Análisis de los diálogos con ChatGPT del profesorado en formación durante el diseño de actividades didácticas de Química

This study analyzes changes in the structure and content of dialogues between pre-service primary teachers and ChatGPT during the design of chemistry teaching activities. A total of 223 prompts from 28 conversations (14 PRE and 14 POS) carried out by third-year students were examined, revealing a progressive enrichment in scientific and pedagogical content, as well as increased sophistication in the interactions. These included more complex questions, appropriate use of technical terminology, and consideration of students’ prior conceptions. Qualitative transformations in the conversational dynamics were also identified, characterized by greater complexity, pedagogical intentionality, and the emergence of expansive prompts, absent in the initial phase and prominent in the final one. Prompts were categorized according to their thematic content (scientific, pedagogical, contextual, and practical) and communicative function within the dialogue (addition, expansion, modulation, or refutation), allowing for a comparative analysis between phases. The study adopted a mixed-methods approach, combining descriptive statistics and qualitative content analysis. Findings suggest that structured activities designed to support pedagogical uses of ChatGPT can enhance pre-service teachers’ competencies related to pedagogical content knowledge (PCK) and foster a critical and reflective approach to the use of generative artificial intelligence technologies in science education.