INTELLIGENCE-BASED ADAPTIVE LEARNING ENVIRONMENTS IN HIGHER EDUCATION

Mazkur tadqiqot oliy ta‘lim tizimida sun‘iy intellektga asoslangan adaptiv o‘quv muhitlarining talabalarning kognitiv imkoniyatlarini rivojlantirish va baholashdagi rolini o‘rganishga bag‘ishlangan. Tadqiqotda AI texnologiyalari yordamida individual ta‘lim trayektoriyalarini shakllantirish, bilimlarni o‘zlashtirish jarayonini optimallashtirish hamda kognitiv jarayonlarni (diqqat, xotira, tafakkur va metakognitsiya) rivojlantirish mexanizmlari tahlil qilinadi. Shuningdek, adaptiv o‘quv muhitlarining kognitiv samaradorligini baholash modeli ishlab chiqiladi. Tadqiqot natijalari oliy ta‘limda raqamli transformatsiyani kuchaytirish va ta‘lim sifatini oshirishga xizmat qiladi.

INTELLIGENCE-BASED ADAPTIVE LEARNING ENVIRONMENTS IN HIGHER EDUCATION Abdurazoqov Boburjon Sobirjon o‘g‘li PhD, Associate Professor at Samarkand State Pedagogical Institute https://doi.org/10.5281/zenodo.21444029 Abstract: This study examines the role of AI-based adaptive learning environments in higher education in assessing and enhancing students’ cognitive capabilities. It analyzes mechanisms of personalized learning pathways enabled by artificial intelligence, optimization of knowledge acquisition processes, and the development of cognitive functions such as attention, memory, thinking, and metacognition. Additionally, a model for evaluating the cognitive effectiveness of adaptive learning environments is proposed. The findings contribute to strengthening digital transformation in higher education and improving educational quality. Keywords: artificial intelligence, adaptive learning environment, cognitive capabilities, higher education, digital pedagogy, personalized learning, metacognition, educational technologies, learning analytics, modeling.

OLIY TA’LIMDA SUN’IY INTELLEKTGA ASOSLANGAN ADAPTIV O‘QUV MUHITLARINING KOGNITIV IMKONIYATLARINI BAHOLASH Abdurazoqov Boburjon Sobirjon o‘g‘li Samarqand davlat pedagogika instituti PhD, dotsenti Annotatsiya: Mazkur tadqiqot oliy ta’lim tizimida sun’iy intellektga asoslangan adaptiv o‘quv muhitlarining talabalarning kognitiv imkoniyatlarini rivojlantirish va baholashdagi rolini o‘rganishga bag‘ishlangan. Tadqiqotda AI texnologiyalari yordamida individual ta’lim trayektoriyalarini shakllantirish, bilimlarni o‘zlashtirish jarayonini optimallashtirish hamda kognitiv jarayonlarni (diqqat, xotira, tafakkur va metakognitsiya) rivojlantirish mexanizmlari tahlil qilinadi. Shuningdek, adaptiv o‘quv muhitlarining kognitiv samaradorligini baholash modeli ishlab chiqiladi. Tadqiqot natijalari oliy ta’limda raqamli transformatsiyani kuchaytirish va ta’lim sifatini oshirishga xizmat qiladi. Kalit so‘zlar: sun’iy intellekt, adaptiv o‘quv muhit, kognitiv imkoniyatlar, oliy ta’lim, raqamli pedagogika, individual ta’lim, metakognitsiya, ta’lim texnologiyalari, o‘quv analitikasi, modellashtirish.

ОЦЕНКА КОГНИТИВНЫХ ВОЗМОЖНОСТЕЙ АДАПТИВНЫХ ОБРАЗОВАТЕЛЬНЫХ СРЕД НА ОСНОВЕ ИСКУССТВЕННОГО ИНТЕЛЛЕКТА В ВЫСШЕМ ОБРАЗОВАНИИ Абдуразоков Бобуржон Собиржон угли PhD, доцент Самаркандского государственного педагогического института Аннотация: Данное исследование посвящено изучению роли адаптивных образовательных сред, основанных на искусственном интеллекте, в оценке и развитии когнитивных возможностей студентов в системе высшего образования. В работе анализируются механизмы формирования индивидуальных образовательных траекторий с использованием технологий ИИ, оптимизация процесса усвоения знаний, а также развитие когнитивных процессов (внимание, память, мышление и метакогниция). Также разрабатывается модель оценки когнитивной эффективности адаптивных образовательных сред. Результаты исследования направлены на усиление цифровой трансформации высшего образования и повышение качества обучения.

Ключевые слова: искусственный интеллект, адаптивная образовательная среда, когнитивные возможности, высшее образование, цифровая педагогика, индивидуальное обучение, метакогниция, образовательные технологии, learning analytics, моделирование.

INTRODUCTION Recent research on artificial intelligence (AI) in education highlights a rapid shift toward adaptive learning environments that personalize instruction based on learners’ cognitive profiles, performance data, and behavioral patterns. Scholars such as Holmes et al. (2019) and Luckin et al. (2016) emphasize that AI-powered systems can enhance learning efficiency by dynamically adjusting content difficulty, pacing, and feedback, thereby supporting individualized learning trajectories in higher education. A significant body of literature focuses on adaptive learning systems (ALS) and their impact on cognitive development. These systems utilize machine learning algorithms to analyze student interactions and provide tailored instructional content. Studies show that such environments improve knowledge retention, problem-solving skills, and metacognitive awareness compared to traditional instruction models. However, researchers also note challenges related to data privacy, algorithm transparency, and pedagogical alignment. From a cognitive perspective, theories such as Cognitive Load Theory (Sweller, 1988) and Constructivist Learning Theory (Piaget, Vygotsky) provide a theoretical foundation for understanding how adaptive systems influence learning processes. AI-based systems reduce extraneous cognitive load by structuring information efficiently while promoting active knowledge construction through interactive tasks and simulations. MATERIALS AND METHODS Research purpose. The purpose of this study is to investigate and model the cognitive capabilities of students in higher education within AI-based adaptive learning environments. It aims to analyze how artificial intelligence technologies personalize learning processes, optimize knowledge acquisition, and influence key cognitive functions such as attention, memory, critical thinking, problem-solving, and metacognitive skills. Research tasks: 1. To analyze the theoretical and methodological foundations of AI-based adaptive learning environments in higher education and their role in cognitive development. 2. To identify and systematize key cognitive domains affected by adaptive learning systems, including attention, memory, critical thinking, problem-solving, and metacognitive skills. 3. To develop criteria and indicators for assessing students’ cognitive capabilities within AI-driven adaptive educational environments. 4. To design a conceptual model for evaluating the cognitive effectiveness of adaptive learning systems based on artificial intelligence. 5. To validate the proposed model through analytical methods and provide practical recommendations for implementing AI-based adaptive learning environments in higher education institutions. Research methodology. This study is based on an interdisciplinary methodological framework combining educational science, cognitive psychology, artificial intelligence in education, and data-driven learning analytics. The research adopts constructivist and competency-based approaches, emphasizing the active role of learners in constructing knowledge within AI-supported adaptive environments.

The methodological foundation also incorporates cognitive load theory, information processing theory, and personalized learning theory, which explain how adaptive systems influence cognitive development and optimize learning efficiency in higher education contexts. Research methods. This study uses a combination of theoretical, empirical, and analytical research methods to investigate students’ cognitive capabilities in AI-based adaptive learning environments. Theoretical methods include analysis, synthesis, and comparative study of scientific literature on artificial intelligence in education, adaptive learning systems, and cognitive development theories. Empirical methods involve pedagogical observation, surveys, cognitive testing, expert evaluation, and an experimental design with pre-test and post-test control and experimental groups to measure learning outcomes. The table systematically classifies the research methods employed in the study. It presents theoretical, empirical, experimental, analytical-statistical, and digital (AI-based) methods as interconnected components of the research design. Each group of methods serves a specific stage of the investigation: theoretical methods provide the scientific foundation, while empirical methods enable the examination of real educational processes and learner behaviors. Experimental methods are used to test the practical impact of AI-based adaptive learning environments on cognitive development. Analytical and statistical methods play a crucial role in processing collected data and generating scientific conclusions. At the same time, digital approaches such as learning analytics and data mining make it possible to analyze students’ learning activities in real time. Collectively, these methods contribute to the reliability, objectivity, and scientific validity of the research findings. Table 1. Innovative Cognitive Assessment Criteria For Ai-Based Adaptive Learning Environments In Higher Education Assessment Innovative Measurement Representation in Dimension Criterion Indicator AI/Adaptive Systems Duration of learner Attention Focus Stability Eye-tracking, interaction attention Management Index time (seconds/minutes) Time required to Cognitive Adaptive System response and adjust to task Adaptation Response Speed learner adaptation rate complexity Decision- Algorithmic Percentage of correct Scenario-based AI tasks Making Choice Quality decisions (%) Metacognitive Self-Correction Number of detected AI feedback correction Control Coefficient and corrected errors loop Ratio between task Optimal Load Cognitive load estimation Cognitive Load complexity and Index model performance Digital Interactive Number of solution Multi-path AI problem Intelligence Solution Strategy pathways used solving Continuous Learning Dynamics of platform Learning analytics Engagement Analytics activity dashboard Profile

Assessment Innovative Measurement Representation in Dimension Criterion Indicator AI/Adaptive Systems Number of original Creative Generative AI-generated task responses and Thinking Solution Quality evaluation alternatives Emotional Emotional Changes in stress and Sentiment and behavior Cognition Stability Index focus levels analysis Application of Knowledge Transfer Cross-context AI knowledge in new Transfer Coefficient assessment situations The table represents an innovative approach to assessing the cognitive capabilities of AI-based adaptive learning environments in higher education. Unlike traditional assessment systems that primarily focus on measuring knowledge acquisition, this framework evaluates learners’ real-time cognitive behaviors, adaptability, decision-making strategies, and metacognitive activities through a comprehensive and multidimensional perspective.

First, the Attention Management criterion, represented by the Focus Stability Index, measures how long students can maintain concentration on a specific learning task. This indicator can be assessed through eye-tracking technologies or platform interaction statistics. Although often overlooked in conventional assessment methods, sustained attention is a critical factor in cognitive performance and learning effectiveness.

The Cognitive Adaptation criterion evaluates the speed at which students adjust to tasks of varying complexity. AI systems continuously analyze learner behavior and automatically modify task difficulty according to individual performance. This adaptive capability is one of the key advantages of intelligent learning environments, allowing for personalized educational experiences.

The Decision-Making Quality criterion measures students’ ability to make accurate and well-reasoned decisions in different academic or problem-solving situations. Through AI-driven scenario-based tasks, learners are presented with multiple options, and their choices are analyzed algorithmically to determine the effectiveness of their reasoning processes. Metacognitive Control assesses students’ capacity to evaluate their own knowledge, identify mistakes, and implement corrective actions. As a fundamental component of self-regulated learning, metacognitive awareness is increasingly recognized as a critical competency in modern education. AI systems monitor this process through feedback loops and adaptive guidance mechanisms.

The Optimal Cognitive Load Index determines whether the level of mental effort required by learning activities remains within an effective range. Excessive cognitive load may reduce learning efficiency, whereas insufficient challenge may limit intellectual development. Adaptive learning systems can dynamically regulate task complexity to maintain an optimal balance between challenge and performance.

The Digital Intelligence criterion and Interactive Solution Strategies evaluate learners’ ability to solve problems using multiple approaches and digital resources. This reflects creative and divergent thinking in technology-enhanced learning environments and demonstrates learners’ capacity to navigate complex digital tasks effectively.

The Learning Analytics dimension provides real-time indicators of cognitive engagement by analyzing platform activity, including login frequency, learning duration, and task completion patterns. Such data offer valuable insights into student participation and learning behaviors that may not be visible through traditional assessments. The Creative Thinking criterion focuses on the originality and diversity of learnergenerated responses. AI-based evaluation systems can assess the novelty of ideas and the variety of alternative solutions proposed by students, thereby supporting the development and measurement of innovation-related competencies. The Emotional Cognition criterion examines psychological factors such as stress, motivation, emotional regulation, and attentional stability. Through sentiment analysis and behavioral tracking technologies, AI systems can identify emotional states that may influence learning outcomes and provide timely interventions when necessary. Finally, Knowledge Transfer, represented by the Transfer Coefficient, evaluates students’ ability to apply acquired knowledge and skills in unfamiliar contexts. This criterion is considered one of the highest indicators of cognitive development because it reflects deep understanding and the capacity to generalize learning across different situations. Overall, this innovative assessment framework positions AI-based adaptive learning environments not merely as tools for delivering educational content but as intelligent systems capable of analyzing cognitive processes in depth. Such an approach provides a scientific foundation for implementing personalized learning pathways, real-time monitoring, predictive assessment, and data-driven educational decision-making in higher education.

Figure 1. Innovative assessment criteria for cognitive capabilities in AI-based adaptive learning environments in higher education The proposed framework presents a comprehensive system of ten unconventional assessment criteria designed to evaluate cognitive capabilities in AI-based adaptive learning environments within higher education. Unlike traditional assessment approaches that primarily focus on academic achievement, this model emphasizes a broader evaluation of learners’ cognitive, behavioral, emotional, and creative competencies. The integration of artificial intelligence technologies enables continuous monitoring, analysis, and adaptation of the learning process according to individual learner characteristics.

The first criterion, Attention Management, evaluates the learner’s ability to maintain concentration, sustain focus over time, and resist distractions during learning activities. The second criterion, Cognitive Adaptability, measures how effectively students adjust to changing task requirements, respond to varying levels of difficulty, and recover from mistakes. The third criterion, Decision-Making Quality, assesses the accuracy and effectiveness of decisions made in problem-solving situations, as well as the learner’s capacity to justify and select optimal solutions.

The fourth criterion, Metacognitive Control, focuses on learners’ ability to monitor their own understanding, identify errors, and regulate their learning strategies. The fifth criterion, Cognitive Load Management, examines how efficiently learners process information while maintaining an optimal balance between mental effort and performance. The sixth criterion, Digital Intelligence, reflects the capacity to employ multiple problem-solving strategies, utilize digital resources effectively, and demonstrate strategic thinking in technology-enhanced learning environments.

The seventh criterion, Learning Analytics, measures engagement through indicators such as participation frequency, task completion patterns, and consistency of involvement in educational activities. The eighth criterion, Creative Thinking, evaluates originality, divergent thinking, and the ability to generate innovative solutions to complex problems. The ninth criterion, Emotional Cognition, examines emotional stability, motivation management, and the influence of emotions on learning performance. Finally, the tenth criterion, Knowledge Transfer, assesses the learner’s ability to apply acquired knowledge in new contexts, transfer skills across domains, and solve unfamiliar problems using previously learned concepts.

Collectively, these ten criteria provide a multidimensional framework for assessing cognitive development in AI-supported adaptive learning systems. By combining learning analytics, behavioral data, and intelligent assessment tools, the model offers a more accurate and personalized understanding of student performance. As a result, it supports the development of individualized learning pathways, enhances educational effectiveness, and contributes to the advancement of data-driven decision-making in higher education.

RESULTS The research resulted in the development of an innovative framework consisting of ten assessment criteria for evaluating cognitive capabilities in AI-based adaptive learning environments in higher education. The proposed model expands traditional assessment approaches by incorporating cognitive, metacognitive, behavioral, emotional, and creative dimensions of learning. Through the integration of artificial intelligence technologies, the framework enables continuous monitoring of learner performance and provides personalized insights into individual learning processes. The findings indicate that Attention Management, Cognitive Adaptability, and Decision-Making Quality serve as fundamental indicators of learners’ cognitive performance in adaptive educational settings. Students demonstrating higher levels of concentration, adaptability, and decision-making accuracy showed improved learning outcomes and greater success in completing complex tasks. Furthermore, the results highlight the importance of Metacognitive Control and Cognitive Load Management, which contribute significantly to self-regulated learning and efficient information processing.

The study also revealed that AI-driven analytics can effectively measure engagement and learning behaviors through criteria such as Learning Analytics and Digital Intelligence. These indicators provide valuable information regarding student participation, interaction patterns, and strategic problem-solving abilities. In addition, the inclusion of Creative Thinking and Knowledge Transfer allows educators to assess learners’ capacity to generate original ideas and apply acquired knowledge in new and unfamiliar contexts, which are essential competencies for the twenty-first century. Another significant outcome of the research is the recognition of Emotional Cognition as a critical factor influencing academic performance. The results suggest that emotional stability, motivation, and effective emotional regulation positively affect learning engagement and achievement. AI-supported sentiment analysis and behavioral tracking tools can provide timely feedback for identifying learners who may require additional support. Overall, the research demonstrates that the proposed ten-criterion framework offers a comprehensive and multidimensional approach to cognitive assessment. The model enhances the accuracy of learner evaluation, supports personalized adaptive learning pathways, and facilitates data-driven educational decision-making. These findings confirm the potential of artificial intelligence technologies to improve the quality, effectiveness, and inclusiveness of higher education learning environments.

DISCUSSION The findings of this study support the growing view that traditional assessment methods are insufficient for evaluating the complex cognitive demands of modern digital learning environments. While conventional assessments primarily focus on knowledge acquisition and academic achievement, the proposed framework introduces a multidimensional perspective that captures cognitive, behavioral, emotional, and creative aspects of learning. This broader approach aligns with contemporary educational theories emphasizing learner-centered and competency-based education.

The results suggest that cognitive performance in AI-based adaptive learning environments cannot be adequately explained by a single indicator. Instead, effective learning emerges from the interaction of multiple factors, including attention regulation, adaptability, metacognitive awareness, and emotional stability. The inclusion of Attention Management and Cognitive Adaptability reflects the importance of maintaining focus and responding effectively to dynamic learning conditions. These findings are consistent with previous studies demonstrating that learners who adapt quickly to changing educational challenges tend to achieve higher levels of academic success.

The study also highlights the significance of Metacognitive Control and Cognitive Load Management in supporting self-regulated learning. Learners who are able to monitor their understanding, identify mistakes, and manage mental effort efficiently are more likely to engage in deeper learning processes. Artificial intelligence technologies provide unique opportunities to monitor these dimensions in real time, enabling adaptive systems to offer timely interventions and personalized support.

Another important contribution of the framework is the integration of Creative Thinking and Knowledge Transfer as assessment dimensions. These competencies are increasingly recognized as essential outcomes of higher education, particularly in knowledge-based economies where innovation and problem-solving skills are highly valued. By evaluating learners’ ability to generate original ideas and apply knowledge across different contexts, the framework extends beyond traditional measures of academic performance and addresses broader educational goals.

This study proposed a novel framework for assessing cognitive capabilities in AI-based adaptive learning environments in higher education. The framework consists of ten interconnected assessment criteria: Attention Management, Cognitive Adaptability, Decision-Making Quality, Metacognitive Control, Cognitive Load Management, Digital Intelligence, Learning Analytics, Creative Thinking, Emotional Cognition, and Knowledge Transfer. Together, these dimensions provide a comprehensive and multidimensional approach to evaluating learner performance beyond traditional academic measures. The findings demonstrate that artificial intelligence technologies can support more accurate, personalized, and continuous assessment of students’ cognitive development. By integrating behavioral data, learning analytics, and adaptive assessment mechanisms, the proposed model enables educators to identify individual strengths and weaknesses, monitor learning progress, and provide targeted instructional support. The framework also highlights the importance of higher-order cognitive skills, creativity, emotional regulation, and knowledge application in modern educational contexts. Furthermore, the study emphasizes the potential of AI-driven adaptive learning systems to enhance educational effectiveness through personalized learning pathways and data-informed decision-making. The proposed criteria contribute to a deeper understanding of learner behavior and support the development of more responsive and inclusive learning environments.