Leveraging AI in E-Learning: Personalized Learning and Adaptive Assessment through Cognitive Neuropsychology—A Systematic Analysis

May combine two claims.

AI can improve students' performance, engagement, and motivation.

“The results indicate that AI can improve students' performance, engagement, and motivation; at the same time, some challenges like bias and discrimination should be noted.”

Probably not a substantial claim: may not pass S4 (based).

The historical literature on AI in education has largely remained descriptive, listing technological advances without deep critical analysis.

“While the historical progress in AI in education is impressive, the literature has largely remained descriptive, listing technological advances without deep critical analysis.”

Probably not a substantial claim: may not pass S4 (based).

Most early learning theories have yet to be integrated with state-of-the-art AI technologies.

“In addition, most of the early learning theories have yet to be integrated with state-of-the-art AI technologies.”

Probably not a substantial claim: may not pass S4 (based).

The current literature on AA remains for the most part descriptive.

“Yet, while the promise of AA is clear, the current literature remains for the most part descriptive.”

Probably not a substantial claim: may not pass S4 (based).

The existing literature on AI in e-learning is mostly theoretical and not deep, especially regarding how to apply cognitive neuropsychology to inform AI-driven systems.

“Although the existing literature has demonstrated various possibilities using AI, they are mostly theoretical and not deep, especially where they concern how to apply cognitive neuropsychology to inform AI-driven systems.”

May combine two claims.

Lesser-explored objectives included memory retention (n = 5) and student motivation (n = 8).

“Lesser-explore objectives included “Memory Retention” (n = 5) and “Student Motivation” (n = 8).”

May combine two claims.

Most research on AI in e-learning is descriptive and has not explored realistic challenges of large-scale implementation.

“Most research is of a descriptive nature, and the more realistic challenges that are in such implementations on a larger scale have not been explored in single-case studies, pilot programs, or small batches of adopters.”

Probably not a substantial claim: may not pass S3 (load-bearing). The quote only partly states this claim.

The pie chart shows the proportions of statistical techniques used across reviewed studies, including descriptive statistics, regression analysis, ANOVA, factor analysis, and ML-based statistical models.

“The resulting pie chart shows the proportion of each statistical method, providing an at-a-glance understanding of which techniques are most used in the field (Figure 4).”

May combine two claims.

The effectiveness and fairness of AA systems are still in early stages of exploration, and their full impact has yet to be realized.

“While research has demonstrated the potential of AA, the effectiveness and fairness of this system are still in its early stages of exploration, and the full impact of such systems has yet to be realized [96][97][98][99].”

• cites [96][97][98][99]Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

The distribution of research outcomes across all 85 studies was categorized into positive, negative, mixed, and neutral outcomes.

“The pie chart (Figure 5) summarizing the distribution of research outcomes across all 85 studies provides a comprehensive overview of the findings categorized into positive, negative, mixed, and neutral outcomes.”

May combine two claims.

Most studies on AI in education have emphasized AI-driven systems for personalization but lack critical discussion of long-term effectiveness and scalability.

“Most of the studies have emphasized the role of AI-driven systems for personalization in education.However, critical discussions regarding their long-term effectiveness and scalability are required.”

May combine two claims.

By synthesizing the results of 85 studies, the review found significant evidence supporting AI's transformative role in education, particularly in improving student engagement, learning outcomes, and the efficiency of assessment strategies.

“By synthesizing the results of 85 studies, we found significant evidence supporting AI's transformative role in education, particularly in improving student engagement, learning outcomes, and the efficiency of assessment strategies.”

May combine two claims.

The most frequently explored topics in the reviewed studies were personalized learning (n=25), AI-driven tutoring (n=22), and adaptive assessment (n=20).

“The objectives include”

May combine two claims.

Lesser-explored objectives in the reviewed studies included memory retention (n=5) and student motivation (n=8).

“Lesser-explored objectives included”

Classified as probably not a substantial claim.

This review followed the PRISMA systematic review methodology.

“This review follows the PRISMA systematic review methodology and synthesizes the results of 85 studies that were selected from an initial pool of 818 records across several databases.”

• Also stated in: § 3.1. Data Collection and Analysis · plain

  “The methodology employed in this systematic review adhered to the guidelines outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses by the PRISMA Statement [127].”

• cites [127] Page, 2021, Updating guidance for reporting systematic reviews: Development of the PRISMA 2020 statement, J. Clin. Epidemiol. sourceNot checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S2 (checkable).

AI has transformative potential in developing personalized and adaptive learning environments.

“The review identifies the transformative potential of AI in developing personalized and adaptive learning (AL) environments, thus, it advocates continued development and exploration as a means to improve educational outcomes.”

May combine two claims.

Engagement Metrics can monitor interaction levels, using data to adjust learning experiences and re-engage students when attention wanes.

“Engagement Metrics can monitor interaction levels, using data to adjust learning experiences and re-engage students when attention wanes [139].”

• cites [139] Dabingaya, 2022, Analyzing the effectiveness of AI-powered adaptive learning platforms in mathematics education, Int. J. Educ. Pract.Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

Neutral and negative outcomes highlight challenges or limitations encountered in the research.

“Mixed outcomes indicate areas where results were less clear or context-dependent, while neutral and negative outcomes highlight challenges or limitations encountered in the research.”

May combine two claims.

AI-powered tutors using Intelligent Feedback Mechanisms offer automated, metric-based feedback, at times outperforming traditional expert instruction.

“AI-powered tutors, using Intelligent Feedback Mechanisms, offer automated, metric-based feedback, at times outperforming traditional expert instruction [143].”

• cites [143] Fazlollahi, 2022, Effect of artificial intelligence tutoring vs expert instruction on learning simulated surgical skills among medical students, JAMA Netw. Open sourceNot checked yet

  Not checked in the free run.

Classified as probably not a substantial claim. May combine two claims.

Studies that lacked clear methodological design or had small sample sizes were excluded.

“Studies that lacked clear methodological design or had small sample sizes were excluded.”

Classified as probably not a substantial claim.

Studies lacking complete data reporting or with vague descriptions of AI applications were excluded to base the review on high-quality and transparent research.

“Due to a lack of complete data reporting or vague descriptions regarding the AI applications, such studies were excluded to base the review on high-quality and transparent research.”

May combine two claims.

Formative Feedback and Assessment tools together with Real-Time Feedback systems like EEG-Based Neurofeedback provide immediate actionable insights that substantially improve learning performance.

“Formative Feedback and Assessment tools [178], together with Real-Time Feedback systems like EEG-Based Neurofeedback, provide immediate actionable insights that substantially improve learning performance [160].”

• cites [178] Nazari, 2021, Application of Artificial Intelligence powered digital writing assistant in higher education: Randomized controlled trial, Heliyon sourceNot checked yet

  Not checked in the free run.

• cites [160] Kim, 2023, Development of a computer-aided education system inspired by face-to-face learning by incorporating EEG-based neurofeedback into online video lectures, IEEE Trans. Learn. Technol. sourceNot checked yet

  Not checked in the free run.

Classified as probably not a substantial claim.

Cognitive neuropsychology helps improve teaching methodology for better learning processes.

“Cognitive neuropsychology, in its study of the natural processes of the development of the brain and learning, helps to work out and improve the methodology of teaching for better processes of learning.”

• cites [114][115][116][117][118]Not checked yet

  Not checked in the free run.

May combine two claims.

Affective Reinforcement Learning uses emotional and engagement cues to personalize educational content, which may enhance learning outcomes.

“Affective Reinforcement Learning uses emotional and engagement cues to personalize educational content, which may enhance learning outcomes [182].”

• cites [182] Park, 2019, A model-free affective reinforcement learning approach to personalization of an autonomous social robot companion for early literacy education, Proc. AAAI Conf. Artif. Intell.Not checked yet

  Not checked in the free run.

May combine two claims.

AA involves more accurate estimates of student skills and increased efficiency in test development.

“This technique thus involves more accurate estimates of student skills and increased efficiency in test development, as fewer questions are placed well beyond a student's skill level.”

Classified as probably not a substantial claim.

The review's quality assessment prioritized randomized controlled trials, longitudinal studies, and systematic reviews.

“Study Design: Priority was given to randomized controlled trials, longitudinal studies, and systematic reviews, which provided more reliable data on AI's impact on learning outcomes.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Combining AI-driven feedback with human input may provide comprehensive guidance tailored to individual needs.

“Combining AI-driven feedback with human input may provide comprehensive guidance tailored to individual needs [142].”

• cites [142] Escalante, 2023, AI-generated feedback on writing: Insights into efficacy and ENL student preference, Int. J. Educ. Technol. High. Educ. sourceNot checked yet

  Not checked in the free run.

May combine two claims.

Representations and algorithms in digital technologies can model how brain systems are organized and work, suggesting theoretical connections between neuroscience and digital technologies used in e-learning.

“For example, representations and algorithms in digital technologies can model how brain systems are organized and work, suggesting sound theoretical connections between neuroscience and digital technologies used in e-learning [119-122].”

• cites [119] Ruiz, 2020, A Meta-Analysis of Neuropsychological Effort Test Performance in Psychotic Disorders, Neuropsychol. Rev. sourceNot checked yet

  Not checked in the free run.

• cites [120] Morris, 2022, Disordered Decision Making: A Cognitive Framework for Apathy and Impulsivity in Huntington’s Disease, Mov. Disord. sourceNot checked yet

  Not checked in the free run.

• cites [121] Kanner, 2020, Cognitive disorders in epilepsy I: Clinical experience, real-world evidence and recommendations, Seizure sourceNot checked yet

  Not checked in the free run.

• cites [122] Friedman, 2022, The role of prefrontal cortex in cognitive control and executive function, Neuropsychopharmacology sourceNot checked yet

  Not checked in the free run.

Classified as probably not a substantial claim. May combine two claims.

Positive outcomes dominated the chart, suggesting many interventions or methodologies explored are effective or beneficial in e-learning and AI integration.

“The dominance of positive outcomes in the chart might suggest that many of the interventions or methodologies explored are effective or beneficial in the context of e-learning and AI integration.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Mixed outcomes indicate areas where results were less clear or context-dependent.

“Mixed outcomes indicate areas where results were less clear or context-dependent, while neutral and negative outcomes highlight challenges or limitations encountered in the research.”

May combine two claims.

Micro-Break Activities supported by AI-based decision-making frameworks enable learners to manage emotions such as anxiety, refocus, and improve performance.

“Micro-Break Activities, supported by AI-based decision-making frameworks, enable the learner to manage their emotions, such as anxiety, to refocus and improve their performance [130,140,192,201].”

• cites [130] Falih, 2019, The effect of the intelligent tutoring systems on the education, J. Inf. Res.Not checked yet

  Not checked in the free run.

• cites [140] Darejeh, A., Moghadam, T.S., Delaramifar, M., and Mashayekh, S. (2024, January 27–29). A framework for AI-powered decision making in developing adaptive e-learning systems to impact learners’ emotional responses. Proceedings of the IEEE International Conference on e-Learning, e-Management and e-Services, Isfahan, Iran.Not checked yet

  Not checked in the free run.

• cites [192] Spain, 2021, A reinforcement learning approach to adaptive remediation in online training, Int. J. Artif. Intell. Educ.Not checked yet

  Not checked in the free run.

• cites [201] Wang, 2011, Data mining for adaptive learning in a TESL-based e-learning system, Expert Syst. Appl.Not checked yet

  Not checked in the free run.

May combine two claims.

PL enhances learners' motivation and optimizes learning.

“It does, indeed, open new perspectives on the stimulation of learners and on enhancing their motivation to learn, hence optimizing learning [41][42][43].”

• cites [41][42][43]Not checked yet

  Not checked in the free run.

Classified as probably not a substantial claim. May combine two claims.

The review excluded studies that lacked clear methodological design or had small sample sizes.

“Studies that lacked clear methodological design or had small sample sizes were excluded.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The AI system may schedule tasks at more appropriate times and offer personalized reading to improve mood if necessary.

“The AI system may schedule the tasks at more appropriate times and offer the learner some personalized reading to get them in a good mood if necessary [157].”

• cites [157] Kaur, 2021, Affective state and learning environment-based analysis of students’ performance in online assessment, Int. J. Cogn. Comput. Eng.Not checked yet

  Not checked in the free run.

Classified as probably not a substantial claim.

Each included study was evaluated for methodological rigor using criteria including study design, replicability, and data reporting.

“Each study was evaluated for methodological rigor using the following criteria:”

Classified as probably not a substantial claim.

Priority was given to randomized controlled trials, longitudinal studies, and systematic reviews because they provide more reliable data on AI's impact on learning outcomes.

“Priority was given to randomized controlled trials, longitudinal studies, and systematic reviews, which provided more reliable data on AI's impact on learning outcomes.”

Classified as probably not a substantial claim.

A multidimensional method of personalizing learning can incorporate principles of attention and perception into an AI system.

“The multidimensional method of the personalization of learning is considered one which can incorporate the principles of attention and perception into an AI system.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Field-dependent learners need more guidance while field-independent learners require more freedom.

“Further considerations of cognitive style might, for instance, be that field-dependent learners need more guidance while field-independent learners require more freedom [197].”

• cites [197] 2019, Effect of a metacognitive scaffolding on self-efficacy, metacognition, and achievement in e-learning environments, Knowl. Manag. E-Learn. Int. J.Not checked yet

  Not checked in the free run.

Classified as probably not a substantial claim.

AI assessments could employ multi-modal data fusion techniques that estimate and react to cognitive loads by adapting the learning environment to the learner's state.

“Empowered by AI, the assessments could employ multi-modal data fusion techniques that estimate and react to cognitive loads by adapting the learning environment to the state of the learner [128].”

• cites [128] Abdurrahman, U.A., Yeh, S., Wong, Y., and Wei, L. (2021). Effects of neuro-cognitive load on learning transfer using a virtual reality-based driving system. Big Data Cogn. Comput., 5. sourceNot checked yet

  Not checked in the free run.

May combine two claims.

AI can facilitate collaborative learning by forming heterogeneous groups that maximize cognitive diversity, fostering critical reflection and improving learning outcomes.

“Additionally, AI can facilitate collaborative learning by forming heterogeneous groups that maximize cognitive diversity, fostering critical reflection and improving learning outcomes [177].”

• cites [177] Nalli, G., Amendola, D., and Smith, S. (2022, January 21–23). Artificial intelligence to improve learning outcomes through online collaborative activities. Proceedings of the 21st International Conference on E-learning and Education Technology, Yamanashi, Japan. sourceNot checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing).

Integrating Social Cognitive Theory may promote social engagement and interaction, contributing to a more supportive learning environment.

“Integrating Social Cognitive Theory may promote social engagement and interaction, contributing to a more supportive learning environment [131].”

• cites [131] Almulla, M., and Al-rahmi, W. (2023). Integrated social cognitive theory with learning input factors: The effects of problem-solving skills and critical thinking skills on learning performance sustainability. Sustainability, 15. sourceNot checked yet

  Not checked in the free run.

May combine two claims. The quote only partly states this claim.

VR and AI applications reported include enhancements in diagnostic skills and more stimulating environments for learning.

“Applications reported include enhancements in diagnostic skills, along with more stimulating environments for learning [147].”

• cites [147] Gutiérrez-Maldonado, J., Peñaloza, C., Jarne-Esparcia, A., Talarn, A., Andrés-Pueyo, A., Aguilar-Alonso, A., and Ferrer-García, M. (2010, January 12–14). A training program of differential diagnosis skills based on virtual reality and artificial intelligence. Proceedings of the 7th International Conference on Information Technology: New Generations, Las Vegas, NV, USA. sourceNot checked yet

  Not checked in the free run.

May combine two claims.

Tools such as the Natural Semantic Network can provide relevant personalized content, which has been shown to increase engagement and performance.

“Tools such as the Natural Semantic Network can help provide relevant personalized content, which has been shown to increase engagement and performance [175].”

• cites [175] 2024, Cognitive technology to evaluate the academic learning of computational cognition in psychology students, Int. J. Educ. Methodol.Not checked yet

  Not checked in the free run.

May combine two claims.

Multimodal learning methods such as the VAK model engage different parts of the brain, enhancing comprehension and recall.

“Multimodal learning methods, such as the VAK model, engage different parts of the brain, enhancing comprehension and recall [163].”

• cites [163] Kozierkiewicz-Hetmanska, A. (2010). Evaluating the Effectiveness of Intelligent Tutoring System Offering Personalized Learning Scenario. Advanced Learning Technologies, Springer.Not checked yet

  Not checked in the free run.

May combine two claims.

Adaptive AI-enabled e-learning maintains student motivation, an important factor in optimizing attention.

“It has been shown that adaptive AI-enabled e-learning maintains student motivation, and it is an important factor in optimizing attention-an extremely perishable cognitive resource [206].”

• cites [206] Younes, 2021, Examining the effectiveness of using adaptive AI-enabled e-learning during the pandemic of COVID-19, Educ. Inf. Technol.Not checked yet

  Not checked in the free run.

May combine two claims.

Electrophysiology and hyperscanning research provide insights into brain activity during learning, which AI systems can use to adapt learning environments based on real-time cognitive data.

“Electrophysiology and hyperscanning research provide insights into brain activity during learning, which AI systems can use to adapt learning environments based on real-time cognitive data [134].”

• cites [134] Balconi, 2023, Electrophysiology and hyperscanning applied to e-learning for organizational training, Learn. Organ. sourceNot checked yet

  Not checked in the free run.

May combine two claims.

Predictive algorithms may reduce cognitive load and improve perception by forecasting learners' needs.

“Predictive algorithms may reduce cognitive load and improve perception by forecasting learners' needs [145].”

• cites [145] Flores, E.G.R., Mena, J., López-Camacho, E., and López, O.O. (2019, January 12–23). Adaptive learning based on AI with predictive algorithms. Proceedings of the 2019 International Conference on Artificial Intelligence, Wuhan, China.Not checked yet

  Not checked in the free run.

May combine two claims.

AI-powered decision models could assess the emotional state of learners and select appropriate activities to mitigate anxiety or negative feelings to maintain focused attention and fluent language use.

“For instance, AI-powered decision models could assess the emotional state of learners and select appropriate activities for them, which could mitigate anxiety or negative feelings to maintain focused attention and the fluent use of language [140].”

• cites [140] Darejeh, A., Moghadam, T.S., Delaramifar, M., and Mashayekh, S. (2024, January 27–29). A framework for AI-powered decision making in developing adaptive e-learning systems to impact learners’ emotional responses. Proceedings of the IEEE International Conference on e-Learning, e-Management and e-Services, Isfahan, Iran.Not checked yet

  Not checked in the free run.

May combine two claims.

Affect-aware feedback has become a crucial part of personalized learning because emotional states impact learning.

“Affect-aware feedback has become a crucial part of PL, since emotional states have a great impact on the learning process itself [146].”

• cites [146] Grawemeyer, 2017, Affective learning: Improving engagement and enhancing learning with affect-aware feedback, User Model. User-Adapt. Interact. sourceNot checked yet

  Not checked in the free run.

May combine two claims.

Semantic network analysis can help optimize language learning by assessing changes in knowledge structures and enabling AI to tailor content to align with learners' existing cognitive patterns.

“Semantic network analysis can help optimize language learning by assessing changes in knowledge structures and enabling AI to tailor content to align with learners' existing cognitive patterns.”

• cites [175] 2024, Cognitive technology to evaluate the academic learning of computational cognition in psychology students, Int. J. Educ. Methodol.Not checked yet

  Not checked in the free run.

May combine two claims.

Interactive and Adaptive Content, such as quizzes embedded within videos, can ensure students understand concepts before progressing, helping maintain engagement.

“Interactive and Adaptive Content, such as quizzes embedded within videos, can ensure students understand concepts before progressing, helping to maintain engagement [159].”

• cites [159] Khalil, 2022, Impact of quiz-based interactive videos into personal learning environment on regulating e-portfolio design and learning engagement: An experimental study, Univers. J. Educ. Res. sourceNot checked yet

  Not checked in the free run.

May combine two claims.

AI-enabled recommendation agents can facilitate content personalization in videos to increase engagement and improve learning outcomes.

“AI-enabled recommendation agents can facilitate content personalization in videos to increase engagement and improve learning outcomes [153].”

• cites [153] Huang, 2023, Effects of artificial intelligence-enabled personalized recommendations on learners’ learning engagement, motivation, and outcomes in a flipped classroom, Comput. Educ. sourceNot checked yet

  Not checked in the free run.

May combine two claims.

Adaptation of content presentation by prior knowledge and learners' individual preferences may ensure the learner receives proper material at the right moment, satisfying continuous assessment and instruction by adaptive learning platforms.

“Adaptation of content presentation by prior knowledge and to the learners' individual preferences may also ensure that the learner receives the proper material at the right moment, thus satisfying continuous assessment and instruction provided by ALPs [203].”

• cites [203] Wang, 2020, When adaptive learning is effective learning: Comparison of an adaptive learning system to teacher-led instruction, Interact. Learn. Environ.Not checked yet

  Not checked in the free run.

May combine two claims.

AI-assisted tools including ChatGPT can positively influence writing skills and motivation.

“It has also been shown that AI-assisted tools, including ChatGPT, can positively influence writing skills and motivation.”

• cites [191] Song, C., and Song, Y. (2023). Enhancing academic writing skills and motivation: Assessing the efficacy of ChatGPT in AI-assisted language learning for EFL students. Front. Psychol., 14. sourceNot checked yet

  Not checked in the free run.

May combine two claims.

Integration of biofeedback mechanisms tracking physiological states enables personalized instruction strategies and improves engagement and performance.

“The integration of biofeedback mechanisms, which track and react to the physiological states of the learners, will enable personalized instruction strategies and improve their engagement and performance [153].”

• cites [153] Huang, 2023, Effects of artificial intelligence-enabled personalized recommendations on learners’ learning engagement, motivation, and outcomes in a flipped classroom, Comput. Educ. sourceNot checked yet

  Not checked in the free run.

May combine two claims.

Individualizing problem-solving tasks to the learner's current knowledge and gradually raising task difficulty enhances numeric concept acquisition and improves numeric information storing and retrieval.

“Numeric cognition can be optimized by individualizing problem-solving tasks to the learner's current knowledge and by gradually raising task difficulty.It enhances learners' numeric concept acquisition and improves their numeric information storing and retrieval capabilities [203].”

• cites [203] Wang, 2020, When adaptive learning is effective learning: Comparison of an adaptive learning system to teacher-led instruction, Interact. Learn. Environ.Not checked yet

  Not checked in the free run.

May combine two claims.

Interactive videos with embedded quizzes adapt to learner performance, enhancing engagement in numeric tasks.

“Interactive videos with embedded quizzes adapt to learner performance, enhancing engagement in numeric tasks [159].”

• cites [159] Khalil, 2022, Impact of quiz-based interactive videos into personal learning environment on regulating e-portfolio design and learning engagement: An experimental study, Univers. J. Educ. Res. sourceNot checked yet

  Not checked in the free run.

May combine two claims.

Cognitive graphs can support personalized learning paths as learners progress, ensuring content relevance and promoting cognitive processing and mastery of problem-solving skills.

“Cognitive graphs can further support PL paths as learners progress, ensuring that the delivery of content remains relevant to the learner's needs, promoting better cognitive processing and mastery of problem-solving skills [176,189,203].”

• cites [176] Mu, 2023, Research on a PL path recommendation system based on cognitive graph with a cognitive graph, Interact. Learn. Environ.Not checked yet

  Not checked in the free run.

• cites [189] Rosen, Y., Rushkin, I., Rubin, R., Munson, L., Ang, A.M., Weber, G., Lopez, G., and Tingley, D. (2018, January 7–9). The effects of adaptive learning in a massive open online course on learners’ skill development. Proceedings of the 8th International Conference on Learning Analytics and Knowledge, Sydney, NSW, Australia. sourceNot checked yet

  Not checked in the free run.

• cites [203] Wang, 2020, When adaptive learning is effective learning: Comparison of an adaptive learning system to teacher-led instruction, Interact. Learn. Environ.Not checked yet

  Not checked in the free run.

May combine two claims.

Neurofeedback systems monitoring state of mind in real-time improve memory retention and numeric problem-solving.

“Neurofeedback systems, which monitor the state of the mind in real-time, improve the capability of memory retention and numeric problem-solving [160].”

• cites [160] Kim, 2023, Development of a computer-aided education system inspired by face-to-face learning by incorporating EEG-based neurofeedback into online video lectures, IEEE Trans. Learn. Technol. sourceNot checked yet

  Not checked in the free run.

May combine two claims.

Automated courses through virtual assistants are cost-effective and provide immediate feedback and support for problem resolution.

“Studies have also highlighted the cost-effectiveness of automated courses through virtual assistants that provide immediate feedback and support for problem resolution [138].”

• cites [138] Bobrytska, 2020, Effectiveness and stakeholders’ perceptions of the integration of automated e-learning courses into vocational education programmes in universities in Ukraine, Int. J. Learn. Teach. Educ. Res.Not checked yet

  Not checked in the free run.

May combine two claims.

Emotion recognition systems support a positive learning environment, promoting memory retention and cognitive engagement.

“Emotion recognition systems support a positive learning environment, promoting memory retention and cognitive engagement [129].”

• cites [129] Alam, F., Yang, Q., Bhutto, M.Y., and Akhtar, N. (2021). The influence of e-learning and emotional intelligence on psychological intentions: Study of stranded Pakistani students. Front. Psychol., 12. sourceNot checked yet

  Not checked in the free run.

May combine two claims.

Adaptive tutorials that adjust content based on learners' responses increase engagement and comprehension.

“Video-based feedback has been shown to enhance understanding and retention, while adaptive tutorials that adjust content based on learners' responses increase engagement and comprehension [180,204].”

• cites [180] Ostrow, K.S., and Heffernan, N. (2014, January 4–7). Testing the multimedia principle in the real world: A comparison of video vs. text feedback in authentic middle school math assignments. Proceedings of the 2014 International Conference on Educational Data Mining, London, UK.Not checked yet

  Not checked in the free run.

• cites [204] Wong, 2015, Adaptive tutorials versus web-based resources in radiology: A mixed methods comparison of efficacy and student engagement, Acad. Radiol. sourceNot checked yet

  Not checked in the free run.

May combine two claims.

AI-powered formative feedback tools help sustain engagement and improve problem-solving skills through timely and actionable insights.

“AI-powered formative feedback tools will help in sustaining engagement and improving problem-solving skills through timely and actionable insights [178].”

• cites [178] Nazari, 2021, Application of Artificial Intelligence powered digital writing assistant in higher education: Randomized controlled trial, Heliyon sourceNot checked yet

  Not checked in the free run.

May combine two claims.

Research has proven a correlation between adaptive learning technologies and improved learning outcomes via continuous assessments and adaptive instructional content with significant efficiency gains.

“Research has proven a correlation between AL technologies and improved learning outcomes via continuous assessments and adaptive instructional content with significant efficiency gains [201].”

• cites [201] Wang, 2011, Data mining for adaptive learning in a TESL-based e-learning system, Expert Syst. Appl.Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

Integration of memory storage and retrieval principles with numeric cognition into AI-driven PL systems provides an overall robust framework for optimizing educational outcomes.

“Therefore, the integration of memory storage and retrieval principles with numeric cognition into the AI-driven PL systems means an overall robust framework for optimizing educational outcomes.”

May combine two claims.

Interactive videos with embedded quizzes that adapt according to performance enhance learner engagement and understanding.

“Interactive videos with embedded quizzes that adapt according to performance further enhance learner engagement and understanding [159].”

• cites [159] Khalil, 2022, Impact of quiz-based interactive videos into personal learning environment on regulating e-portfolio design and learning engagement: An experimental study, Univers. J. Educ. Res. sourceNot checked yet

  Not checked in the free run.

May combine two claims.

AI-based decision frameworks focusing on assessment of emotions and introduction of micro-break activities have provided evidence that emotional states may be better regulated with overall improvement in learning outcomes.

“Because emotional states are important determinants of effective PL, the recent use of AI-based decision frameworks focusing on the assessment of emotions and introduction of micro-break activities has provided evidence that emotional states may be better regulated with an overall improvement in learning outcomes.”

• cites [140] Darejeh, A., Moghadam, T.S., Delaramifar, M., and Mashayekh, S. (2024, January 27–29). A framework for AI-powered decision making in developing adaptive e-learning systems to impact learners’ emotional responses. Proceedings of the IEEE International Conference on e-Learning, e-Management and e-Services, Isfahan, Iran.Not checked yet

  Not checked in the free run.

May combine two claims.

Metacognitive scaffolding improves metacognitive abilities and learning outcomes.

“Metacognitive scaffolding, which supports the planning, monitoring, and evaluation of activities, improves metacognitive abilities and learning outcomes [197].”

• cites [197] 2019, Effect of a metacognitive scaffolding on self-efficacy, metacognition, and achievement in e-learning environments, Knowl. Manag. E-Learn. Int. J.Not checked yet

  Not checked in the free run.

Classified as probably not a substantial claim. May combine two claims.

An adaptive and personalized AI-driven learning system can employ real-time feedback, emotional and cognitive engagement, interactive content, data-driven approaches, scaffolding, social learning, and specialized technologies to bring efficiency to personalized learning experiences.

“An adaptive and personalized AI-driven learning system can employ real-time feedback, emotional and cognitive engagement, interactive content, data-driven approaches, scaffolding, social learning, and specialized technologies to bring efficiency to PL experiences.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

Understanding how language systems work in the brain could support continuous updating of learning experiences by AI-driven systems based on the learner's emotional state, cognitive style, and learning objectives.

“An understanding of how the systems of language work in the brain could conceivably support the continuous updating of learning experiences by AI-driven systems in view of the learner's emotional state, cognitive style, and learning objectives.”

May combine two claims.

AL paths that track knowledge retention over time enable timely reviews and provide continuous assessment, enabling personalized instruction.

“AL paths that track knowledge retention over time enable timely reviews and provide continuous assessment, enabling personalized instruction [152,201].”

• cites [152] Hinkle, 2020, Pilot of an adaptive learning platform in a graduate nursing education pathophysiology course, J. Nurs. Educ. sourceNot checked yet

  Not checked in the free run.

• cites [201] Wang, 2011, Data mining for adaptive learning in a TESL-based e-learning system, Expert Syst. Appl.Not checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

AI powers the intelligent tutoring systems of ALEKS and Carnegie Learning, whereby real-time adaptation can be done on each performance by students.

“AI powers the intelligent tutoring systems of ALEKS and Carnegie Learning, whereby real-time adaptation can be done on each performance by students.”

Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

Continuous adaptive scaffolding should foster goal setting, reflections on one's understanding, and self-regulation of strategies.

“The continuous adaptive scaffolding should foster goal setting, reflections on one's understanding, and self-regulation of strategies.”

May combine two claims.

On-the-fly adjustment in neuroadaptive systems not only results in better learning but also prevents cognitive burnout, such that the trainee is able to maintain high performance over extended training periods.

“Such on-the-fly adjustment not only results in better learning but also prevents cognitive burnout, such that the trainee is able to maintain high performance over extended training periods [217].”

• cites [217] Chen, 2022, Artificial Intelligence-Virtual Trainer: Innovative Didactics Aimed at Personalized Training Needs, J. Knowl. Econ. sourceNot checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

AI-powered systems such as NeuroNation and iSTART utilize cognitive performance for managing cognitive load, enhancing memory retention, and regulating attention.

“This ensures that AI-powered systems, such as NeuroNation and iSTART, utilize cognitive performance for managing cognitive load, enhancing memory retention, and regulating attention.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

Integration of learning styles within PL scenarios is particularly relevant for improving affective and motivational outcomes.

“Integration of learning styles within PL scenarios is particularly relevant for improving affective and motivational outcomes.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

Social constructivist approaches support metacognitive skills development where learning arises through interaction with knowledgeable peers or AI systems.

“Social constructivist approaches also support metacognitive skills development in a considerable way, where learning arises through interaction either with knowledgeable peers or AI systems.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

AI systems analyze data on behavior and performance and can congregate heterogeneous groupings that maximize collaboration and learning consequences.

“Drawing on the principles of social learning, AI systems analyze data on the behavior and performance of the students and, hence, have the chance to congregate heterogeneous groupings that maximize collaboration and learning consequences.”

Probably not a substantial claim: may not pass S4 (based).

AI constantly adapts to the cognitive and emotional needs of the learner through optimized engagement and outcomes from personalized real-time feedback and intervention.

“Figure visually sums up how AI constantly adapts to the cognitive and emotional needs of the learner through optimized engagement and outcomes from personalized real-time feedback and intervention.”

Probably not a substantial claim: may not pass S2 (checkable).

Integration of tools like self-assessment quizzes, reflective prompts, and progress tracking would enable learners to plan, monitor, and evaluate their learning processes.

“For improving meta-cognitive strategies, integration of tools like self-assessment quizzes, reflective prompts, and progress tracking would be possible.These activities would enable learners to plan, monitor, and evaluate their learning processes.”

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing). May combine two claims.

Regular auditing of bias in algorithmic outputs, combined with the execution of fairness metrics like demographic parity, can help mitigate these risks.

“Regular auditing of bias in algorithmic outputs, combined with the execution of fairness metrics like demographic parity, can help mitigate these risks.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

State-of-the-art systems like Affective AutoTutor adapt content and emotional feedback based on the learner's cognition state and emotional engagement.

“The full integration of AI and PL brings further advances from cognitive neuropsychology whereby state-of-the-art systems like Affective AutoTutor adapt content and emotional feedback based on the learner's cognition state and emotional engagement.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

Scaffolding motivation in e-learning environments could lead to greater self-efficacy and more engaged learners.

“Scaffolding motivation in e-learning environments could lead to greater self-efficacy and more engaged learners.”

• AssumptionRests on an assumption: Integrating motivational scaffolding into e-learning environments can heighten confidence in learners about their memory and numeric abilities.

May combine two claims.

Based on emotional signals, AI can adapt the learning environment, such as by using motivational messages or simplifying the task at moments of frustration of the learner, or gamifying when the learner is bored.

“Based on such emotional signals, AI can adapt the learning environment, such as by using motivational messages or simplifying the task at moments of frustration of the learner, or gamifying when the learner is bored [216].”

• cites [216] Ouyang, 2022, Artificial intelligence in online higher education: A systematic review of empirical research from 2011 to 2020, Educ. Inf. Technol. sourceNot checked yet

  Not checked in the free run.

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

AI algorithms that are trained on data containing biases due to past discriminations will replicate those biases, whether in learning recommendations or updates of assessment.

“For example, AI algorithms that are trained on data containing biases due to past discriminations will replicate those biases, whether in learning recommendations or updates of assessment.”

Probably not a substantial claim: may not pass S4 (based).

Real-world applications of PL and AA, with the integration of AI, have already been able to show measurable improvements in educational outcomes.

“Real-world applications of PL and AA, with the integration of AI, have already been able to show measurable improvements in educational outcomes.”

Probably not a substantial claim: may not pass S4 (based).

This dynamic adjustment ensures that learners remain within their optimal cognitive zone, where they can achieve the best learning outcomes.

“This dynamic adjustment ensures that learners remain within their optimal cognitive zone, where they can achieve the best learning outcomes.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

Adaptive fading, where cognitive load is reduced as learners progress, maintains optimal cognitive engagement and fosters self-regulation.

“Adaptive fading, where the cognitive load is reduced as learners progress, maintains optimal cognitive engagement and fosters self-regulation.”

May combine two claims.

Personalized adaptation in neuroadaptive training can help optimize learning because training does not become too easy or too overwhelming, thus enhancing the rate at which skills are learned and improving their retention.

“Such personalized adaptation can help optimize learning because training does not become too easy or too overwhelming, thus enhancing the rate at which skills are learned and improving their retention, according to [215].”

• cites [215] Giannakos, 2021, Systematic Literature Review of E-Learning Capabilities to Enhance Organizational Learning, Inf. Syst. Front. sourceNot checked yet

  Not checked in the free run.

Classified as probably not a substantial claim. May combine two claims.

AI-driven education has the potential to offer not only personalized content, but also optimized content with respect to cognitive and emotional states that will further enhance learning outcomes.

“This also highlights how AI-driven education has the potential to offer not only personalized content, but also optimized content with respect to cognitive and emotional states that will further enhance learning outcomes.”

Probably not a substantial claim: may not pass S4 (based).

Emotional Intelligence plays an important role in AI systems when dealing with students' affective states, directly influencing motivation and learning processes.

“Emotional Intelligence plays an important role in AI systems when dealing with students' affective states, directly influencing motivation and learning processes.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

AI systems using historical data to inform their decision-making processes might carry over pre-existing biases.

“AI systems using historical data to inform their decisionmaking processes might carry over pre-existing biases.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

AI enhances PL by constantly analyzing students' progress and accordingly adapting the learning path, which might allow students to proceed at their own pace.

“AI enhances PL by constantly analyzing students' progress and accordingly adapting the learning path, which might allow students to proceed at their own pace.”