A study note on AI tool usage and students' intellectual learning capability: for students doing dissertation works

A study note on AI tool usage and students' intellectual learning capability: for students doing dissertation works (using "MBA students" as an example; it is also applicable to Housing Studies students) (information from perplexity.ai)

My question 1: "What could be the reasons that, as MBA dissertation students, we need to do literature review by studying academic articles and reference books, when we can obtain (a) organized and concise academic information and (supporting line of reasoning and idea justifications) direct from perplexity?

The reply: "MBA dissertation students need to conduct literature reviews by studying academic articles and reference books rather than relying solely on AI tools like Perplexity for several key reasons:

1. Comprehensive Understanding and Critical Engagement

A literature review involves gaining a deep understanding of the existing research relevant to the dissertation topic. It is not just about summarizing information but critically evaluating and synthesizing research findings to identify gaps, debates, and methodologies in the field. This process helps students develop insights and a theoretical framework for their own research, which cannot be fully achieved by simply obtaining organized and concise information from AI tools alone.

2. Building Research Skills and Academic Rigor

Conducting a literature review by engaging directly with scholarly sources helps students develop essential research skills such as source evaluation, critical thinking, and academic writing. Over-reliance on AI-generated summaries risks diminishing these skills and may lead to accepting information without sufficient scrutiny, which is problematic given the risks of AI hallucinations, bias, and inaccuracies in AI outputs.

3. Ensuring Accuracy and Reliability

AI tools like Perplexity can sometimes produce misleading or false information (AI hallucinations) and may perpetuate biases present in their training data. Academic articles and reference books undergo peer review and editorial processes that ensure the credibility and reliability of the information, which is crucial for maintaining academic integrity in dissertations.

4. Contextualizing Research and Justifying Originality

A literature review situates the dissertation within the broader academic conversation by showing how the research relates to existing studies and where it contributes new knowledge. This contextualization is vital for justifying the research problem and demonstrating the originality and relevance of the dissertation topic, which AI tools cannot fully replicate.

5. Identifying Research Gaps and Methodological Approaches

Through detailed examination of academic literature, students can identify gaps in knowledge and refine their research questions and methodologies accordingly. This process guides the direction of their research and helps avoid duplication of effort. AI tools may provide summaries but lack the nuanced understanding necessary to map out these research gaps comprehensively.

6. Academic Communication and Scholarly Standards

Writing a literature review is a fundamental academic exercise that prepares students for scholarly communication. It demonstrates familiarity with the discipline, ability to critically analyze sources, and adherence to academic standards. These skills are essential for graduate-level research and future academic or professional work.

In summary, while AI tools like Perplexity can assist by providing quick and organized information, MBA dissertation students must engage with academic articles and reference books to develop a rigorous, critical, and comprehensive literature review. This process ensures accuracy, fosters essential research skills, situates their work within the scholarly community, and ultimately strengthens the quality and credibility of their dissertation."

My question 2: "Is there harmful effect on our development of intellectual competence if we rely too much on getting all the "answers" from perplexity direct when doing our dissertation project?"

The reply: "Yes, there are harmful effects on the development of intellectual competence if MBA dissertation students rely too much on getting all the "answers" directly from AI tools like Perplexity when doing their dissertation projects. Several studies and expert analyses highlight these risks:

1. Decline in Critical Thinking and Problem-Solving Skills

Over-reliance on AI for direct answers can lead to "intellectual deskilling," where students become less capable of independent critical thinking and problem-solving. By outsourcing cognitive tasks to AI, students may engage less deeply with complex problems, reducing their perseverance, intellectual curiosity, and ability to analyze and synthesize information critically.

2. Reduced Creativity and Original Thinking

Heavy use of AI-generated content can weaken creativity and original thinking. When students substitute their own analysis and ideas with AI outputs, they risk producing work that lacks personal insight and depth, which undermines the development of their intellectual competence.

3. Inhibited Development of Metacognitive Skills

Frequent reliance on AI can impair metacognitive abilities—the capacity to reflect on and regulate one’s own thinking processes. This cognitive offloading leads to superficial understanding and poor retention of knowledge, as students do not actively engage in deep learning or critical reflection.

4. Ethical and Academic Integrity Concerns

Dependence on AI can increase risks of plagiarism and academic dishonesty, as AI-generated content may be used without proper understanding or citation. This not only threatens intellectual integrity but also prevents students from developing their own scholarly voice and ethical academic habits.

5. Impact on Long-Term Academic and Professional Skills

Students who rely excessively on AI may struggle with writing, analysis, and communication skills essential for academic success and professional careers. The diminished engagement with challenging tasks can leave them ill-prepared for real-world problems that require independent judgment and creativity.

6. Reduced Engagement and Brain Activity

Research such as the MIT Media Lab study found that users relying on AI tools like ChatGPT showed lower brain engagement and poorer learning outcomes compared to those who worked independently or used traditional search engines. This suggests that AI reliance can reduce active cognitive involvement necessary for intellectual growth.

Summary

While AI tools like Perplexity can be valuable aids, excessive dependence on them for direct answers in dissertation work harms intellectual competence by weakening critical thinking, creativity, metacognition, and ethical academic practices. To develop strong intellectual skills, students must actively engage with research, analyze information critically, and produce original work rather than passively consuming AI-generated responses."

A collection of blog notes on using chatgpt for research purpose.