This thesis aims to analyze the implications of the advent of Artificial Intelligence in the educational field, with a primary focus on Japanese language learning. In the current context, where the rapid spread of Large Language Models (LLMs) is redefining educational and language teaching paradigms, this dissertation highlights the need for methodological adaptation, while also focusing on experimentation with the Japanese language. Academic literature, thus far, has indeed favored the study of AI applied to L2/FL English over other languages. Specifically, this research aims to investigate the perception and readiness for change among students at Ca' Foscari University of Venice regarding the use of latest-generation technologies, such as the ChatGPT-5 model (released in September 2025), equipped with advanced speech synthesis capabilities. The objective is to understand whether and how an informed implementation of these tools, integrated through Prompt Engineering and supported by a language teaching approach such as Task-Based Language Learning (TBLL), can foster personalized learning and mitigate negative affective factors—such as classroom anxiety (FLCAS)—which risk reducing motivation and hindering progress during language learning. The study is based on the premise, supported by Self-Determination Theory, that AI cannot replace the human component and the need for relatedness (one of the three basic psychological needs according to this theory), but rather must be guided by an expert figure to be effective. The work is divided into four chapters: the first chapter provides a historical overview of the relationship between artificial intelligence and language pedagogy, examining the advantages and critical issues of current models in language learning, while also addressing ethical considerations on the subject. The second chapter describes the theoretical integration between Task-Based Language Learning and Prompt Engineering for formulating effective instructions. The third chapter outlines the methodology and research results based on a sample of 30 students from Ca' Foscari University of Venice, who were involved in an experiment combining two questionnaires and an oral interaction task with ChatGPT-5. Finally, the fourth chapter draws conclusions regarding the potential of these tools in the current academic context, taking into account the results obtained through the experiment.
La presente tesi si propone di analizzare le implicazioni dell’avvento dell’Intelligenza Artificiale in ambito educativo, con un’attenzione rivolta principalmente all’apprendimento della lingua giapponese. Nel contesto attuale in cui la rapida diffusione di Large Language Models (LLMs) sta ridefinendo i paradigmi didattici e glottodidattici, l’elaborato sottolinea la necessità di un adattamento metodologico, cercando altresì di focalizzarsi su una sperimentazione con la lingua giapponese. La letteratura accademica, finora, ha infatti privilegiato lo studio dell’IA applicata alla lingua inglese L2/LS rispetto ad altre lingue. In particolare, la ricerca mira a indagare la percezione e la disponibilità al cambiamento degli studenti dell’Università Ca’ Foscari Venezia di fronte all’utilizzo di tecnologie di ultima generazione, come il modello ChatGPT-5 (rilasciato a settembre 2025), dotato di avanzate capacità di sintesi vocale. L’obiettivo è comprendere se e come un’implementazione consapevole di questi strumenti, integrata attraverso il Prompt Engineering e supportata da un approccio glottodidattico come il Task-Based Language Learning (TBLL), possa favorire un apprendimento personalizzato e mitigare fattori affettivi negativi come l’ansia in aula (FLCAS) che rischiano di ridurre la motivazione e quindi di intralciare il progresso durante l’apprendimento linguistico. Lo studio si fonda sul presupposto, supportato dalla SelfDetermination Theory, che l’IA non possa sostituire la componente umana e la necessità di relatedness (o senso di appartenenza, ovvero uno dei tre bisogni psicologici di base secondo questa teoria), ma debba essere guidata da una figura esperta per essere efficace. Il lavoro si articola in quattro capitoli: il primo capitolo fornirà una panoramica storica del rapporto tra intelligenza artificiale e glottodidattica, esaminando vantaggi e criticità dei modelli attuali nell’apprendimento linguistico, con uno sguardo rivolto anche a questioni etiche sull’argomento. Il secondo capitolo descriverà l’integrazione teorica tra il Task-Based Language Learning e il Prompt Engineering per la formulazione di istruzioni efficaci. Il terzo capitolo illustrerà la metodologia e i risultati della ricerca su un campione di 30 studenti dell’Università Ca’ Foscari Venezia coinvolti in un esperimento che combina due questionari e un task di interazione orale con ChatGPT-5. Infine, nel quarto capitolo verranno tracciate le conclusioni relative alle potenzialità di questi strumenti nel contesto accademico attuale tenendo conto dei risultati ottenuti tramite l’esperimento.
Didattica della lingua giapponese e Intelligenza Artificiale: analisi di un caso studio sull'uso di ChatGPT per la produzione orale
D'ALOISIO, FRANCESCO
2024/2025
Abstract
This thesis aims to analyze the implications of the advent of Artificial Intelligence in the educational field, with a primary focus on Japanese language learning. In the current context, where the rapid spread of Large Language Models (LLMs) is redefining educational and language teaching paradigms, this dissertation highlights the need for methodological adaptation, while also focusing on experimentation with the Japanese language. Academic literature, thus far, has indeed favored the study of AI applied to L2/FL English over other languages. Specifically, this research aims to investigate the perception and readiness for change among students at Ca' Foscari University of Venice regarding the use of latest-generation technologies, such as the ChatGPT-5 model (released in September 2025), equipped with advanced speech synthesis capabilities. The objective is to understand whether and how an informed implementation of these tools, integrated through Prompt Engineering and supported by a language teaching approach such as Task-Based Language Learning (TBLL), can foster personalized learning and mitigate negative affective factors—such as classroom anxiety (FLCAS)—which risk reducing motivation and hindering progress during language learning. The study is based on the premise, supported by Self-Determination Theory, that AI cannot replace the human component and the need for relatedness (one of the three basic psychological needs according to this theory), but rather must be guided by an expert figure to be effective. The work is divided into four chapters: the first chapter provides a historical overview of the relationship between artificial intelligence and language pedagogy, examining the advantages and critical issues of current models in language learning, while also addressing ethical considerations on the subject. The second chapter describes the theoretical integration between Task-Based Language Learning and Prompt Engineering for formulating effective instructions. The third chapter outlines the methodology and research results based on a sample of 30 students from Ca' Foscari University of Venice, who were involved in an experiment combining two questionnaires and an oral interaction task with ChatGPT-5. Finally, the fourth chapter draws conclusions regarding the potential of these tools in the current academic context, taking into account the results obtained through the experiment.| File | Dimensione | Formato | |
|---|---|---|---|
|
Tesi_D'Aloisio.pdf
accesso aperto
Dimensione
2.23 MB
Formato
Adobe PDF
|
2.23 MB | Adobe PDF | Visualizza/Apri |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14247/27265