The large multimodal model GPT-4 with the chatbot ChatGPT can operate with text, images etc. With the further aim of understanding how to employ GPT-4 to design a classroom activity, we want to investigate if ChatGPT's multimodal functionalities allow it to effectively handle different (semiotic) representations while solving an assigned mathematical task with operations such as treatment and conversion of data, and provide potentially useful new representations for students. We conducted a qualitative test in which two tasks containing graphs were submitted to GPT-4, and we analysed its responses with a focus on used and produced representations.
Chat GPT as a didactic tool to improve students' abilities in managing different semiotic representations
Giorgia Nieddu;Antioco Luigi Zurru
2025-01-01
Abstract
The large multimodal model GPT-4 with the chatbot ChatGPT can operate with text, images etc. With the further aim of understanding how to employ GPT-4 to design a classroom activity, we want to investigate if ChatGPT's multimodal functionalities allow it to effectively handle different (semiotic) representations while solving an assigned mathematical task with operations such as treatment and conversion of data, and provide potentially useful new representations for students. We conducted a qualitative test in which two tasks containing graphs were submitted to GPT-4, and we analysed its responses with a focus on used and produced representations.| File | Dimensione | Formato | |
|---|---|---|---|
|
GiorgiaNieddu_TWG16.pdf
accesso aperto
Licenza:
DRM non definito
Dimensione
489.65 kB
Formato
Adobe PDF
|
489.65 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


