we bring together #L2 #diagnostic and #dynamic #assessment, #theory and #practice, #high-stakes assessment with assessment supporting #learning, and #computerised dynamic assessment with learning and #teaching in the classroom

Dynamic Diagnostic Language Assessment - Conceptual and Practical Innovation in Foreign Language Education and Assessment Dynamic and diagnostic assessment of foreign language

@ddlangproject.bsky.social
Dynamic Diagnostic Language Assessment - Conceptual and Practical Innovation in Foreign Language Education and Assessment is a project I have an honour to be a part of, giving me an opportunity to work with inspiring teachers and wonderful colleagues.
r.jyu.fi/DDLANG_en