Conference Agenda

The purpose of this study is to analyze the influence of key competencies in the management of Personal Learning Environments (PLEs) on the academic satisfaction of university students. The research was conducted using a quantitative approach, with a non-experimental, cross-sectional design and an explanatory scope. The sample consisted of university students from a private institution in Peru. The methodology employed an exploratory factor analysis, which was validated through measurement model analysis. Data processing and analysis were performed using PLS-SEM, a structural equation modeling approach that enabled comparisons of direct effects among the model's variables. The results show that information organization, creative collaboration, and participation in communities and discussion groups have direct, positive, and statistically significant effects on academic satisfaction. In contrast, online information searching did not show a significant direct effect on this variable. Furthermore, the structural model presented a moderate level of explanation for academic satisfaction, confirming the multifactorial nature of this construct in university contexts mediated by digital technologies. The main conclusion is that academic satisfaction does not depend solely on access to digital information but is shaped by cognitive and social skills associated with the effective management of PLEs, especially those linked to the organization of information and collaborative interaction.

This study aimed to examine how academic self-efficacy influenced dependence on generative artificial intelligence among Peruvian university students, while testing the mediating role of performance expectancy. A quantitative, non-experimental, cross-sectional
explanatory design was implemented with 430 students from public universities in northern Peru (Piura, Tumbes, Lambayeque, La Libertad–Trujillo, and Cajamarca). Data were collected in November–December 2025 using a structured 5-point Likert questionnaire
measuring academic self-efficacy, AI-related performance expectancy, and AI dependence. The model was estimated through PLS-SEM in SmartPLS and supported strong measurement quality, with high indicator loadings and robust internal consistency, convergent validity, and
discriminant validity. Structural results showed that academic self-efficacy exerted a strong negative direct effect on AI dependence (β = −0.707; p < 0.001) and a positive effect on performance expectancy (β = 0.55; p < 0.001). Performance expectancy, in turn, significantly
increased AI dependence (β = 0.707; p < 0.001) and mediated the relationship between self-efficacy and dependence (indirect β = 0.389; p < 0.001), with substantial explained variance for dependence (R² = 0.45). Overall, self-efficacy functioned as a direct protective factor, yet a
concurrent indirect pathway increased dependence via heightened expectations of academic gains from AI. The study recommended strengthening student autonomy and metacognitive regulation, implementing critical AI literacy, and redesigning assessment tasks and
guidance to encourage responsible, complementary use rather than cognitive substitution.

This study aimed to identify the level of adoption and adaptation of artificial intelligence (AI) tools among students and teachers in technical education at an institute in Lima. Using a quantitative approach, two validated questionnaires were applied to a sample of 118 students and 10 teachers. The instruments measured variables related to knowledge, usage, educational purpose, and critical integration of AI in the learning process. The results showed that students have occasional access to these technologies, but their academic use and adaptability remain limited. Teachers showed a high level of familiarity, although with significant differences in pedagogical integration. It is concluded that, although AI is present in the technical educational environment, its effective incorporation requires institutional strategies to strengthen digital competencies, continuous teacher training, and critical and responsible student engagement.

The implementation of generative artificial intelligence (AI) and Machine Learning (ML) is revolutionizing higher education. These technologies enable more personalized teaching, optimize academic management processes, and promote innovative pedagogical methodologies. This study aims to systematically review the state of the art of AI and ML in the university context between 2015 and 2024, based on 50 articles selected from the Scopus database. A mixed methodology was used that combined bibliometric and content analysis to identify trends, technological tools, and their influence on academic performance. The findings highlight ChatGPT, Microsoft Insights, and Turnitin as the most used generative artificial intelligence tools, contributing to academic writing, plagiarism detection, and educational data analysis. Furthermore, evidence shows that the application of machine learning models, such as Random Forest and Support Vector Machine, favors the prediction of student performance and the detection of at-risk students. Despite significant benefits, challenges persist related to ethics, data privacy, and technology dependence. In conclusion, the integration of AI and machine learning drives efficiency in teaching processes and promotes educational innovation; however, it is essential to develop regulatory frameworks and adaptive pedagogical approaches to ensure its sustainable implementation.

The rapid expansion of digital technologies, higher education expansion and persistent inequalities have turned data into a strategic asset for education systems in Latin America. Universities and schools generate large volumes of data on enrolment, performance, virtual learning and institutional management, yet the systematic use of these data for decision making remains incipient and uneven across the region. This article examines how educational analytics, and specifically learning analytics, can contribute to improving quality, equity and efficiency while acknowledging structural constraints to their adoption. It draws on a narrative review of recent literature, complemented by descriptive analyses of official connectivity indicators, to situate analytics within contexts marked by deep digital and social divides. Evidence is organized at three levels: system level policies and infrastructure, institutional governance and management, and pedagogical practices in classrooms and virtual environments. Across these levels, analytics are understood as socio technical practices that reconfigure how actors perceive teaching, learning and accountability. The findings reveal a growing but fragmented landscape of projects, dashboards, early warning systems and predictive models, concentrated in institutions with stronger digital capacities and revealing gaps in data literacy, teacher development, ethical frameworks and regulation. The article argues that analytics can reinforce exclusion or become a lever for inclusion, depending on how they are embedded in broader strategies for digital transformation and social justice, and concludes by outlining context sensitive design principles that align data with educational purposes and local inequalities.

Abstract: Introduction: Through the design of activities related to active methodologies to promote challenging learning, research, creativity and greater prominence of students in their learning experience, the innovation and creativity project with Canvas Studio arises through which It is proposed to innovate the forms of evaluation in the courses. Methods: The project consisted of the adaptation of the learning activities, integrating elements of challenging learning and digital tools that are part of Canvas and that are available to the student, for this the first courses and their teachers were identified, general meetings were organized and specific to each course, teachers were guided and monitored in the application of active methodologies. Results: As a result of this process, 40% of the evaluations in 41 current courses were updated and piloted in 29 courses with the participation of 3,058 students. In addition, a pilot was carried out in 2 completely new courses and with 100% of evaluations under this approach, with the participation of 349 students. Conclusion: The application of this educational innovation through Canvas Studio allowed the teacher to encourage participation, interaction and active communication in the course, identify and direct students at key points through multimedia elements and identify the level of understanding procedures and exercise development.

Keywords: Learning, active methodologies, evaluations