Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
ChatGPT as a Cognitive Tool in Sociology: Students’ Perceptions and Its Impact on Academic Writing Practices
0
Zitationen
1
Autoren
2025
Jahr
Abstract
This study examines how undergraduate sociology students perceive and use ChatGPT as a cognitive tool in their academic writing, focusing on its influence on reasoning, writing strategies, and ethical judgment. Using a qualitative case study design, data were collected from 18 sociology students through in-depth semi-structured interviews and comparative analysis of writing samples produced with and without ChatGPT. Thematic analysis generated three key findings. First, students described ChatGPT as cognitive scaffolding that helped them clarify abstract sociological concepts, generate ideas, and articulate theoretical explanations more coherently. Many viewed the tool as a “thinking companion” that reduced conceptual barriers when engaging with complex theories. Second, ChatGPT contributed to shifts in writing strategies, with students moving from unstructured drafting to more systematic practices such as outlining, coherence checking, and iterative refinement. These changes were evident in writing samples, which demonstrated clearer argumentation and improved organization after AI-assisted drafting. Third, students negotiated a tension between assistance and dependence. While recognizing the tool’s benefits, they expressed concerns about losing their academic voice, weakening critical thinking, or relying too heavily on AI-generated structures. Consequently, they adopted selective strategies, including verifying theoretical content and revising AI-generated text. ChatGPT emerged as a supplementary cognitive partner rather than a replacement for sociological reasoning. The findings underscore the need to integrate AI literacy into sociology curricula to promote critical and ethical AI use in academic writing.
Ähnliche Arbeiten
Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI
2019 · 8.402 Zit.
Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead
2019 · 8.270 Zit.
High-performance medicine: the convergence of human and artificial intelligence
2018 · 7.702 Zit.
Proceedings of the 19th International Joint Conference on Artificial Intelligence
2005 · 5.781 Zit.
Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)
2018 · 5.507 Zit.