The Impact of Artificial Intelligence Tools on Human Cognitive Abilities: A Comprehensive Review
DOI:
https://doi.org/10.63501/hsdq5611Keywords:
Artificial Intelligence, cognitive abilities, critical thinking, cognitive offloading, educational technology, metacognitionAbstract
Background: The rapid integration of artificial intelligence (AI) tools into daily cognitive tasks has raised questions about their impact on human cognitive abilities.
Objective: This review synthesizes current research on how the use of AI tools affects cognitive domains, including critical thinking, memory, creativity, and decision-making.
Methods: We reviewed empirical studies, including randomized controlled trials, field experiments, and neurophysiological investigations, that examined AI's cognitive effects across educational and professional settings.
Results: Evidence reveals a complex, non-linear relationship between AI use and cognition. Moderate AI usage shows minimal cognitive impact, while excessive reliance correlates with decreased critical thinking abilities (cognitive offloading effect), reduced metacognitive accuracy, and lower retention on delayed assessments. Field studies with approximately 1,000 students demonstrated improved immediate performance but worse unassisted outcomes following AI removal, particularly with unrestricted AI access. Neurophysiological data showed reduced brain network connectivity during AI-assisted tasks. Age-related differences emerged, with younger users (17-25 years) exhibiting greater dependence and lower critical-thinking scores. However, well-designed AI tutoring systems with scaffolding produced superior learning gains compared with traditional instruction.
Discussion: The findings reveal a fundamental paradox: AI assistance frequently improves immediate task performance while simultaneously undermining durable skill acquisition. Three primary mechanisms mediate these effects: reduced cognitive effort through offloading, diminished metacognitive monitoring, and altered practice patterns. Critically, the cognitive impact of AI is not deterministic but depends on implementation design, with scaffolded AI systems that maintain active engagement producing dramatically different outcomes than those providing direct solutions. Age-related vulnerabilities, particularly among younger users (17-25 years), raise developmental concerns, though higher educational attainment appears to provide protective effects against cognitive offloading tendencies.
Conclusions: AI tools do not inherently impair or enhance cognition; rather, their impact depends critically on implementation design, user agency, and interaction patterns. Strategic use that maintains active cognitive engagement can augment human capabilities, while passive reliance risks skill atrophy. Educational interventions promoting AI literacy and metacognitive awareness are essential for optimizing cognitive outcomes.
References
Bastani, H., Bastani, O., Sungu, A., Ge, H., Kabakcı, Ö., & Mariman, R. (2024). Generative AI can harm learning. SSRN Working Paper. https://doi.org/10.2139/ssrn.4895486
Bjork, R. A. (1994). Memory and metamemory considerations in the training of human beings. In J. Metcalfe & A. P. Shimamura (Eds.), Metacognition: Knowing about knowing (pp. 185-205). MIT Press. https://psycnet.apa.org/record/1994-97967-009
Clark, A., & Chalmers, D. (1998). The extended mind. Analysis, 58(1), 7-19. https://doi.org/10.1093/analys/58.1.7
Fernandes, D., Villa, S., Nicholls, S., Haavisto, O., Buschek, D., Schmidt, A., Kosch, T., Shen, C., & Welsch, R. (2026). AI makes you smarter but none the wiser: The disconnect between performance and metacognition. Computers in Human Behavior, 175, Article 108779. https://doi.org/10.1016/j.chb.2025.108779
Kestin, G., Miller, K., Klales, A., Milbourne, T., & Ponti, G. (2025). AI tutoring outperforms in-class active learning: An RCT introducing a novel research-based design in an authentic educational setting. Scientific Reports, 15(1), Article 17458. https://doi.org/10.1038/s41598-025-97652-6
Kocoń, J., Cichecki, I., Kaszyca, O., Kochanek, M., Szydło, D., Baran, J., ... & Kazienko, P. (2023). ChatGPT: Jack of all trades, master of none. Information Fusion, 99, Article 101861. https://doi.org/10.1016/j.inffus.2023.101861
Kosmyna, N., Hauptmann, E., Yuan, Y. T., Situ, J., Liao, X. H., Beresnitzky, A. V., Braunstein, I., & Maes, P. (2025). Your brain on ChatGPT: Accumulation of cognitive debt when using an AI assistant for essay writing task. arXiv preprint arXiv:2506.08872. https://doi.org/10.48550/arXiv.2506.08872
Ophir, E., Nass, C., & Wagner, A. D. (2009). Cognitive control in media multitaskers. Proceedings of the National Academy of Sciences, 106(37), 15583-15587. https://doi.org/10.1073/pnas.0903620106
Risko, E. F., & Gilbert, S. J. (2016). Cognitive offloading. Trends in Cognitive Sciences, 20(9), 676-688. https://doi.org/10.1016/j.tics.2016.07.002
Slamecka, N. J., & Graf, P. (1978). The generation effect: Delineation of a phenomenon. Journal of Experimental Psychology: Human Learning and Memory, 4(6), 592-604. https://doi.org/10.1037/0278-7393.4.6.592
Storm, B. C., & Stone, S. M. (2015). Saving-enhanced memory: The benefits of saving on the learning and remembering of new information. Psychological Science, 26(2), 182-188. https://doi.org/10.1177/0956797614559285
Downloads
Published
Issue
Section
License
This article is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits use, sharing, adaptation, distribution, and reproduction in any medium or format, provided appropriate credit is given to the author(s) and the source. To view a copy of this license, visit: https://creativecommons.org/licenses/by/4.0