Scaffolding Higher-Order Thinking Through AI Chatbots: A Multi-Domain Study

<p dir="ltr">The integration of artificial intelligence (AI) into higher education is transforming pedagogical approaches, particularly in fostering higher-order thinking skills. While AI language models have been praised for their potential to enhance higher-order thinking (Lu et al., 2024; Putra et al., 2024), the direct relationship between AI-facilitated cognitive engagement and competency development remains insufficiently explored. The impact of AI chatbot interactions on student performance and engagement remains an under-researched area, particularly in its ability to scaffold different cognitive processes. Whilst recent research has explored the use of AI in academic writing (Kim et al, 2024), and the utility of Socratic prompting (Chukhlomin, 2024), few studies have examined chatbot implementations across different cognitive processes. This study investigates how purpose-built AI chatbots scaffold higher-order thinking by focusing on two key domains: analytical thinking in academic writing and critical thinking in ethical case analysis. The study employs two parallel implementations across student cohorts in two distinct subjects: Effective Communication and Leadership Fundamentals. The first implementation involves writing-focused chatbots that facilitate analytical thinking through structured questioning. The second utilises critical thinking chatbots designed to scaffold reasoning and ethical decision-making. Both implementations, developed on the Mizou platform, adopt a mixed-methods approach, incorporating pre- and post-surveys with 50 participants per subject to assess AI chatbots' effectiveness to ascertain the perceived usefulness of the AI chatbots and students' confidence levels in completing the analytical tasks set. Preliminary findings indicate a significant increase in student confidence and engagement in analytical tasks, with students particularly valuing chatbots as thought-provokers rather than mere knowledge providers. These findings suggest that AI chatbots, when designed as cognitive scaffolds, can effectively support higher-order thinking while maintaining student agency. Future research should explore the long-term impact of AI-facilitated cognitive scaffolding across multiple domains and its role in digital literacy development. The success of these implementations lies in their design as cognitive tools that guide discovery rather than prescribe solutions. Future research directions include investigating the transferability of thinking skills and digital literacy developed through chatbot interactions across multiple cognitive domains. Key words : AI, cognitive scaffolding, higher-order thinking, HOTS, educational technology, metacognition, student-AI interaction</p><p><br></p>