AI tools may weaken critical thinking skills by encouraging cognitive offloading, study suggests

A new study published in the journal Societies suggests that frequent reliance on artificial intelligence tools may negatively affect critical thinking skills. People who used AI tools more frequently demonstrated weaker critical thinking abilities, largely due to a cognitive phenomenon known as cognitive offloading. This effect was particularly pronounced among younger individuals, while those with higher education levels tended to retain stronger critical thinking skills regardless of AI tool usage.

As AI continues to integrate into education, workplaces, and daily decision-making, researchers have raised concerns about its potential effects on cognitive skills. Critical thinking, the ability to analyze, evaluate, and synthesize information to form reasoned conclusions, is essential for problem-solving and independent decision-making. AI tools can provide users with quick solutions, but over-reliance on these tools may reduce opportunities for deep cognitive engagement.

“The motivation for this study stemmed from the increasing integration of AI tools into everyday life and their potential cognitive consequences. While AI offers efficiency and convenience, there is growing concern that reliance on these tools could diminish critical thinking abilities,” said study author Michael Gerlich, the Head of Center for Strategic Corporate Foresight and Sustainability at SBS Swiss Business School.

“Cognitive offloading, where individuals delegate cognitive tasks to AI rather than engaging in deep analytical reasoning, is an emerging phenomenon that has not been extensively studied. Given the crucial role of critical thinking in decision-making, education, and professional competence, understanding how AI influences cognitive engagement is imperative.”

“The study aimed to explore these implications, providing insights that can inform educators, policymakers, and technology designers in fostering a balanced approach to AI adoption. As generative AI enables the offloading of cognitive processes – not merely providing information but allowing users to bypass critical thinking by delivering direct answers – it presents a potential risk that this study sought to examine.”

Free daily newsletter

Gerlich used a mixed-method approach, combining survey data with in-depth interviews. A total of 666 participants from diverse age groups and educational backgrounds completed a survey assessing their AI tool usage, cognitive offloading habits, and critical thinking skills. The survey incorporated standardized measures of critical thinking, including self-reported assessments and performance-based evaluations.

To supplement these findings, Gerlich conducted interviews with 50 participants, allowing them to gain deeper insights into how people perceive their dependence on AI tools and its impact on their cognitive engagement. The interview transcripts were analyzed thematically to identify recurring patterns and concerns.

For the quantitative data, Gerlich used statistical techniques, including correlation analysis and analysis of variance, to examine relationships between AI tool usage, cognitive offloading, and critical thinking performance. A machine learning model known as random forest regression was also employed to determine the relative importance of different factors influencing critical thinking.

The results indicated a strong negative correlation between frequent AI tool usage and critical thinking abilities. Participants who reported heavy reliance on AI tools performed worse on critical thinking assessments compared to those who used these tools less frequently.

Cognitive offloading played a significant role in this relationship. Participants who frequently delegated cognitive tasks to AI tools, such as using search engines for quick answers or relying on recommendation algorithms for decision-making, exhibited weaker critical thinking skills. This pattern suggests that AI tools may be reducing the need for individuals to engage in independent analysis and evaluation.

“The key takeaway is that while AI tools can enhance productivity and information accessibility, their overuse may lead to unintended cognitive consequences,” Gerlich told PsyPost. “The findings reveal a strong negative correlation between frequent AI tool usage and critical thinking abilities, mediated by cognitive offloading. This suggests that reliance on AI tools could reduce opportunities for deep, reflective thinking.

“For individuals, this means being mindful of how they interact with AI and ensuring they continue to engage in critical analysis and independent problem-solving. AI should complement cognitive engagement rather than replace it. This is particularly relevant for education, where AI-driven learning platforms must be integrated in ways that encourage active thinking rather than passive dependence.”

Age and education level were also important factors. Younger participants (aged 17–25) showed higher AI tool usage and greater cognitive offloading, which coincided with lower critical thinking scores. In contrast, older participants (aged 46 and above) demonstrated stronger critical thinking skills and were less reliant on AI tools.

“One surprising finding was the extent to which younger participants exhibited higher dependence on AI tools and, correspondingly, lower critical thinking scores,” Gerlich said. “This suggests that digital natives, who have grown up with AI-integrated technologies, might be more prone to cognitive offloading than older generations.”

Education level played a protective role—those with higher education tended to maintain strong critical thinking skills, even if they used AI tools frequently. This suggests that formal education may provide individuals with strategies to critically assess AI-generated information rather than accepting it uncritically.

“The role of education level was striking – higher educational attainment was consistently associated with stronger critical thinking skills, regardless of AI usage,” Gerlich told PsyPost. “This implies that educational interventions could play a crucial role in mitigating the cognitive costs of AI dependence. Similar results showed the Microsoft study that showed that knowledge workers with lower self confidence trust AI more and offload cognitive processes compared to people with higher self confidence.”

The qualitative data from interviews reinforced these findings. Many participants expressed concerns about their growing dependence on AI tools. Some younger users admitted that they rarely questioned the accuracy of AI-generated recommendations, while others worried that they were losing their ability to analyze and solve problems independently. Participants with higher education backgrounds were more likely to critically evaluate AI outputs, demonstrating skepticism and cross-checking information from multiple sources.

The findings highlight the need to encourage users to engage with information actively rather than passively accepting AI-generated conclusions. But as with all research, there are some caveats. The study does not establish a direct causal relationship—while AI tool usage is associated with lower critical thinking scores, it is possible that individuals with weaker critical thinking skills are simply more inclined to rely on AI tools in the first place.

“The study highlights correlations rather than direct causation, meaning that while AI usage and cognitive offloading are linked to reduced critical thinking, additional factors may also contribute to this relationship,” Gerlich noted. “Moreover, the study focuses on self-reported and assessment-based measures of critical thinking, which may not capture all dimensions of cognitive engagement. The findings should be interpreted in the broader context of evolving AI-human interactions, and further research is needed to explore long-term effects across different demographic groups and technological contexts.”

Future research could use experimental designs to test whether interventions, such as teaching users how to critically engage with AI-generated information, can mitigate cognitive offloading effects. Longitudinal studies tracking individuals’ AI tool usage and critical thinking development over time could also provide deeper insights into how technology shapes cognitive abilities in the long run.

“The long-term objective of this research is to deepen our understanding of how AI influences human cognition, particularly in the areas of critical thinking, decision-making, and trust,” Gerlich explained. “Future studies will explore how individuals develop cognitive dependencies on AI, the extent to which AI-generated content affects analytical reasoning, and strategies for mitigating potential cognitive decline.

“Additionally, I aim to investigate the role of AI in educational settings, assessing how AI-driven learning tools can be designed to enhance rather than diminish critical thinking. Beyond academia, the goal is to contribute to policy discussions and educational frameworks that ensure AI integration supports, rather than undermines, cognitive engagement and intellectual autonomy. All of my research focuses on the Human Factor of AI implementation, which is too often overlooked when discussing the amazing technical opportunities AI offers.”

“As AI becomes increasingly embedded in daily life, it is essential to foster awareness of its cognitive implications,” Gerlich concluded. “AI tools are not inherently detrimental; rather, their impact depends on how they are used. The challenge lies in striking a balance between leveraging AI’s capabilities for efficiency and ensuring that individuals remain cognitively engaged and capable of independent critical thought. Moving forward, interdisciplinary collaboration between educators, policymakers, and technologists will be crucial in designing AI systems that support human cognition rather than replace it.”

The study, “AI Tools in Society: Impacts on Cognitive Offloading and the Future of Critical Thinking,” was published January 3, 2025.