Development of a Personalized Gamified Mobile Learning App Using an AI-Based Virtual Assistant for Financial Literacy

Development of a Personalized Gamified Mobile Learning App Using an AI-Based Virtual Assistant for Financial Literacy | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Development of a Personalized Gamified Mobile Learning App Using an AI-Based Virtual Assistant for Financial Literacy Angie Nayeli Ruiz-Carhuamaca, Juliana Alexandra Yauricasa-Seguil, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7170120/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This study presents the implementation of a mobile application that integrates gamification and artificial intelligence to strengthen financial literacy in college students. The proposed solution combines a user-centered approach, educational and structured content, and interactive challenges, integrating a virtual assistant powered by Gemini AI. The application is composed of three stages: an initial diagnostic assessment, gamified challenges, and a final assessment. The system offers personalized feedback, financial topics with levels of difficulty, and a system of achievements and rewards, which seeks to enhance engagement and learning outcomes. The development stage was carried out under the Scrum framework over eight sprints, where Unity and FastAPI were used for frontend and backend development, respectively. On the other hand, usability and user experience were evaluated using two instruments: the System Usability Scale (SUS) and the User Experience Questionnaire (UEQ). The results obtained from the tests with 50 university students show a high usability (SUS = 91.9) and a positive experience in all the dimensions evaluated by the UEQ, which confirms the effectiveness of the application. This work contributes to the field of educational technologies by providing a scalable and validated architecture for financial education through mobile learning. Special Education Finance Mobile Learning Gamification Financial Literacy Artificial Intelligence Virtual Assistant Personalized learning Educational Technology Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 1. Introduction The economic environment is becoming increasingly complex; therefore, financial education, which is an essential component in this area, must provide comprehensive training for young people that is directly proportional to the complexity of the financial system. However, several studies have shown deficiencies in the level of financial literacy among university students, which demonstrates a deficient education. In (Ruiz et al., 2024), the level of financial knowledge of university students from various Peruvian universities was analyzed, and the results showed that there is a lack of basic financial concepts, which limits their ability to correctly manage their financial resources and achieve economic planning for the future. Likewise, in other international contexts, they have also reported low levels of financial literacy as in (Torma et al., 2023), who found that after a financial literacy study of university students in Croatia, more than 70% indicated that they were rarely or seldom informed about finances and two thirds of said survey stated that they were not familiar with financial concepts, which showed that the level of financial literacy of said sample is moderately low. On the other hand, the development of innovative solutions to provide quality education is valued. As reported by Ruiz (2024), 52% of students prefer learning methods based on interactive games, and 81% value receiving personalized recommendations. This drives the development of innovative solutions such as the proposed solution, which incorporates a form of gamification to promote user engagement. This approach achieves the objectives of the applications, since implementing game mechanics, such as reward systems, challenges, achievements, and interactive messaging, has proven to be effective in encouraging user participation and solving problems to meet game goals (French et al., 2021). It should be noted that students who used financial applications with gamified elements showed better results in achieving financial goals, such as saving money, due to the motivation generated by the rewards, which highlights the value of personalizing motivation in digital environments (French et al., 2021). Concerning the inclusion of financial topics in educational institutions, there are still significant gaps in the effectiveness of the methodologies they employ, because despite the inclusion of financial content in their curricula, they lack dynamic strategies to capture and maintain the interest of young people. Even though financial education courses exist, they tend to be optional, of short duration, and do not always manage to improve financial wellbeing or reduce students' financial stress, since the content is limited and of little impact (Robb & Chy, 2023). Studies reveal that many financial education initiatives maintain the traditional teaching approach of addressing financial topics such as money management, with simplified and poorly contextualized methods that do not address the realities and challenges of the students' real environment (Björklund & Sandahl, 2023). This lack of contextualization and dynamism reduces student interest and participation. In this sense, research such as (Sinnewe & Nicholson, 2023) highlights that financial education that is more aligned with the real environment of young people contributes directly to developing better financial habits in adulthood. This reinforces the need to redesign existing methodologies and integrate emerging technologies that allow personalizing the learning experience according to the profile of each young person so that they feel more motivated and committed to learning. Regarding the playful approach to financial education. It was shown that the use of board games in educational contexts can significantly improve the level of knowledge of university students, as well as present better behavior and attitude towards financial challenges (Reisdorfer-da-Silva et al., 2025). Likewise, there is a positive impact on game mechanics and reflection prompts as demonstrated in the financial game Moonshot, which uses the combination of these elements to significantly increase the useful value perceived by users, emphasizing the importance of the instructional design of these elements (Platz & Zauner, 2025). Similarly, the traditional game Monopoly, which was adapted as a financial learning tool, managed to stimulate critical reflection on financial decisions, irrationality in investing, and financial behavior, which shows that well-designed games manage to facilitate the learning of deep-rooted topics and that it is long-lasting (Lew & Saville, 2021). Thus, the ludic approach allows the development of decision-making skills in an entertaining and high-impact way for users. In this scenario, the need arises to design solutions that combine playful approaches with technological tools to strengthen financial literacy among young university students, such as the use of mobile applications, gamification, AI, and personalized recommendations. This is how the proposed solution is presented as an innovative alternative to meet this demand. Thus, strengthening the financial competencies of students effectively and sustainably. It should be noted that, in a previous study, the conceptual design of this solution was presented, where the design of the proposal was based on the results obtained from the survey conducted on the level of financial knowledge and learning preferences, resulting in the wireframes and mockups of the application. In this context, the main objective of this article is to give continuity to that initial proposal, documenting the technical development process and implementation of the application. It describes the use of the agile Scrum methodology to organize the project in development sprints, the design of the physical and logical architecture of the system, the database, and the integration of the AI-based virtual assistant. In addition, it includes the results of the validation phase with beta users, which was obtained by using the tools: System Usability Scale (SUS) and User Experience Questionnaire (UEQ). To evaluate the usability perceived by the user, the SUS was used, which is a simple 10-item scale that allows a global view of the subjective evaluations of usability, where each item has a score that is used to calculate the SUS score ranging from 0 to 100 (Brooke, 1996). In contrast, to measure not only technical aspects, but also emotional and perceptual components about user experience (attractiveness, pragmatic quality, hedonic quality) the UEQ was used, which is a quick and direct measurement of user experience where each item consists of a pair of terms with opposite meanings, with a total of 26 items grouped into 6 scales (Schrepp et al., 2017). These tools complement each other; by combining them, it is possible to obtain a comprehensive view of the system, both the efficiency of use and the subjective experience of the users. The article is organized as follows: the methodology section describes the development approach, the tools used, and the architectural design; the results section shows the development of the solution. Finally, the discussion section interprets the findings, the main limitations, and the comparison of results with other studies, and concludes by naming the main contributions and proposals for future research. 2. Background information 2.1 Mobile Learning Applications in Education (Hassan et al., 2023 ) presents an adaptive usability model for mobile learning applications, focused on students with cognitive difficulties. Their proposal seeks to improve the mobile learning experience through a hybrid and personalized recommendation system. For this purpose, a systematic usability analysis and evaluation of applications developed in HTML was used. The results showed that there are significant deficiencies in the existing applications of the study group, and a more accessible and personalized navigation model was suggested. However, the focus of this study is limited to students with special abilities and does not consider the university population. (Frisancho et al., 2023 ) analyzed the impact of a behavioral intervention based on a mobile application on financial literacy and behavior for young Peruvian high school graduates. The intervention combined an application to record transactions, biweekly SMS messages, and periodic visits for 27 weeks. The results showed significant improvements in financial knowledge and price awareness, although there was no change in savings and budgeting habits. Data from credit bureaus revealed an increase in credit usage and subsequent financial inclusion. On the other hand, the design did not include explicit educational content but promoted self-reflection and financial curiosity. Thus, it shows that simple digital tools, combined with behavioral nudging strategies, can partially substitute traditional financial education in contexts of difficult access and vulnerability. (Yánez-Pérez et al., 2024 ) presents the design, development, and evaluation of an educational mobile application (IndagApp) that facilitates science teaching through school inquiry methodology. The artifact was developed under an iterative approach of continuous improvement by applying usability tests such as SUS, UES (User Engagement Scale), and MAU (Mobile Application Usability). The results showed an adequate usability of IndagApp, which is a resource that captures attention, generates interest and enjoyment, and is a useful tool for teachers. Although the study is not based on financial education, it provides valuable methodological input for the development of user-centered educational mobile applications. However, the study does not explore metrics of effectiveness in financial literacy. In (Asmawi et al., 2024 ), the development of BlockScholar, a mobile educational application focused on blockchain technology using gamification, is presented. It was based on the Rapid Application Development (RAD) model, which encourages collaboration between developers and end users to accelerate the development process. The results show that the combination of game-based learning and mobile technology is successful in increasing participation and retention of information among young people. However, the study does not include long-term follow-up metrics on the impact of game-based learning on the development of cognitive skills and knowledge retention in youth. 2.2 Gamification in Educational Technology (Grijalvo et al., 2022 ) presents a gamified learning framework based on the principles of Mechanics, Dynamics, and Emotions (MDE) for the integration of digital business games in university contexts. It is supported by two strategic simulators: Gestionet and Global Management Challenge - GMC, which allows the development of financial competencies and soft skills. We addressed how to effectively integrate gamified tools into the formal curriculum, promoting the acquisition of financial knowledge and the development of competencies demanded by the labor market. The results highlight that intrinsic motivation and self-motivation influence the improvement of competencies, while participation in simulators increases satisfaction and recommendation intention. However, the study is limited by the single institutional context and does not explore the integration of dynamic personalization or AI. Nor do they address learning outcomes in terms of effective acquisition of financial content. (Pitthan & Witte, 2025) analyzed how gamification can increase completion rates in adaptive learning environments using a financial simulation tool for adults. The study validates a theoretical model that integrates gamification with content personalization and gamification elements. The results demonstrate the importance of psychological factors as mediators of user retention, providing evidence that the strategic inclusion of gamified elements during critical phases of learning can significantly increase retention rates. This reinforces the need to integrate adaptive dynamics and gamification at key moments in the user's journey. In (Lisana et al., 2025 ) present the design, implementation, and evaluation of a gamified educational application to promote financial literacy in young university students. The developed video game simulates realistic economic decisions and is supported by immersive narratives, instant feedback, progressive challenges, and virtual rewards. The results showed a significant improvement in financial literacy scores after using the game, as well as an assessment of its usefulness and attractiveness. However, the sustained impact on financial behavior is not evaluated, nor are the mechanisms for personalization or adaptation of the content according to the user's profile explored. In (Imam et al., 2022 ), a gamified quiz video game is developed to teach technical concepts of corporate finance in an Australian university context. It was developed under the “cognitive apprenticeship” model and seeks to facilitate active learning through questions with random values, variable scores, and immediate feedback, promoting the transfer of knowledge to real-world scenarios. The evaluation was based on a structured perception questionnaire, with Bayesian regression analysis revealing a high relationship between game usability, learning perception, and overall satisfaction. Furthermore, it reinforces the idea that effective gamified environments should focus not only on content but also on user experience, professional relevance, and cognitive load reduction. 2.3 AI-Based Virtual Assistants for Learning (Hean et al., 2025 ) investigate the ability of large-scale language models (LLMs) to provide personalized and reliable financial advice. A systematic comparative analysis of multiple LLMs-ChatGPT (3.5, 4, 4o), Claude (3 Haiku, 3.5 Sonnet, 3 Opus), Gemini, and LLaMA-was conducted using real financial literacy tests such as Money Counts and NFEC. The study reveals that the most advanced models exceeded 79% accuracy on topics such as insurance and student loans, whereas they failed on financial psychology. Although the pedagogical quality of the answers and adaptive personalization mechanisms were not evaluated, it is shown that LLMs can function as predictive agents that adapt their performance according to topic difficulty. This reinforces their potential use as virtual assistants in gamified educational applications to provide adaptive support, assess prior knowledge, and generate dynamic content. (Mabwe et al., 2025 ) examines the role of generative artificial intelligence (GenAI) chatbots, such as ChatGPT, Microsoft Bing, and Google Bard (Gemini), in supporting investment decision making through standardized prompts about stocks, ETFs, diversification, and market trends. The results indicate medium-to-high performance and a tendency to avoid risky decisions. However, they also made significant errors, such as references to non-existent assets. Although the study focuses on investment contexts, its findings are relevant, as they show the ability of these systems to generate relevant and specific content. However, the content analysis does not consider the level of understanding or bias of the language used by the chatbots, a crucial element if these systems are to be applied to educational or gamified contexts. (Erdem et al., 2025 ) evaluates the literature accuracy of GenAI-based virtual assistants (ChatGPT-4o, o1-preview, and Gemini Advanced), focusing on their application for financial literature reviews. The results reveal that ChatGPT-4o and o1-preview had hallucination rates of 20.0% and 21.3% respectively (binary scale). In contrast, Gemini Advanced had a significantly higher rate: 76.7% (binary). Also, a higher incidence of hallucinations was evidenced in “new” subjects. This study is highly relevant because it warns about the risks of blindly trusting content generated by LLMs without validation and curation processes. These findings underscore the need to incorporate layers of automated verification, bibliographic traceability, and explainability in conversational agents, especially when promoting learning based on academic sources. 2.4 AI for Financial Intelligence: Educational Implications In (Qatawneh et al., 2024 ), the impact of AI on financial decision-making is analyzed, highlighting the mediating role of financial technologies (Fin-Tech). A structural model is proposed that examines how seven AI techniques-NLP, machine learning algorithms, computer vision, predictive analytics, robotic process automation (RPA), blockchain, and deep learning-influence financial decision making, mediated by Fin-Tech tools. Their findings confirm that AI has a significant impact on decision-making, in addition to an indirect effect mediated by Fin-Tech, which showed statistically significant mediation, underscoring its role as a catalyst for the adoption of AI. While the study provides an integrative framework and solid empirical evidence, its applicability is limited by the sample size. (Sujith et al., 2022 ) Examine the key components of machine learning in informed financial decision making and explore how machine learning can optimize decision making within data-intensive business environments. They used surveys of employees and industry leaders and analyzed the impact of the use of Machine Learning models. The results showed that these technologies allow the identification of useful patterns to make more efficient decisions. However, although this study demonstrates the potential of using AI for decision making, its application is limited to the corporate environment and not to educational environments. (Akour et al., 2024 ) Investigates the impact of various dimensions of AI, such as natural language processing (NLP), machine learning, expert systems, and computer vision, on financial decision making in pharmaceutical companies in Jordan. They used surveys and applied them to accounting and financial professionals and conducted an analysis using structural models. As a result, they identified positive effects of AI on business decision making, making business decisions more efficient and rational. However, the approach is limited to the corporate environment and does not consider the use of AI to provide personalized education to the user to develop financial skills. In (Acharya et al., 2024 ), a framework for developing explainable and fair machine learning models applied to financial and real estate contexts, such as loan approval and housing price prediction, is proposed. Advanced algorithms such as LightGBM, XGBoost, transparency techniques such as SHAP, and intersectional fairness techniques such as Calibrated Equalized Odds and Intersectional Fairness were used. The results showed that it is possible to achieve a balance between accuracy, transparency, and fairness, although fairness implied a slight reduction in predictive performance. However, the study provides valuable insights into responsible practices in decision-making processes but does not address adaptive interactions for learning. 2.5 Financial Literacy Education and Digital Interventions (Malik, 2023 ) Conducted a content analysis of 163 financial education mobile apps identified in Google Play and App Store. They evaluated 13 key features related to instructional design, Nielsen heuristics, and personalization principles. As a result, they found that 82% of apps included personalized education and 49% applied gamification. In addition, principles of meaningful learning such as personalization, pre-training, and multimedia were analyzed. However, the study shows a high variability in pedagogical quality, little systematic validation of educational effectiveness, and limited integration of emerging technologies such as AI. This highlights the need to implement more robust frameworks for designing effective mobile financial applications, where not only interaction but also didactic structure and adaptive personalization are considered. (Reisdorfer-da-Silva et al., 2025 ) evaluated the use of board games as a financial education teaching strategy in public schools in Brazil using a quasi-experimental methodology and propensity score matching. The study focused on training teachers and applying games in the classroom, which resulted in significant improvements in financial knowledge, behavior, and attitudes. It is supported by theoretical frameworks that highlight the effectiveness of active learning and the role of financial preferences in adolescence. Although they demonstrated the potential of interactive games, their use is restricted to the school context and lacks adaptive mechanisms, and neither long-term follow-up strategies nor integration with digital environments are explored. (Liu et al., 2024 ) developed the Collaborative Structure Search Framework (CSSF) algorithm to optimize personalized learning paths using big data and AI. It used real educational datasets and graph optimization techniques to identify optimal learning sequences. As a result, they obtained improvements in accuracy (F1-score) over traditional methods. However, their proposal is focused on offering personalized educational trajectories in real time without considering aspects of motivation, user engagement, nor is it focused on financial education. However, their contributions are relevant for the design of intelligent systems capable of offering personalized educational trajectories. (Blanco et al., 2023 ) presents a community-based digital intervention aimed at improving financial capability and reducing financial stress among low- and middle-income Hispanic adults. Mind Your Money (MYM), a digital financial education program designed with a participatory and culturally tailored approach, was developed and validated based on content from the CFPB's Your Money, Your Goals program. It sought to address the gap in financial literacy and well-being with mobile technologies and behavioral change strategies based on “nudges” (reminders, incentives, and follow-up phone calls). The results show significant effects on financial capability and self-efficacy. In addition, a 17% reduction in financial stress levels and a 66% reduction in retention rates are reported, reinforcing the feasibility of sustained digital interventions. However, the design does not allow isolation of the individual effect of each nudge, and the sample, although locally representative, limits generalizability. 3. Methodology This study adopts a methodological approach, which is based on the design, development, and implementation of a gamified mobile application with the integration of a virtual assistant developed with AI. It is oriented to financial literacy in young university students; in addition, it combines agile development principles, state-of-the-art technologies, and user-centered evaluation methods. 3.1 Overview of the Solution The proposed solution consists of a gamified mobile application for strengthening financial literacy in young university students. This application, developed for Android devices, integrates elements of artificial intelligence, game dynamics, and a user-centered design to provide an interactive, personalized, and motivating learning experience. 3.2 Agile Development Approach The framework for the development process is Scrum, which allowed us to organize the work in eight sprints of three weeks each. Each sprint included planning, development, review, and retrospective activities. In addition, Scrum's artifacts were used: product backlog, sprint backlog, and functional increments, which ensured continuous progress control and the incorporation of iterative improvements. This approach has proven to be effective in the development of educational mobile applications. In (Iwaya et al., 2023 ) developed a digital health mobile application “Early Labour App”, in which they used Scrum in a continuous development environment (CI/CD), highlighting its usefulness for iteratively managing requirements and user testing throughout the development process. Likewise, (Hadi et al., 2022 ) applied the agile Scrum approach to create an adaptive educational system, where they showed how the sprint structure facilitates the progressive integration of key functionalities in educational environments. 3.3 Version management and collaboration For version management and development team coordination, the Git Flow model was used in private repositories on GitHub, which allowed for a collaborative and structured workflow, enabling branch management to work through features, greater version control, and bug fixes. In addition, GitHub served as the primary means for asynchronous collaboration, code review, and progress tracking. The use of these tools has been supported by previous research (Hundhausen et al., 2023 ) analyzed collaborative development projects in university courses, where they combined objective data from GitHub and messaging platforms with peer-to-peer evaluations. Their findings show that the contributions recorded on GitHub correlate significantly with the perceived individual contribution of the development team, which reinforces the usefulness of this tool to foster equity and traceability in academic projects. Similarly, the scientific project UnDifi-2D documents the use of Git as a version control system for software development projects to facilitate collaborative maintenance and code development (Campoli et al., 2022 ). Thus, the choice of Git Flow and GitHub as central elements in the organization, traceability, collaboration, and control in the development stage of the proposed solution is validated. 3.4 Project management tools During development, Kanban methodology was used to visually organize workflow, prioritize user stories, and monitor the progress of functional modules. This agile management tool allows setting work-in-progress (WIP) limits, which encourages team members to finish tasks before starting new ones, thus avoiding overload and fostering continuous value delivery. In the context of software development, Kanban allows achieving a continuous workflow and improving project process management, team coordination, and final product quality by implementing practices such as Kanban board, WIP management, and constant feedback (Damij & Damij, 2024 ). Also, the combination of Kanban with methodologies such as Scrum, along with the integration of the Drum-Buffer-Rope (DBR) method, has proven to be effective in volatile and complex environments. In (Mayo-Alvarez et al., 2024 ), the simulation of different agile scenarios evidenced that the use of Scrum-Kanban together with DBR allowed completing more tasks during the sprint and keeping fewer tasks accumulated in process, which allowed achieving a more agile and efficient workflow. Therefore, the approach of using Scrum and Kanban in the project is empirically supported as an effective strategy for control, adaptation, and efficiency in iterative solution development. 3.5 Development technologies Backend: Developed in Python using the FastAPI framework, which allowed to build a robust REST API, based on OpenAPI and JSON Schema standards, deployed in the cloud through Railway. Frontend: Implemented in Unity with C#, which facilitated the creation of the interactive graphic interface, the gamification mechanics, and the creation of the APK. Database: MySQL was used as the database management system, deployed in the cloud through Railway. Virtual Assistant: Personalization and educational feedback were supported through the integration of Gemini AI as an intelligent assistance engine. 3.6 User experience evaluation To evaluate the user experience, two widely validated evaluation instruments were applied: System Usability Scale (SUS): It is used after the respondent has used the system being evaluated. In addition, all items should be marked; in case the respondent feels that he/she cannot answer, he/she should mark the central point of the scale. This is used to calculate the SUS score by having the totality of contributions per item (Brooke, 1996 ). The ease of application and its ability to synthesize the overall perception of usability make it a widely adopted tool in similar studies. User Experience Questionnaire (UEQ): It is used post-use of the system, where it captures scales of user experience, such as attractiveness, which is a pure valence dimension. Perspicuity, efficiency, and reliability are aspects of pragmatic quality, while stimulation and novelty are aspects of hedonic quality (Schrepp et al., 2017 ). These dimensions allow for a more holistic evaluation of the system. 4. Results This section describes the results obtained from the implementation of the gamified mobile application with artificial intelligence, which focuses on strengthening the financial literacy of university students. 4.1 System Architecture The solution's architecture combines a multi-tier logical structure and a distributed physical implementation, which optimizes both data flow and application performance. In Fig. 1 , the system is composed of three main layers: Frontend: This module manages direct interaction with the user on Android devices, which allows access to the learning modules, gamification, and chat with the virtual assistant. Backend: This module manages business logic and data management. Among its main functions are the authentication and access control of users, management of progress in financial challenges, evaluations (initial, final), achievements and rewards system, and personalized recommendations. In addition, it establishes bidirectional communication with the AI module. AI Server: This module manages the recommendations, where it processes the data sent by the backend to generate personalized recommendations for each user, real-time feedback on the user's decisions during the game, and suggestions based on their performance. Figure 2 shows the deployment architecture of the solution, which is distributed among different system components in independent but interconnected environments. Frontend: The frontend is developed in Unity, where it is packaged and distributed for Android devices. This interface allows the user to interact with the gamification modules and the virtual assistant fluidly and intuitively. Backend: The backend is developed in Python, which is hosted in the cloud through Railway services, which allows the management of requests, manages the business logic, and communication between system components. Database: MySQL was used as the database management system, which is also hosted by Railway to simplify administration and connectivity with the backend. AI Server: Integration with Gemini AI is done independently of the backend, facilitating natural language processing (NLP) and the generation of recommendations tailored to each user. Interoperability: The different modules communicate with each other through REST APIs, allowing an efficient and secure data flow between the frontend, the backend, the database, and the assistant. 4.2 User Module: Authentication and Personalization This module manages secure access to the application through JSON Web Token (JWT), which allows user authentication securely and efficiently. In addition, each user has a personalized profile where it is stored: Personal data. Initial level determined in the diagnostic evaluation. Progress history of both evaluations and games. Rewards obtained. Configuration of notifications and usage preferences. Personalization is also handled, which allows adjusting the content according to the level of knowledge detected in the initial evaluation of each subject, which offers an adaptive experience from the first login for each user. 4.3 Financial Learning Module The educational content is organized around three main topics: Investment. Savings. Credits and debts. Each of these axes is available in three levels of difficulty (basic, intermediate, advanced), where the user has the option to choose the desired level or the level recommended by the system after the initial diagnosis. As visualized in Fig. 3 , within each level, users face: Practical challenges: simulated situations where financial decisions must be made. Formative feedback: The assistant provides immediate explanations in the event of erroneous financial decisions to strengthen learning. Performance feedback: The assistant analyzes the user's performance after completing the game and provides recommendations. Final evaluation (Quiz): A Quiz that appears at the end of the game, which is adapted according to the theme and level of the game. 4.4 Gamification Module The gamification is implemented through: Hit system: Increase or reduction of the initial balance (S/1200) according to the decisions made. As shown in Fig. 4 , at the end of a game, the impact of your decisions on your final balance is visualized; when you get a hit, your balance increases, and when you make a wrong decision, your balance decreases. Achievements for consecutive hits: Collectible coins are rewarded for 3 and/or 5 consecutive hits, and the number of collectible coins achieved is displayed in the score screen at the end of the game (Fig. 4 ). Collectible coins: Symbolic rewards that reinforce the engagement (Fig. 5 ). Unlocking ranking: The user's performance unlocks a new ranking (beginner, master), which encourages continued use of the application. 4.5 Integration with Virtual Assistant A virtual assistant based on Gemini AI was integrated, which plays a tutoring role within the learning process: Provides explanatory feedback after erroneous decisions (Fig. 6 ). Answers queries through an interactive chat (Fig. 7 ). Provides customized definitions from the financial glossary (Fig. 8 ) Accompanies the user throughout the navigation within the gamified scenarios. 4.6 Backend The backend centralizes and manages: Record of each user's progress per game and evaluations made. Record of decisions made, correct and incorrect answers. Control of achievements obtained and rewards unlocked. Management of the use of the question bank for the evaluations (diagnostic and quiz) and the situations of each game according to theme and level. 4.7 Database model A relational model was designed in MySQL to manage in a structured way the different learning, progress, and evaluation processes of the users within the application. As shown in Fig. 9 , the model is composed of the following entities: users: Contains the basic information of each registered user, such as name, email, and personal information. evaluacion_inicial: Records the results of the initial diagnostic test, where a recommended level is determined for each topic. temas_seleccionados: Stores the topics enabled for each user. juego: Records the game sessions per user, including decisions made, accumulated virtual balance, hits, rewards, and achievements. situaciones y opciones: Define the different financial simulation situations presented in the game's subtopics, along with their possible answer options for each selected topic and level. preguntas_quiz: This is the bank of theoretical questions for the quiz. Quiz: Presents the questions according to the game topic and level, to store the results. preguntas_evaluacion: Repositories of questions used in the initial evaluation. progreso_general: Allows tracking the progress of each user, consolidating the results obtained throughout their experience on the platform. monedas: Records the types of collectible game coins that the user can obtain at the end of a game and its quiz. 4.8 Usability Evaluation To evaluate the usability of the proposed mobile application, tests were conducted with 50 young university students between 18 and 25 years of age, belonging to different careers and universities. Each participant received: A user manual with clear instructions on how to use the app A guided session to explore the basic functions of the app. Access to the deployed functional APK, with operational backend. Two evaluation questionnaires: the System Usability Scale (SUS) and the User Experience Questionnaire (UEQ) The data collected in both questionnaires were analyzed in order to identify strengths, opportunities for improvement, and verify if the application meets the usability and user experience criteria of the educational context. 4.8.1 System Usability Scale (SUS) Results The SUS questionnaire is composed of 10 items with positive and negative statements. It provides a score between 0 and 100 that allows interpretation of the perception of system usability, where a score above 68 is considered above average. Figure 9 shows the Bangor scale and the score of 91.9. This score is classified as grade A with an adjective of Excellent, which indicates a high acceptance and ease of use perceived by users. This suggests that the application is understandable and easy to use, even for users with no prior financial knowledge. 4.8.2 User Experience Questionnaire (UEQ) Results Figure 10 shows the various dimensions of the UEQ questionnaire, dimensions that evaluate the user experience, through negative and positive scales. These dimensions are: Attractiveness Clarity Efficiency Accuracy Stimulation Originality Likewise, the values obtained in all dimensions are positive, which reflects a highly satisfactory user experience. Among the dimensions, the perception of originality and stimulation stands out, demonstrating a favorable perception of the playful and interactive approach. Likewise, the clarity dimension was highly rated, reflecting that users easily understood the dynamics of the game. 5. Discussion The results of the usability and user experience validations show the great acceptance and effectiveness of the application developed in this study. The SUS questionnaire obtained an average score of 91.9 points, which places the application in grade A, indicating excellent usability. These results demonstrate that users perceived the mobile application as highly intuitive and easy to use. As shown in Fig. 9 , the value obtained falls within the Excellent category, which reflects high levels of satisfaction and willingness to recommend the application to a friend. This is consistent with the study by (Li et al., 2024 ), who applied SUS to evaluate an interprofessional healthcare education platform, where they obtained high scores reflecting a positive perception in terms of ease of use, navigation, and efficiency. In addition, they emphasize that a high SUS score is related to a shorter learning curve and greater acceptance by end users. Similarly, (Razak & Senan, 2022 ) show similar results in their mobile learning system based on augmented reality as an interactive learning medium, where they obtained 94% acceptance in usability tests applying SUS. Concerning the UEQ results, Fig. 10 shows an outstanding performance in all dimensions, with average scores above 5 (positive) in each one. These results show that users consider the application functional, pleasant, innovative, and attractive. These results align with (Marques & Pombo, 2023 ), who evaluated a gamified application for primary education using UEQ, obtaining high scores in attractiveness, novelty, and efficiency. Their study highlights how a well-designed interface with gamified approaches can significantly increase the level of motivation and engagement of users. Similarly, (Ramli et al., 2024 ) evaluated an educational application with elements of gamification and augmented reality for biology students, obtaining as results values above 0.8 in pragmatic and hedonic scales of the UEQ, which validates its effectiveness in improving the user experience in digital learning environments. Along the same lines, (Lučić & Uzelac, 2024 ) proposed a behavioral approach based on the behavioral theoretical framework “Capacity-Opportunity-Botivation” (COM-B), in which they highlight that gamification and persuasion can increase automatic motivation and facilitate sustainable changes in behavior. This research underlines that effective educational interventions must be considered, in addition to knowledge, motivational, and social aspects that influence economic decisions, which supports the pedagogical approach adopted in this solution. Thus, the validation stage of a system requires the use of complementary instruments, such as the SUS and the UEQ. While the SUS synthesizes the general perception of usability, the UEQ delves into hedonic and pragmatic aspects of the user experience. This combination allows for a more holistic and reliable analysis of the system, which is key in gamified learning contexts. On the other hand, key factors that contributed to the high scores include the integration of the AI-based virtual assistant (Gemini), the personalization of learning through the initial assessment, the achievement and reward system, the gamification, and the feedback provided by the assistant in different sections of the application. However, it is important to recognize that the validation tests were conducted with a limited sample of 50 university students, which, while re-presenting a manageable sample, needs to be expanded in future studies to improve the generalizability of the results, in addition, the study focused only on three financial topics, so the scalability to other content is yet to be validated. It would also be valuable to make comparisons with traditional methods of financial education and evaluate the impact of prolonged use of the application on users' real decision-making. 6. Conclusion This study presents the technical development and implementation of a mobile application, which is supported by a user-centered pedagogical approach, personalized learning elements, game mechanics, and emerging technologies to promote responsible financial decision making. As main findings, the application shows high user acceptance and satisfaction with its use, as well as a positive perception in pragmatic and hedonic terms. These support the effectiveness of the strategy employed, which integrates diagnostic phases, interactive challenges, and personalized feedback. Likewise, the modular structure of the system, the progressive design of the levels, and the incorporation of the virtual assistant as decision support prove to be key factors for the success of this type of educational tool. Finally, this study provides an innovative user-based solution that contributes to the field of digital educational technologies applied to financial education, where it promotes a functional, scalable, and validated architecture that can be adapted to other training contexts and educational levels. Declarations Consent Statement: The participants involved in the study were individuals aged 18 to 25, all belonging to the specific target segment of the research. No minors were involved. All participants provided informed consent prior to their participation, in accordance with the ethical guidelines approved by the university. Acknowledgements The researchers would like to express their gratitude to the Research Department of the Universidad Peruana de Ciencias Aplicadas for funding this research. Author’s contributions: Angie Ruiz wrote the article and led the data analysis and interpretation. Survey design and data collection were carried out by Angie Ruiz and Juliana Yauricasa, and both contributed to the development of the mobile application. Juliana Yauricasa contributed to the development of the diagrams. Juan Morales contributed to the supervision, revision and submission of the paper. All authors contributed to the manuscript and have read and approved the final version. Funding: Provided by the Universidad Peruana de Ciencias Aplicadas Data availability: The data sets used and analyzed during this study are available upon request from the corresponding author. Code availability: The code used to develop the backend, frontend, and virtual assistant is available upon request from the corresponding author. Conflict of interest: The authors declare that they have no conflict of interest Originality and review status: The authors declare that this manuscript is original and has not been published in any other journal. References Acharya, D. B., Divya, B., & Kuppan, K. (2024). Explainable and Fair AI: Balancing Performance in Financial and Real Estate Machine Learning Models. IEEE Access, 12, 154022–154034. https://doi.org/10.1109/ACCESS.2024.3484409 Akour, I., Alzyoud, M., Alquqa, E. K., Tariq, E., Alzboun, N., Al-Hawary, S. I. S., & Alshurideh, M. T. (2024). Artificial intelligence and financial decisions: Empirical evidence from developing economies. International Journal of Data and Network Science, 8(1), 101–108. https://doi.org/10.5267/j.ijdns.2023.10.013 Asmawi, A., Saifulbahri, E. M., & Arifin, N. A. M. (2024). Development of BlockScholar as an Educational Mobile Application on Blockchain Technology. Journal of Advanced Research in Applied Sciences and Engineering Technology, 34(1), 15–23. https://doi.org/10.37934/araset.34.1.1523 Björklund, M., & Sandahl, J. (2023). Teaching and learning financial literacy within social studies–a case study on how to realise curricular aims and ambitions. Journal of Curriculum Studies, 55(3), 325 – 338. https://doi.org/10.1080/00220272.2023.2203771 Blanco, L. R., Hernandez, I., Thames, A. D., Chen, L., & Serido, J. (2023). Mind your money: A community-based digital intervention for improving financial capability among Hispanics. Journal of Economic Behavior & Organization, 212, 629–643. https://doi.org/10.1016/J.JEBO.2023.05.030 Brooke, J. (1996). SUS – a quick and dirty usability scale (pp. 189–194). Campoli, L., Assonitis, A., Ciallella, M., Paciorri, R., Bonfiglioli, A., & Ricchiuto, M. (2022). UnDiFi-2D: An unstructured discontinuity fitting code for 2D grids. Computer Physics Communications, 271. https://doi.org/10.1016/j.cpc.2021.108202 Damij, N., & Damij, T. (2024). An Approach to Optimizing Kanban Board Workflow and Shortening the Project Management Plan. IEEE Transactions on Engineering Management, 71, 13266 – 13273. https://doi.org/10.1109/TEM.2021.3120984 Erdem, O., Hassett, K., & Egriboyun, F. (2025). Hallucination in AI-generated financial literature reviews: evaluating bibliographic accuracy. International Journal of Data Science and Analytics. https://doi.org/10.1007/s41060-025-00731-0 French, D., McKillop, D., & Stewart, E. (2021). Personal finance apps and low-income households. Strategic Change, 30(4), 367–375. https://doi.org/10.1002/jsc.2430 Frisancho, V., Herrera, A., & Prina, S. (2023). Can a mobile-app-based behavioral intervention teach financial skills to youth? Experimental evidence from a financial diaries study. Journal of Economic Behavior and Organization, 214, 595–614. https://doi.org/10.1016/j.jebo.2023.08.014 Grijalvo, M., Segura, A., & Núñez, Y. (2022). Computer-based business games in higher education: A proposal of a gamified learning framework. Technological Forecasting and Social Change, 178. https://doi.org/10.1016/j.techfore.2022.121597 Hadi, S. H., Permanasari, A. E., Hartanto, R., Sakkinah, I. S., Sholihin, M., Sari, R. C., & Haniffa, R. (2022). Developing augmented reality-based learning media and users’ intention to use it for teaching accounting ethics. Education and Information Technologies, 27(1), 643–670. https://doi.org/10.1007/s10639-021-10531-1 Hassan, J. U., Missen, M. M. S., Firdous, A., Maham, A., & Ikram, A. (2023). An Adaptive M-Learning Usability Model for Facilitating M-Learning for Slow Learners. International Journal of Interactive Mobile Technologies, 17(19), 48–69. https://doi.org/10.3991/ijim.v17i19.42153 Hean, O., Saha, U., & Saha, B. (2025). Can AI help with your personal finances? Applied Economics. https://doi.org/10.1080/00036846.2025.2450384 Hundhausen, C., Conrad, P., Adesope, O., & Tariq, A. (2023). Combining GitHub, Chat, and Peer Evaluation Data to Assess Individual Contributions to Team Software Development Projects. ACM Transactions on Computing Education, 23(3). https://doi.org/10.1145/3593592 Imam, T., Cowling, M., & Das, N. (2022). Designing Computer Games to Teach Finance and Technical Concepts in an Online Learning Context: Potential and Effectiveness. Mathematics, 10(22). https://doi.org/10.3390/math10224205 Iwaya, L. H., Nordin, A., Fritsch, L., Børøsund, E., Johansson, M., Varsi, C., & Ängeby, K. (2023). Early Labour App: Developing a practice-based mobile health application for digital early labour support. International Journal of Medical Informatics, 177. https://doi.org/10.1016/j.ijmedinf.2023.105139 Lew, C., & Saville, A. (2021). Game-based learning: Teaching principles of economics and investment finance through Monopoly. International Journal of Management Education, 19(3). https://doi.org/10.1016/j.ijme.2021.100567 Li, Q., Zhao, J., Yan, R., Gao, Q., Zhen, Y., Li, X., Liang, Y., Min, S., & Yang, L. (2024). WeChat mini program in laboratory biosafety education among medical students at Guangzhou Medical University: a mixed method study of feasibility and usability. BMC Medical Education, 24(1). https://doi.org/10.1186/s12909-024-05131-9 Lisana, L., Dinata, H., & Valencia Tanudjaja, G. (2025). Playing to learn: Game-based approach to financial literacy for generation Z. Entertainment Computing, 52, 100896. https://doi.org/10.1016/J.ENTCOM.2024.100896 Liu, T.-Y., Jiang, Y.-H., Wei, Y., Wang, X., Huang, S., & Dai, L. (2024). Educational Practices and Algorithmic Framework for Promoting Sustainable Development in Education by Identifying Real-World Learning Paths. Sustainability (Switzerland), 16(16). https://doi.org/10.3390/su16166871 Lučić, A., & Uzelac, M. (2024). Investigating alternative avenues for financial behaviour change: moving beyond the traditional approach. Young Consumers. https://doi.org/10.1108/YC-05-2023-1748 Mabwe, K., Aminu, N., Ivanov, S. H., & Dimov, D. (2025). Generative artificial intelligence chatbots in investment decision-making: a phantom menace or a new hope? Foresight, 27(4), 820 – 863. https://doi.org/10.1108/FS-06-2024-0122 Malik, A. (2023). Assessing the Effectiveness of Financial Literacy Mobile Apps Using the Content Analysis Approach. International Journal of Interactive Mobile Technologies, 17(23), 68–84. https://doi.org/10.3991/IJIM.V17I23.42213 Marques, M. M., & Pombo, L. (2023). User Experience of a Mobile App in a City Tour Game for International Doctoral Students. Education Sciences, 13(12). https://doi.org/10.3390/educsci13121221 Mayo-Alvarez, L., Del-Aguila-Arcentales, S., Alvarez-Risco, A., Chandra Sekar, M., Davies, N. M., & Yáñez, J. A. (2024). Innovation by integration of Drum-Buffer-Rope (DBR) method with Scrum-Kanban and use of Monte Carlo simulation for maximizing throughput in agile project management. Journal of Open Innovation: Technology, Market, and Complexity, 10(1). https://doi.org/10.1016/j.joitmc.2024.100228 Pitthan, F., & Witte, K. De. (2025). Game over or continue? How gamification can improve completion rate in adaptive learning. Education and Information Technologies, 30(3), 2757–2786. https://doi.org/10.1007/S10639-024-12928-0 Platz, L., & Zauner, M. (2025). Financial Literacy Games—Increasing Utility Value by Instructional Design in Upper Secondary Education. Education Sciences, 15(2). https://doi.org/10.3390/educsci15020227 Qatawneh, A. M., Lutfi, A., & Al Barrak, T. (2024). Effect of Artificial Intelligence (AI) on Financial Decision-Making: Mediating Role of Financial Technologies (Fin-Tech). HighTech and Innovation Journal, 5(3), 759–773. https://doi.org/10.28991/HIJ-2024-05-03-015 Ramli, R. Z., Sahari Ashaari, N., Mat Noor, S. F., Noor, M. M., Yadegaridehkordi, E., Abd Majid, N. A., Dahlan, H. A., & Abdul Wahab, A. N. (2024). Designing a mobile learning application model by integrating augmented reality and game elements to improve student learning experience. Education and Information Technologies, 29(2), 1981–2008. https://doi.org/10.1007/s10639-023-11874-7 Razak, R. N. I. A., & Senan, N. (2022). Mobile Learning for Manually Coded Malay Sign Language Using Augmented Reality. Journal of Soft Computing and Data Mining, 3(1), 86–94. https://doi.org/10.30880/jscdm.2022.03.01.009 Reisdorfer-da-Silva, R. C., Becker, K. L., & Vieira, K. M. (2025). The impact of board games on the financial literacy of public-school students. Journal of Behavioral and Experimental Economics, 114, 102331. https://doi.org/10.1016/J.SOCEC.2024.102331 Robb, C. A., & Chy, S. (2023). Undergraduate financial knowledge, attitudes, and behaviors: The impact of financial life skills course on college students. Financial Planning Review, 6(1). https://doi.org/10.1002/cfp2.1155 Ruiz, A. N., Yauricasa, J. A., & Morales, J. C. (2024). Design of a Mobile Learning App for Financial Literacy in Young People Using Gamification. IJACSA) International Journal of Advanced Computer Science and Applications, 15(12). www.ijacsa.thesai.org Schrepp, M., Hinderks, A., & Thomaschewski, J. (2017). Construction of a Benchmark for the User Experience Questionnaire (UEQ). International Journal of Interactive Multimedia and Artificial Intelligence, 4, 40–44. https://doi.org/10.9781/ijimai.2017.445 Sinnewe, E., & Nicholson, G. (2023). Healthy financial habits in young adults: An exploratory study of the relationship between subjective financial literacy, engagement with finances, and financial decision-making. Journal of Consumer Affairs, 57(1), 564 – 592. https://doi.org/10.1111/joca.12512 Sujith, A. V. L. N., Qureshi, N. I., Dornadula, V. H. R., Rath, A., Prakash, K. B., & Singh, S. K. (2022). A Comparative Analysis of Business Machine Learning in Making Effective Financial Decisions Using Structural Equation Model (SEM). Journal of Food Quality, 2022. https://doi.org/10.1155/2022/6382839 Torma, J., Barbić, D., & Ivanov, M. (2023). Analyzing the Effects of Financial Education on Financial Literacy and Financial Behaviour: A Randomized Field Experiment in Croatia. South East European Journal of Economics and Business, 18(2), 63–86. https://doi.org/10.2478/jeb-2023-0019 Yánez-Pérez, I., Toma, R. B., & Meneses-Villagrá, J. Á. (2024). Design and usability of IndagApp: an app for inquiry-based science education | Diseño y usabilidad de IndagApp: una app para la enseñanza de las ciencias por indagación. RIED-Revista Iberoamericana de Educacion a Distancia, 27(2). https://doi.org/10.5944/ried.27.2.39109 Additional Declarations The authors declare no competing interests. Supplementary Files Appendix.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7170120","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":488088342,"identity":"3c8e1a15-bfcd-4952-a90d-34b83c462822","order_by":0,"name":"Angie Nayeli Ruiz-Carhuamaca","email":"","orcid":"https://orcid.org/0009-0009-8420-2649","institution":"Universidad Peruana de Ciencias Aplicadas","correspondingAuthor":false,"prefix":"","firstName":"Angie","middleName":"Nayeli","lastName":"Ruiz-Carhuamaca","suffix":""},{"id":488088509,"identity":"ed96ecca-ce01-4274-8c73-e1d914e6c8e6","order_by":1,"name":"Juliana Alexandra Yauricasa-Seguil","email":"","orcid":"https://orcid.org/0009-0001-7540-7820","institution":"Universidad Peruana de Ciencias Aplicadas","correspondingAuthor":false,"prefix":"","firstName":"Juliana","middleName":"Alexandra","lastName":"Yauricasa-Seguil","suffix":""},{"id":488088510,"identity":"9065438a-01a4-459e-b1c1-1955455880ac","order_by":2,"name":"Juan Carlos Morales-Arevalo","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/UlEQVRIiWNgGAWjYBACxgYYix2IP0CYBkRqYQZyZhCjBQGAWph5iNHCPCP52QPGHJt8/mbmw59t2+7IM7A3b5Ng3GGD22Ez0swNGLelWc44zJYmndv2zLCB51iZBOOZNDxaEswkGLcdNmA4zGPGnNt2OIFBIgco0nYYj5b0b2At8of5P3+2BGmRfwPS8h+PlhyILQaHeRikGcG28IC0HMCtpedNmUTitjQDw8NsZpI9554ZtvGkFVsknknGqcWwPX2bxMdtNgZyx5sff/hRdkeen/3wxhsfd9jh1tIAJBIQ/AMMbCAqsQGnDgZ5ND7UC4hEMQpGwSgYBaOAAQBsoE7AD1IzDQAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-1834-6451","institution":"Universidad Peruana de Ciencias Aplicadas","correspondingAuthor":true,"prefix":"","firstName":"Juan","middleName":"Carlos","lastName":"Morales-Arevalo","suffix":""}],"badges":[],"createdAt":"2025-07-20 14:03:34","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-7170120/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7170120/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":87476652,"identity":"a7953ffe-a973-4fd2-a041-a095de29503c","added_by":"auto","created_at":"2025-07-24 09:15:32","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":113199,"visible":true,"origin":"","legend":"\u003cp\u003eLogical architecture\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7170120/v1/e918a24d927d672d601f7201.png"},{"id":87477381,"identity":"caa5b253-5e11-488d-988a-504d1f3c1127","added_by":"auto","created_at":"2025-07-24 09:23:32","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":61333,"visible":true,"origin":"","legend":"\u003cp\u003ePhysical architecture\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7170120/v1/ccb160e3c92476f5b0d6b191.png"},{"id":87477385,"identity":"1678b96f-c7bc-4ed6-9b84-59e8a257eafe","added_by":"auto","created_at":"2025-07-24 09:23:32","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":330550,"visible":true,"origin":"","legend":"\u003cp\u003eFlowchart of the learning modules\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7170120/v1/70759e955d2b9d1ed4195a71.png"},{"id":87476653,"identity":"aa0ecb0f-ddb0-4f62-a2cf-a98c64cdcb2d","added_by":"auto","created_at":"2025-07-24 09:15:32","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":88499,"visible":true,"origin":"","legend":"\u003cp\u003eScore screen at the end of a game\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7170120/v1/fe07bbb454c21b24f8c57d7f.png"},{"id":87477383,"identity":"b3b98c56-53f8-4010-b220-0b84e0682b57","added_by":"auto","created_at":"2025-07-24 09:23:32","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":86921,"visible":true,"origin":"","legend":"\u003cp\u003eRewards screen\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7170120/v1/8b7f25322dbb74e11675c287.png"},{"id":87479722,"identity":"67f7b35c-e479-4c82-a4d7-82a87abf3467","added_by":"auto","created_at":"2025-07-24 09:39:32","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":188099,"visible":true,"origin":"","legend":"\u003cp\u003eWizard screen after erroneous decision feedback\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-7170120/v1/c6260087b8f3786fbe56a1f7.png"},{"id":87476667,"identity":"4ccefef6-b74b-484b-a584-dc5899d27cbf","added_by":"auto","created_at":"2025-07-24 09:15:32","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":163936,"visible":true,"origin":"","legend":"\u003cp\u003eVirtual assistant chat\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-7170120/v1/e24a584ee7a40ae0047effec.png"},{"id":87477388,"identity":"b1c8fe36-37de-4c69-ace2-3c375f5df69d","added_by":"auto","created_at":"2025-07-24 09:23:33","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":127327,"visible":true,"origin":"","legend":"\u003cp\u003eGlossary\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-7170120/v1/753c155b6aeafee43a442296.png"},{"id":87478784,"identity":"c26d2197-4f0d-4417-a710-a1ed6b5cf881","added_by":"auto","created_at":"2025-07-24 09:31:33","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":313232,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig. 8.\u003c/strong\u003e Database model of the proposed system. Note: The labels are in Spanish to reflect the original implementation\u003c/p\u003e","description":"","filename":"08.png","url":"https://assets-eu.researchsquare.com/files/rs-7170120/v1/2803c5f77ca0b704a0e5d3cf.png"},{"id":87476670,"identity":"70b1912e-62f4-44da-a43a-800d03e2f3bd","added_by":"auto","created_at":"2025-07-24 09:15:33","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":80796,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig. 9.\u003c/strong\u003e SUS score scale\u003c/p\u003e","description":"","filename":"9.png","url":"https://assets-eu.researchsquare.com/files/rs-7170120/v1/91ea3a6e2c3d0de382e89f7a.png"},{"id":87476704,"identity":"0fd20057-a87b-48a1-b625-fbbe4032df5c","added_by":"auto","created_at":"2025-07-24 09:15:33","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":69980,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig. 10.\u003c/strong\u003e UEQ Dimensions\u003c/p\u003e","description":"","filename":"10.png","url":"https://assets-eu.researchsquare.com/files/rs-7170120/v1/142e316e31f39985d5b17624.png"},{"id":87477391,"identity":"09726b32-72f3-44af-b75c-df8af4fee718","added_by":"auto","created_at":"2025-07-24 09:23:33","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":82377,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig. 11.\u003c/strong\u003e UEQ by Items\u003c/p\u003e","description":"","filename":"11.png","url":"https://assets-eu.researchsquare.com/files/rs-7170120/v1/aadacf22c64532e95098a5a5.png"},{"id":87479956,"identity":"d977b9a2-f055-4a10-9b48-ec8bbdb5d685","added_by":"auto","created_at":"2025-07-24 09:47:33","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2511465,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7170120/v1/8771e513-3fc7-45ad-acba-3eebc5b747ba.pdf"},{"id":87476656,"identity":"043592a6-4671-44cc-934e-1b8ab759d18d","added_by":"auto","created_at":"2025-07-24 09:15:32","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":225628,"visible":true,"origin":"","legend":"","description":"","filename":"Appendix.docx","url":"https://assets-eu.researchsquare.com/files/rs-7170120/v1/9dd6ad5a0abac619034f23ca.docx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eDevelopment of a Personalized Gamified Mobile Learning App Using an AI-Based Virtual Assistant for Financial Literacy\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eThe economic environment is becoming increasingly complex; therefore, financial education, which is an essential component in this area, must provide comprehensive training for young people that is directly proportional to the complexity of the financial system. However, several studies have shown deficiencies in the level of financial literacy among university students, which demonstrates a deficient education. In (Ruiz et al., 2024), the level of financial knowledge of university students from various Peruvian universities was analyzed, and the results showed that there is a lack of basic financial concepts, which limits their ability to correctly manage their financial resources and achieve economic planning for the future. Likewise, in other international contexts, they have also reported low levels of financial literacy as in (Torma et al., 2023), who found that after a financial literacy study of university students in Croatia, more than 70% indicated that they were rarely or seldom informed about finances and two thirds of said survey stated that they were not familiar with financial concepts, which showed that the level of financial literacy of said sample is moderately low.\u003c/p\u003e\n\u003cp\u003eOn the other hand, the development of innovative solutions to provide quality education is valued. As reported by Ruiz (2024), 52% of students prefer learning methods based on interactive games, and 81% value receiving personalized recommendations. This drives the development of innovative solutions such as the proposed solution, which incorporates a form of gamification to promote user engagement. This approach achieves the objectives of the applications, since implementing game mechanics, such as reward systems, challenges, achievements, and interactive messaging, has proven to be effective in encouraging user participation and solving problems to meet game goals (French et al., 2021). It should be noted that students who used financial applications with gamified elements showed better results in achieving financial goals, such as saving money, due to the motivation generated by the rewards, which highlights the value of personalizing motivation in digital environments (French et al., 2021).\u003c/p\u003e\n\u003cp\u003eConcerning the inclusion of financial topics in educational institutions, there are still significant gaps in the effectiveness of the methodologies they employ, because despite the inclusion of financial content in their curricula, they lack dynamic strategies to capture and maintain the interest of young people. Even though financial education courses exist, they tend to be optional, of short duration, and do not always manage to improve financial wellbeing or reduce students' financial stress, since the content is limited and of little impact (Robb \u0026amp; Chy, 2023). Studies reveal that many financial education initiatives maintain the traditional teaching approach of addressing financial topics such as money management, with simplified and poorly contextualized methods that do not address the realities and challenges of the students' real environment (Björklund \u0026amp; Sandahl, 2023). This lack of contextualization and dynamism reduces student interest and participation. In this sense, research such as (Sinnewe \u0026amp; Nicholson, 2023) highlights that financial education that is more aligned with the real environment of young people contributes directly to developing better financial habits in adulthood. This reinforces the need to redesign existing methodologies and integrate emerging technologies that allow personalizing the learning experience according to the profile of each young person so that they feel more motivated and committed to learning.\u003c/p\u003e\n\u003cp\u003eRegarding the playful approach to financial education. It was shown that the use of board games in educational contexts can significantly improve the level of knowledge of university students, as well as present better behavior and attitude towards financial challenges (Reisdorfer-da-Silva et al., 2025). Likewise, there is a positive impact on game mechanics and reflection prompts as demonstrated in the financial game Moonshot, which uses the combination of these elements to significantly increase the useful value perceived by users, emphasizing the importance of the instructional design of these elements (Platz \u0026amp; Zauner, 2025). Similarly, the traditional game Monopoly, which was adapted as a financial learning tool, managed to stimulate critical reflection on financial decisions, irrationality in investing, and financial behavior, which shows that well-designed games manage to facilitate the learning of deep-rooted topics and that it is long-lasting (Lew \u0026amp; Saville, 2021). Thus, the ludic approach allows the development of decision-making skills in an entertaining and high-impact way for users.\u003c/p\u003e\n\u003cp\u003eIn this scenario, the need arises to design solutions that combine playful approaches with technological tools to strengthen financial literacy among young university students, such as the use of mobile applications, gamification, AI, and personalized recommendations. This is how the proposed solution is presented as an innovative alternative to meet this demand. Thus, strengthening the financial competencies of students effectively and sustainably. It should be noted that, in a previous study, the conceptual design of this solution was presented, where the design of the proposal was based on the results obtained from the survey conducted on the level of financial knowledge and learning preferences, resulting in the wireframes and mockups of the application.\u003c/p\u003e\n\u003cp\u003eIn this context, the main objective of this article is to give continuity to that initial proposal, documenting the technical development process and implementation of the application. It describes the use of the agile Scrum methodology to organize the project in development sprints, the design of the physical and logical architecture of the system, the database, and the integration of the AI-based virtual assistant. In addition, it includes the results of the validation phase with beta users, which was obtained by using the tools: System Usability Scale (SUS) and User Experience Questionnaire (UEQ). To evaluate the usability perceived by the user, the SUS was used, which is a simple 10-item scale that allows a global view of the subjective evaluations of usability, where each item has a score that is used to calculate the SUS score ranging from 0 to 100 (Brooke, 1996). In contrast, to measure not only technical aspects, but also emotional and perceptual components about user experience (attractiveness, pragmatic quality, hedonic quality) the UEQ was used, which is a quick and direct measurement of user experience where each item consists of a pair of terms with opposite meanings, with a total of 26 items grouped into 6 scales (Schrepp et al., 2017). These tools complement each other; by combining them, it is possible to obtain a comprehensive view of the system, both the efficiency of use and the subjective experience of the users.\u003c/p\u003e\n\u003cp\u003eThe article is organized as follows: the methodology section describes the development approach, the tools used, and the architectural design; the results section shows the development of the solution. Finally, the discussion section interprets the findings, the main limitations, and the comparison of results with other studies, and concludes by naming the main contributions and proposals for future research.\u003c/p\u003e"},{"header":"2. Background information","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Mobile Learning Applications in Education\u003c/h2\u003e\u003cp\u003e(Hassan et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) presents an adaptive usability model for mobile learning applications, focused on students with cognitive difficulties. Their proposal seeks to improve the mobile learning experience through a hybrid and personalized recommendation system. For this purpose, a systematic usability analysis and evaluation of applications developed in HTML was used. The results showed that there are significant deficiencies in the existing applications of the study group, and a more accessible and personalized navigation model was suggested. However, the focus of this study is limited to students with special abilities and does not consider the university population.\u003c/p\u003e\u003cp\u003e(Frisancho et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) analyzed the impact of a behavioral intervention based on a mobile application on financial literacy and behavior for young Peruvian high school graduates. The intervention combined an application to record transactions, biweekly SMS messages, and periodic visits for 27 weeks. The results showed significant improvements in financial knowledge and price awareness, although there was no change in savings and budgeting habits. Data from credit bureaus revealed an increase in credit usage and subsequent financial inclusion. On the other hand, the design did not include explicit educational content but promoted self-reflection and financial curiosity. Thus, it shows that simple digital tools, combined with behavioral nudging strategies, can partially substitute traditional financial education in contexts of difficult access and vulnerability.\u003c/p\u003e\u003cp\u003e(Y\u0026aacute;nez-P\u0026eacute;rez et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) presents the design, development, and evaluation of an educational mobile application (IndagApp) that facilitates science teaching through school inquiry methodology. The artifact was developed under an iterative approach of continuous improvement by applying usability tests such as SUS, UES (User Engagement Scale), and MAU (Mobile Application Usability). The results showed an adequate usability of IndagApp, which is a resource that captures attention, generates interest and enjoyment, and is a useful tool for teachers. Although the study is not based on financial education, it provides valuable methodological input for the development of user-centered educational mobile applications. However, the study does not explore metrics of effectiveness in financial literacy.\u003c/p\u003e\u003cp\u003eIn (Asmawi et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), the development of BlockScholar, a mobile educational application focused on blockchain technology using gamification, is presented. It was based on the Rapid Application Development (RAD) model, which encourages collaboration between developers and end users to accelerate the development process. The results show that the combination of game-based learning and mobile technology is successful in increasing participation and retention of information among young people. However, the study does not include long-term follow-up metrics on the impact of game-based learning on the development of cognitive skills and knowledge retention in youth.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Gamification in Educational Technology\u003c/h2\u003e\u003cp\u003e(Grijalvo et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) presents a gamified learning framework based on the principles of Mechanics, Dynamics, and Emotions (MDE) for the integration of digital business games in university contexts. It is supported by two strategic simulators: Gestionet and Global Management Challenge - GMC, which allows the development of financial competencies and soft skills. We addressed how to effectively integrate gamified tools into the formal curriculum, promoting the acquisition of financial knowledge and the development of competencies demanded by the labor market. The results highlight that intrinsic motivation and self-motivation influence the improvement of competencies, while participation in simulators increases satisfaction and recommendation intention. However, the study is limited by the single institutional context and does not explore the integration of dynamic personalization or AI. Nor do they address learning outcomes in terms of effective acquisition of financial content.\u003c/p\u003e\u003cp\u003e(Pitthan \u0026amp; Witte, 2025) analyzed how gamification can increase completion rates in adaptive learning environments using a financial simulation tool for adults. The study validates a theoretical model that integrates gamification with content personalization and gamification elements. The results demonstrate the importance of psychological factors as mediators of user retention, providing evidence that the strategic inclusion of gamified elements during critical phases of learning can significantly increase retention rates. This reinforces the need to integrate adaptive dynamics and gamification at key moments in the user's journey.\u003c/p\u003e\u003cp\u003eIn (Lisana et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) present the design, implementation, and evaluation of a gamified educational application to promote financial literacy in young university students. The developed video game simulates realistic economic decisions and is supported by immersive narratives, instant feedback, progressive challenges, and virtual rewards. The results showed a significant improvement in financial literacy scores after using the game, as well as an assessment of its usefulness and attractiveness. However, the sustained impact on financial behavior is not evaluated, nor are the mechanisms for personalization or adaptation of the content according to the user's profile explored.\u003c/p\u003e\u003cp\u003eIn (Imam et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), a gamified quiz video game is developed to teach technical concepts of corporate finance in an Australian university context. It was developed under the \u0026ldquo;cognitive apprenticeship\u0026rdquo; model and seeks to facilitate active learning through questions with random values, variable scores, and immediate feedback, promoting the transfer of knowledge to real-world scenarios. The evaluation was based on a structured perception questionnaire, with Bayesian regression analysis revealing a high relationship between game usability, learning perception, and overall satisfaction. Furthermore, it reinforces the idea that effective gamified environments should focus not only on content but also on user experience, professional relevance, and cognitive load reduction.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.3 AI-Based Virtual Assistants for Learning\u003c/h2\u003e\u003cp\u003e(Hean et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) investigate the ability of large-scale language models (LLMs) to provide personalized and reliable financial advice. A systematic comparative analysis of multiple LLMs-ChatGPT (3.5, 4, 4o), Claude (3 Haiku, 3.5 Sonnet, 3 Opus), Gemini, and LLaMA-was conducted using real financial literacy tests such as Money Counts and NFEC. The study reveals that the most advanced models exceeded 79% accuracy on topics such as insurance and student loans, whereas they failed on financial psychology. Although the pedagogical quality of the answers and adaptive personalization mechanisms were not evaluated, it is shown that LLMs can function as predictive agents that adapt their performance according to topic difficulty. This reinforces their potential use as virtual assistants in gamified educational applications to provide adaptive support, assess prior knowledge, and generate dynamic content.\u003c/p\u003e\u003cp\u003e(Mabwe et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) examines the role of generative artificial intelligence (GenAI) chatbots, such as ChatGPT, Microsoft Bing, and Google Bard (Gemini), in supporting investment decision making through standardized prompts about stocks, ETFs, diversification, and market trends. The results indicate medium-to-high performance and a tendency to avoid risky decisions. However, they also made significant errors, such as references to non-existent assets. Although the study focuses on investment contexts, its findings are relevant, as they show the ability of these systems to generate relevant and specific content. However, the content analysis does not consider the level of understanding or bias of the language used by the chatbots, a crucial element if these systems are to be applied to educational or gamified contexts.\u003c/p\u003e\u003cp\u003e(Erdem et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) evaluates the literature accuracy of GenAI-based virtual assistants (ChatGPT-4o, o1-preview, and Gemini Advanced), focusing on their application for financial literature reviews. The results reveal that ChatGPT-4o and o1-preview had hallucination rates of 20.0% and 21.3% respectively (binary scale). In contrast, Gemini Advanced had a significantly higher rate: 76.7% (binary). Also, a higher incidence of hallucinations was evidenced in \u0026ldquo;new\u0026rdquo; subjects. This study is highly relevant because it warns about the risks of blindly trusting content generated by LLMs without validation and curation processes. These findings underscore the need to incorporate layers of automated verification, bibliographic traceability, and explainability in conversational agents, especially when promoting learning based on academic sources.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.4 AI for Financial Intelligence: Educational Implications\u003c/h2\u003e\u003cp\u003eIn (Qatawneh et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), the impact of AI on financial decision-making is analyzed, highlighting the mediating role of financial technologies (Fin-Tech). A structural model is proposed that examines how seven AI techniques-NLP, machine learning algorithms, computer vision, predictive analytics, robotic process automation (RPA), blockchain, and deep learning-influence financial decision making, mediated by Fin-Tech tools. Their findings confirm that AI has a significant impact on decision-making, in addition to an indirect effect mediated by Fin-Tech, which showed statistically significant mediation, underscoring its role as a catalyst for the adoption of AI. While the study provides an integrative framework and solid empirical evidence, its applicability is limited by the sample size.\u003c/p\u003e\u003cp\u003e(Sujith et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) Examine the key components of machine learning in informed financial decision making and explore how machine learning can optimize decision making within data-intensive business environments. They used surveys of employees and industry leaders and analyzed the impact of the use of Machine Learning models. The results showed that these technologies allow the identification of useful patterns to make more efficient decisions. However, although this study demonstrates the potential of using AI for decision making, its application is limited to the corporate environment and not to educational environments.\u003c/p\u003e\u003cp\u003e(Akour et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) Investigates the impact of various dimensions of AI, such as natural language processing (NLP), machine learning, expert systems, and computer vision, on financial decision making in pharmaceutical companies in Jordan. They used surveys and applied them to accounting and financial professionals and conducted an analysis using structural models. As a result, they identified positive effects of AI on business decision making, making business decisions more efficient and rational. However, the approach is limited to the corporate environment and does not consider the use of AI to provide personalized education to the user to develop financial skills.\u003c/p\u003e\u003cp\u003eIn (Acharya et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), a framework for developing explainable and fair machine learning models applied to financial and real estate contexts, such as loan approval and housing price prediction, is proposed. Advanced algorithms such as LightGBM, XGBoost, transparency techniques such as SHAP, and intersectional fairness techniques such as Calibrated Equalized Odds and Intersectional Fairness were used. The results showed that it is possible to achieve a balance between accuracy, transparency, and fairness, although fairness implied a slight reduction in predictive performance. However, the study provides valuable insights into responsible practices in decision-making processes but does not address adaptive interactions for learning.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.5 Financial Literacy Education and Digital Interventions\u003c/h2\u003e\u003cp\u003e(Malik, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) Conducted a content analysis of 163 financial education mobile apps identified in Google Play and App Store. They evaluated 13 key features related to instructional design, Nielsen heuristics, and personalization principles. As a result, they found that 82% of apps included personalized education and 49% applied gamification. In addition, principles of meaningful learning such as personalization, pre-training, and multimedia were analyzed. However, the study shows a high variability in pedagogical quality, little systematic validation of educational effectiveness, and limited integration of emerging technologies such as AI. This highlights the need to implement more robust frameworks for designing effective mobile financial applications, where not only interaction but also didactic structure and adaptive personalization are considered.\u003c/p\u003e\u003cp\u003e(Reisdorfer-da-Silva et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) evaluated the use of board games as a financial education teaching strategy in public schools in Brazil using a quasi-experimental methodology and propensity score matching. The study focused on training teachers and applying games in the classroom, which resulted in significant improvements in financial knowledge, behavior, and attitudes. It is supported by theoretical frameworks that highlight the effectiveness of active learning and the role of financial preferences in adolescence. Although they demonstrated the potential of interactive games, their use is restricted to the school context and lacks adaptive mechanisms, and neither long-term follow-up strategies nor integration with digital environments are explored.\u003c/p\u003e\u003cp\u003e(Liu et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) developed the Collaborative Structure Search Framework (CSSF) algorithm to optimize personalized learning paths using big data and AI. It used real educational datasets and graph optimization techniques to identify optimal learning sequences. As a result, they obtained improvements in accuracy (F1-score) over traditional methods. However, their proposal is focused on offering personalized educational trajectories in real time without considering aspects of motivation, user engagement, nor is it focused on financial education. However, their contributions are relevant for the design of intelligent systems capable of offering personalized educational trajectories.\u003c/p\u003e\u003cp\u003e(Blanco et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) presents a community-based digital intervention aimed at improving financial capability and reducing financial stress among low- and middle-income Hispanic adults. Mind Your Money (MYM), a digital financial education program designed with a participatory and culturally tailored approach, was developed and validated based on content from the CFPB's Your Money, Your Goals program. It sought to address the gap in financial literacy and well-being with mobile technologies and behavioral change strategies based on \u0026ldquo;nudges\u0026rdquo; (reminders, incentives, and follow-up phone calls). The results show significant effects on financial capability and self-efficacy. In addition, a 17% reduction in financial stress levels and a 66% reduction in retention rates are reported, reinforcing the feasibility of sustained digital interventions. However, the design does not allow isolation of the individual effect of each nudge, and the sample, although locally representative, limits generalizability.\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Methodology","content":"\u003cp\u003eThis study adopts a methodological approach, which is based on the design, development, and implementation of a gamified mobile application with the integration of a virtual assistant developed with AI. It is oriented to financial literacy in young university students; in addition, it combines agile development principles, state-of-the-art technologies, and user-centered evaluation methods.\u003c/p\u003e\n\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\n \u003ch2\u003e3.1 Overview of the Solution\u003c/h2\u003e\n \u003cp\u003eThe proposed solution consists of a gamified mobile application for strengthening financial literacy in young university students. This application, developed for Android devices, integrates elements of artificial intelligence, game dynamics, and a user-centered design to provide an interactive, personalized, and motivating learning experience.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\n \u003ch2\u003e3.2 Agile Development Approach\u003c/h2\u003e\n \u003cp\u003eThe framework for the development process is Scrum, which allowed us to organize the work in eight sprints of three weeks each. Each sprint included planning, development, review, and retrospective activities. In addition, Scrum\u0026apos;s artifacts were used: product backlog, sprint backlog, and functional increments, which ensured continuous progress control and the incorporation of iterative improvements.\u003c/p\u003e\n \u003cp\u003eThis approach has proven to be effective in the development of educational mobile applications. In (Iwaya et al., \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e) developed a digital health mobile application \u0026ldquo;Early Labour App\u0026rdquo;, in which they used Scrum in a continuous development environment (CI/CD), highlighting its usefulness for iteratively managing requirements and user testing throughout the development process. Likewise, (Hadi et al., \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e) applied the agile Scrum approach to create an adaptive educational system, where they showed how the sprint structure facilitates the progressive integration of key functionalities in educational environments.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\n \u003ch2\u003e3.3 Version management and collaboration\u003c/h2\u003e\n \u003cp\u003eFor version management and development team coordination, the Git Flow model was used in private repositories on GitHub, which allowed for a collaborative and structured workflow, enabling branch management to work through features, greater version control, and bug fixes. In addition, GitHub served as the primary means for asynchronous collaboration, code review, and progress tracking. The use of these tools has been supported by previous research (Hundhausen et al., \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e) analyzed collaborative development projects in university courses, where they combined objective data from GitHub and messaging platforms with peer-to-peer evaluations. Their findings show that the contributions recorded on GitHub correlate significantly with the perceived individual contribution of the development team, which reinforces the usefulness of this tool to foster equity and traceability in academic projects. Similarly, the scientific project UnDifi-2D documents the use of Git as a version control system for software development projects to facilitate collaborative maintenance and code development (Campoli et al., \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e). Thus, the choice of Git Flow and GitHub as central elements in the organization, traceability, collaboration, and control in the development stage of the proposed solution is validated.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\n \u003ch2\u003e3.4 Project management tools\u003c/h2\u003e\n \u003cp\u003eDuring development, Kanban methodology was used to visually organize workflow, prioritize user stories, and monitor the progress of functional modules. This agile management tool allows setting work-in-progress (WIP) limits, which encourages team members to finish tasks before starting new ones, thus avoiding overload and fostering continuous value delivery. In the context of software development, Kanban allows achieving a continuous workflow and improving project process management, team coordination, and final product quality by implementing practices such as Kanban board, WIP management, and constant feedback (Damij \u0026amp; Damij, \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e). Also, the combination of Kanban with methodologies such as Scrum, along with the integration of the Drum-Buffer-Rope (DBR) method, has proven to be effective in volatile and complex environments. In (Mayo-Alvarez et al., \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e), the simulation of different agile scenarios evidenced that the use of Scrum-Kanban together with DBR allowed completing more tasks during the sprint and keeping fewer tasks accumulated in process, which allowed achieving a more agile and efficient workflow. Therefore, the approach of using Scrum and Kanban in the project is empirically supported as an effective strategy for control, adaptation, and efficiency in iterative solution development.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\n \u003ch2\u003e3.5 Development technologies\u003c/h2\u003e\n \u003cul\u003e\n \u003cli\u003e\n \u003cp\u003eBackend: Developed in Python using the FastAPI framework, which allowed to build a robust REST API, based on OpenAPI and JSON Schema standards, deployed in the cloud through Railway.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eFrontend: Implemented in Unity with C#, which facilitated the creation of the interactive graphic interface, the gamification mechanics, and the creation of the APK.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eDatabase: MySQL was used as the database management system, deployed in the cloud through Railway.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eVirtual Assistant: Personalization and educational feedback were supported through the integration of Gemini AI as an intelligent assistance engine.\u003c/p\u003e\n \u003c/li\u003e\n \u003c/ul\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\n \u003ch2\u003e3.6 User experience evaluation\u003c/h2\u003e\n \u003cp\u003eTo evaluate the user experience, two widely validated evaluation instruments were applied:\u003c/p\u003e\n \u003cul\u003e\n \u003cli\u003e\n \u003cp\u003eSystem Usability Scale (SUS): It is used after the respondent has used the system being evaluated. In addition, all items should be marked; in case the respondent feels that he/she cannot answer, he/she should mark the central point of the scale. This is used to calculate the SUS score by having the totality of contributions per item (Brooke, \u003cspan class=\"CitationRef\"\u003e1996\u003c/span\u003e). The ease of application and its ability to synthesize the overall perception of usability make it a widely adopted tool in similar studies.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eUser Experience Questionnaire (UEQ): It is used post-use of the system, where it captures scales of user experience, such as attractiveness, which is a pure valence dimension. Perspicuity, efficiency, and reliability are aspects of pragmatic quality, while stimulation and novelty are aspects of hedonic quality (Schrepp et al., \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e). These dimensions allow for a more holistic evaluation of the system.\u003c/p\u003e\n \u003c/li\u003e\n \u003c/ul\u003e\n\u003c/div\u003e"},{"header":"4. Results","content":"\u003cp\u003eThis section describes the results obtained from the implementation of the gamified mobile application with artificial intelligence, which focuses on strengthening the financial literacy of university students.\u003c/p\u003e\n\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\n \u003ch2\u003e4.1 System Architecture\u003c/h2\u003e\n \u003cp\u003eThe solution\u0026apos;s architecture combines a multi-tier logical structure and a distributed physical implementation, which optimizes both data flow and application performance.\u003c/p\u003e\n \u003cp\u003eIn Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e, the system is composed of three main layers:\u003c/p\u003e\n \u003cul\u003e\n \u003cli\u003e\n \u003cp\u003eFrontend: This module manages direct interaction with the user on Android devices, which allows access to the learning modules, gamification, and chat with the virtual assistant.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eBackend: This module manages business logic and data management. Among its main functions are the authentication and access control of users, management of progress in financial challenges, evaluations (initial, final), achievements and rewards system, and personalized recommendations. In addition, it establishes bidirectional communication with the AI module.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eAI Server: This module manages the recommendations, where it processes the data sent by the backend to generate personalized recommendations for each user, real-time feedback on the user\u0026apos;s decisions during the game, and suggestions based on their performance.\u003c/p\u003e\n \u003c/li\u003e\n \u003c/ul\u003e\n \u003cp\u003eFigure \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e shows the deployment architecture of the solution, which is distributed among different system components in independent but interconnected environments.\u003c/p\u003e\n \u003cul\u003e\n \u003cli\u003e\n \u003cp\u003eFrontend: The frontend is developed in Unity, where it is packaged and distributed for Android devices. This interface allows the user to interact with the gamification modules and the virtual assistant fluidly and intuitively.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eBackend: The backend is developed in Python, which is hosted in the cloud through Railway services, which allows the management of requests, manages the business logic, and communication between system components.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eDatabase: MySQL was used as the database management system, which is also hosted by Railway to simplify administration and connectivity with the backend.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eAI Server: Integration with Gemini AI is done independently of the backend, facilitating natural language processing (NLP) and the generation of recommendations tailored to each user.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eInteroperability: The different modules communicate with each other through REST APIs, allowing an efficient and secure data flow between the frontend, the backend, the database, and the assistant.\u003c/p\u003e\n \u003c/li\u003e\n \u003c/ul\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\n \u003ch2\u003e4.2 User Module: Authentication and Personalization\u003c/h2\u003e\n \u003cp\u003eThis module manages secure access to the application through JSON Web Token (JWT), which allows user authentication securely and efficiently. In addition, each user has a personalized profile where it is stored:\u003c/p\u003e\n \u003cul\u003e\n \u003cli\u003e\n \u003cp\u003ePersonal data.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eInitial level determined in the diagnostic evaluation.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eProgress history of both evaluations and games.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eRewards obtained.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eConfiguration of notifications and usage preferences.\u003c/p\u003e\n \u003c/li\u003e\n \u003c/ul\u003e\n \u003cp\u003ePersonalization is also handled, which allows adjusting the content according to the level of knowledge detected in the initial evaluation of each subject, which offers an adaptive experience from the first login for each user.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\n \u003ch2\u003e4.3 Financial Learning Module\u003c/h2\u003e\n \u003cp\u003eThe educational content is organized around three main topics:\u003c/p\u003e\n \u003cul\u003e\n \u003cli\u003e\n \u003cp\u003e\u0026nbsp;Investment.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eSavings.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eCredits and debts.\u003c/p\u003e\n \u003c/li\u003e\n \u003c/ul\u003e\n \u003cp\u003eEach of these axes is available in three levels of difficulty (basic, intermediate, advanced), where the user has the option to choose the desired level or the level recommended by the system after the initial diagnosis. As visualized in Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e, within each level, users face:\u003c/p\u003e\n \u003cul\u003e\n \u003cli\u003e\n \u003cp\u003ePractical challenges: simulated situations where financial decisions must be made.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eFormative feedback: The assistant provides immediate explanations in the event of erroneous financial decisions to strengthen learning.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003ePerformance feedback: The assistant analyzes the user\u0026apos;s performance after completing the game and provides recommendations.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eFinal evaluation (Quiz): A Quiz that appears at the end of the game, which is adapted according to the theme and level of the game.\u003c/p\u003e\n \u003c/li\u003e\n \u003c/ul\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\n \u003ch2\u003e4.4 Gamification Module\u003c/h2\u003e\n \u003cp\u003eThe gamification is implemented through:\u003c/p\u003e\n \u003cul\u003e\n \u003cli\u003e\n \u003cp\u003eHit system: Increase or reduction of the initial balance (S/1200) according to the decisions made. As shown in Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e, at the end of a game, the impact of your decisions on your final balance is visualized; when you get a hit, your balance increases, and when you make a wrong decision, your balance decreases.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eAchievements for consecutive hits: Collectible coins are rewarded for 3 and/or 5 consecutive hits, and the number of collectible coins achieved is displayed in the score screen at the end of the game (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eCollectible coins: Symbolic rewards that reinforce the engagement (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eUnlocking ranking: The user\u0026apos;s performance unlocks a new ranking (beginner, master), which encourages continued use of the application.\u003c/p\u003e\n \u003c/li\u003e\n \u003c/ul\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\n \u003ch2\u003e4.5 Integration with Virtual Assistant\u003c/h2\u003e\n \u003cp\u003eA virtual assistant based on Gemini AI was integrated, which plays a tutoring role within the learning process:\u003c/p\u003e\n \u003cul\u003e\n \u003cli\u003e\n \u003cp\u003eProvides explanatory feedback after erroneous decisions (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eAnswers queries through an interactive chat (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003e).\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eProvides customized definitions from the financial glossary (Fig. \u003cspan class=\"InternalRef\"\u003e8\u003c/span\u003e)\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eAccompanies the user throughout the navigation within the gamified scenarios.\u003c/p\u003e\n \u003c/li\u003e\n \u003c/ul\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\n \u003ch2\u003e4.6 Backend\u003c/h2\u003e\n \u003cp\u003eThe backend centralizes and manages:\u003c/p\u003e\n \u003cul\u003e\n \u003cli\u003e\n \u003cp\u003eRecord of each user\u0026apos;s progress per game and evaluations made.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eRecord of decisions made, correct and incorrect answers.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eControl of achievements obtained and rewards unlocked.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eManagement of the use of the question bank for the evaluations (diagnostic and quiz) and the situations of each game according to theme and level.\u003c/p\u003e\n \u003c/li\u003e\n \u003c/ul\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\n \u003ch2\u003e4.7 Database model\u003c/h2\u003e\n \u003cp\u003eA relational model was designed in MySQL to manage in a structured way the different learning, progress, and evaluation processes of the users within the application. As shown in Fig. \u003cspan class=\"InternalRef\"\u003e9\u003c/span\u003e, the model is composed of the following entities:\u003c/p\u003e\n \u003cul\u003e\n \u003cli\u003e\n \u003cp\u003eusers: Contains the basic information of each registered user, such as name, email, and personal information.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eevaluacion_inicial: Records the results of the initial diagnostic test, where a recommended level is determined for each topic.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003etemas_seleccionados: Stores the topics enabled for each user.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003ejuego: Records the game sessions per user, including decisions made, accumulated virtual balance, hits, rewards, and achievements.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003esituaciones y opciones: Define the different financial simulation situations presented in the game\u0026apos;s subtopics, along with their possible answer options for each selected topic and level.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003epreguntas_quiz: This is the bank of theoretical questions for the quiz.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eQuiz: Presents the questions according to the game topic and level, to store the results.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003epreguntas_evaluacion: Repositories of questions used in the initial evaluation.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eprogreso_general: Allows tracking the progress of each user, consolidating the results obtained throughout their experience on the platform.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003emonedas: Records the types of collectible game coins that the user can obtain at the end of a game and its quiz.\u003c/p\u003e\n \u003c/li\u003e\n \u003c/ul\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\n \u003ch2\u003e4.8 Usability Evaluation\u003c/h2\u003e\n \u003cp\u003eTo evaluate the usability of the proposed mobile application, tests were conducted with 50 young university students between 18 and 25 years of age, belonging to different careers and universities. Each participant received:\u003c/p\u003e\n \u003cul\u003e\n \u003cli\u003e\n \u003cp\u003eA user manual with clear instructions on how to use the app\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eA guided session to explore the basic functions of the app.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eAccess to the deployed functional APK, with operational backend.\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eTwo evaluation questionnaires: the System Usability Scale (SUS) and the User Experience Questionnaire (UEQ)\u003c/p\u003e\n \u003c/li\u003e\n \u003c/ul\u003e\n \u003cp\u003eThe data collected in both questionnaires were analyzed in order to identify strengths, opportunities for improvement, and verify if the application meets the usability and user experience criteria of the educational context.\u003c/p\u003e\n \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e\n \u003ch2\u003e4.8.1 System Usability Scale (SUS) Results\u003c/h2\u003e\n \u003cp\u003eThe SUS questionnaire is composed of 10 items with positive and negative statements. It provides a score between 0 and 100 that allows interpretation of the perception of system usability, where a score above 68 is considered above average. Figure \u003cspan class=\"InternalRef\"\u003e9\u003c/span\u003e shows the Bangor scale and the score of 91.9. This score is classified as grade A with an adjective of Excellent, which indicates a high acceptance and ease of use perceived by users. This suggests that the application is understandable and easy to use, even for users with no prior financial knowledge.\u003c/p\u003e\n \u003c/div\u003e\n \u003cdiv id=\"Sec24\" class=\"Section3\"\u003e\n \u003ch2\u003e4.8.2 User Experience Questionnaire (UEQ) Results\u003c/h2\u003e\n \u003cp\u003eFigure \u003cspan class=\"InternalRef\"\u003e10\u003c/span\u003e shows the various dimensions of the UEQ questionnaire, dimensions that evaluate the user experience, through negative and positive scales. These dimensions are:\u003c/p\u003e\n \u003cul\u003e\n \u003cli\u003e\n \u003cp\u003eAttractiveness\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eClarity\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eEfficiency\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eAccuracy\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eStimulation\u003c/p\u003e\n \u003c/li\u003e\n \u003cli\u003e\n \u003cp\u003eOriginality\u003c/p\u003e\n \u003c/li\u003e\n \u003c/ul\u003e\n \u003cp\u003eLikewise, the values obtained in all dimensions are positive, which reflects a highly satisfactory user experience. Among the dimensions, the perception of originality and stimulation stands out, demonstrating a favorable perception of the playful and interactive approach. Likewise, the clarity dimension was highly rated, reflecting that users easily understood the dynamics of the game.\u003c/p\u003e\n \u003c/div\u003e\n\u003c/div\u003e"},{"header":"5. Discussion","content":"\u003cp\u003eThe results of the usability and user experience validations show the great acceptance and effectiveness of the application developed in this study. The SUS questionnaire obtained an average score of 91.9 points, which places the application in grade A, indicating excellent usability. These results demonstrate that users perceived the mobile application as highly intuitive and easy to use. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e9\u003c/span\u003e, the value obtained falls within the Excellent category, which reflects high levels of satisfaction and willingness to recommend the application to a friend. This is consistent with the study by (Li et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), who applied SUS to evaluate an interprofessional healthcare education platform, where they obtained high scores reflecting a positive perception in terms of ease of use, navigation, and efficiency. In addition, they emphasize that a high SUS score is related to a shorter learning curve and greater acceptance by end users. Similarly, (Razak \u0026amp; Senan, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) show similar results in their mobile learning system based on augmented reality as an interactive learning medium, where they obtained 94% acceptance in usability tests applying SUS.\u003c/p\u003e\u003cp\u003eConcerning the UEQ results, Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e10\u003c/span\u003e shows an outstanding performance in all dimensions, with average scores above 5 (positive) in each one. These results show that users consider the application functional, pleasant, innovative, and attractive. These results align with (Marques \u0026amp; Pombo, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), who evaluated a gamified application for primary education using UEQ, obtaining high scores in attractiveness, novelty, and efficiency. Their study highlights how a well-designed interface with gamified approaches can significantly increase the level of motivation and engagement of users. Similarly, (Ramli et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) evaluated an educational application with elements of gamification and augmented reality for biology students, obtaining as results values above 0.8 in pragmatic and hedonic scales of the UEQ, which validates its effectiveness in improving the user experience in digital learning environments.\u003c/p\u003e\u003cp\u003eAlong the same lines, (Lučić \u0026amp; Uzelac, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) proposed a behavioral approach based on the behavioral theoretical framework \u0026ldquo;Capacity-Opportunity-Botivation\u0026rdquo; (COM-B), in which they highlight that gamification and persuasion can increase automatic motivation and facilitate sustainable changes in behavior. This research underlines that effective educational interventions must be considered, in addition to knowledge, motivational, and social aspects that influence economic decisions, which supports the pedagogical approach adopted in this solution.\u003c/p\u003e\u003cp\u003eThus, the validation stage of a system requires the use of complementary instruments, such as the SUS and the UEQ. While the SUS synthesizes the general perception of usability, the UEQ delves into hedonic and pragmatic aspects of the user experience. This combination allows for a more holistic and reliable analysis of the system, which is key in gamified learning contexts.\u003c/p\u003e\u003cp\u003eOn the other hand, key factors that contributed to the high scores include the integration of the AI-based virtual assistant (Gemini), the personalization of learning through the initial assessment, the achievement and reward system, the gamification, and the feedback provided by the assistant in different sections of the application. However, it is important to recognize that the validation tests were conducted with a limited sample of 50 university students, which, while re-presenting a manageable sample, needs to be expanded in future studies to improve the generalizability of the results, in addition, the study focused only on three financial topics, so the scalability to other content is yet to be validated. It would also be valuable to make comparisons with traditional methods of financial education and evaluate the impact of prolonged use of the application on users' real decision-making.\u003c/p\u003e"},{"header":"6. Conclusion","content":"\u003cp\u003eThis study presents the technical development and implementation of a mobile application, which is supported by a user-centered pedagogical approach, personalized learning elements, game mechanics, and emerging technologies to promote responsible financial decision making.\u003c/p\u003e\u003cp\u003eAs main findings, the application shows high user acceptance and satisfaction with its use, as well as a positive perception in pragmatic and hedonic terms. These support the effectiveness of the strategy employed, which integrates diagnostic phases, interactive challenges, and personalized feedback. Likewise, the modular structure of the system, the progressive design of the levels, and the incorporation of the virtual assistant as decision support prove to be key factors for the success of this type of educational tool.\u003c/p\u003e\u003cp\u003eFinally, this study provides an innovative user-based solution that contributes to the field of digital educational technologies applied to financial education, where it promotes a functional, scalable, and validated architecture that can be adapted to other training contexts and educational levels.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eConsent Statement: The participants involved in the study were individuals aged 18 to 25, all belonging to the specific target segment of the research. No minors were involved. All participants provided informed consent prior to their participation, in accordance with the ethical guidelines approved by the university.\u003c/p\u003e\n\u003cp\u003eAcknowledgements The researchers would like to express their gratitude to the Research Department of the Universidad Peruana de Ciencias Aplicadas for funding this research.\u003c/p\u003e\n\u003cp\u003eAuthor\u0026rsquo;s contributions: Angie Ruiz wrote the article and led the data analysis and interpretation. Survey design and data collection were carried out by Angie Ruiz and Juliana Yauricasa, and both contributed to the development of the mobile application. Juliana Yauricasa contributed to the development of the diagrams. Juan Morales contributed to the supervision, revision and submission of the paper. All authors contributed to the manuscript and have read and approved the final version.\u003c/p\u003e\n\u003cp\u003eFunding: Provided by the Universidad Peruana de Ciencias Aplicadas\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eData availability: The data sets used and analyzed during this study are available upon request from the corresponding author.\u003c/p\u003e\n\u003cp\u003eCode availability: The code used to develop the backend, frontend, and virtual assistant is available upon request from the corresponding author.\u003c/p\u003e\n\u003cp\u003eConflict of interest: The authors declare that they have no conflict of interest\u003c/p\u003e\n\u003cp\u003eOriginality and review status: The authors declare that this manuscript is original and has not been published in any other journal.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAcharya, D. B., Divya, B., \u0026amp; Kuppan, K. (2024). Explainable and Fair AI: Balancing Performance in Financial and Real Estate Machine Learning Models. IEEE Access, 12, 154022\u0026ndash;154034. https://doi.org/10.1109/ACCESS.2024.3484409\u003c/li\u003e\n\u003cli\u003eAkour, I., Alzyoud, M., Alquqa, E. K., Tariq, E., Alzboun, N., Al-Hawary, S. I. S., \u0026amp; Alshurideh, M. T. (2024). Artificial intelligence and financial decisions: Empirical evidence from developing economies. International Journal of Data and Network Science, 8(1), 101\u0026ndash;108. https://doi.org/10.5267/j.ijdns.2023.10.013\u003c/li\u003e\n\u003cli\u003eAsmawi, A., Saifulbahri, E. M., \u0026amp; Arifin, N. A. M. (2024). Development of BlockScholar as an Educational Mobile Application on Blockchain Technology. Journal of Advanced Research in Applied Sciences and Engineering Technology, 34(1), 15\u0026ndash;23. https://doi.org/10.37934/araset.34.1.1523\u003c/li\u003e\n\u003cli\u003eBj\u0026ouml;rklund, M., \u0026amp; Sandahl, J. (2023). Teaching and learning financial literacy within social studies\u0026ndash;a case study on how to realise curricular aims and ambitions. Journal of Curriculum Studies, 55(3), 325 \u0026ndash; 338. https://doi.org/10.1080/00220272.2023.2203771\u003c/li\u003e\n\u003cli\u003eBlanco, L. R., Hernandez, I., Thames, A. D., Chen, L., \u0026amp; Serido, J. (2023). Mind your money: A community-based digital intervention for improving financial capability among Hispanics. Journal of Economic Behavior \u0026amp; Organization, 212, 629\u0026ndash;643. https://doi.org/10.1016/J.JEBO.2023.05.030\u003c/li\u003e\n\u003cli\u003eBrooke, J. (1996). SUS \u0026ndash; a quick and dirty usability scale (pp. 189\u0026ndash;194).\u003c/li\u003e\n\u003cli\u003eCampoli, L., Assonitis, A., Ciallella, M., Paciorri, R., Bonfiglioli, A., \u0026amp; Ricchiuto, M. (2022). UnDiFi-2D: An unstructured discontinuity fitting code for 2D grids. Computer Physics Communications, 271. https://doi.org/10.1016/j.cpc.2021.108202\u003c/li\u003e\n\u003cli\u003eDamij, N., \u0026amp; Damij, T. (2024). An Approach to Optimizing Kanban Board Workflow and Shortening the Project Management Plan. IEEE Transactions on Engineering Management, 71, 13266 \u0026ndash; 13273. https://doi.org/10.1109/TEM.2021.3120984\u003c/li\u003e\n\u003cli\u003eErdem, O., Hassett, K., \u0026amp; Egriboyun, F. (2025). Hallucination in AI-generated financial literature reviews: evaluating bibliographic accuracy. International Journal of Data Science and Analytics. https://doi.org/10.1007/s41060-025-00731-0\u003c/li\u003e\n\u003cli\u003eFrench, D., McKillop, D., \u0026amp; Stewart, E. (2021). Personal finance apps and low-income households. Strategic Change, 30(4), 367\u0026ndash;375. https://doi.org/10.1002/jsc.2430\u003c/li\u003e\n\u003cli\u003eFrisancho, V., Herrera, A., \u0026amp; Prina, S. (2023). Can a mobile-app-based behavioral intervention teach financial skills to youth? Experimental evidence from a financial diaries study. Journal of Economic Behavior and Organization, 214, 595\u0026ndash;614. https://doi.org/10.1016/j.jebo.2023.08.014\u003c/li\u003e\n\u003cli\u003eGrijalvo, M., Segura, A., \u0026amp; N\u0026uacute;\u0026ntilde;ez, Y. (2022). Computer-based business games in higher education: A proposal of a gamified learning framework. Technological Forecasting and Social Change, 178. https://doi.org/10.1016/j.techfore.2022.121597\u003c/li\u003e\n\u003cli\u003eHadi, S. H., Permanasari, A. E., Hartanto, R., Sakkinah, I. S., Sholihin, M., Sari, R. C., \u0026amp; Haniffa, R. (2022). Developing augmented reality-based learning media and users\u0026rsquo; intention to use it for teaching accounting ethics. Education and Information Technologies, 27(1), 643\u0026ndash;670. https://doi.org/10.1007/s10639-021-10531-1\u003c/li\u003e\n\u003cli\u003eHassan, J. U., Missen, M. M. S., Firdous, A., Maham, A., \u0026amp; Ikram, A. (2023). An Adaptive M-Learning Usability Model for Facilitating M-Learning for Slow Learners. International Journal of Interactive Mobile Technologies, 17(19), 48\u0026ndash;69. https://doi.org/10.3991/ijim.v17i19.42153\u003c/li\u003e\n\u003cli\u003eHean, O., Saha, U., \u0026amp; Saha, B. (2025). Can AI help with your personal finances? Applied Economics. https://doi.org/10.1080/00036846.2025.2450384\u003c/li\u003e\n\u003cli\u003eHundhausen, C., Conrad, P., Adesope, O., \u0026amp; Tariq, A. (2023). Combining GitHub, Chat, and Peer Evaluation Data to Assess Individual Contributions to Team Software Development Projects. ACM Transactions on Computing Education, 23(3). https://doi.org/10.1145/3593592\u003c/li\u003e\n\u003cli\u003eImam, T., Cowling, M., \u0026amp; Das, N. (2022). Designing Computer Games to Teach Finance and Technical Concepts in an Online Learning Context: Potential and Effectiveness. Mathematics, 10(22). https://doi.org/10.3390/math10224205\u003c/li\u003e\n\u003cli\u003eIwaya, L. H., Nordin, A., Fritsch, L., B\u0026oslash;r\u0026oslash;sund, E., Johansson, M., Varsi, C., \u0026amp; \u0026Auml;ngeby, K. (2023). Early Labour App: Developing a practice-based mobile health application for digital early labour support. International Journal of Medical Informatics, 177. https://doi.org/10.1016/j.ijmedinf.2023.105139\u003c/li\u003e\n\u003cli\u003eLew, C., \u0026amp; Saville, A. (2021). Game-based learning: Teaching principles of economics and investment finance through Monopoly. International Journal of Management Education, 19(3). https://doi.org/10.1016/j.ijme.2021.100567\u003c/li\u003e\n\u003cli\u003eLi, Q., Zhao, J., Yan, R., Gao, Q., Zhen, Y., Li, X., Liang, Y., Min, S., \u0026amp; Yang, L. (2024). WeChat mini program in laboratory biosafety education among medical students at Guangzhou Medical University: a mixed method study of feasibility and usability. BMC Medical Education, 24(1). https://doi.org/10.1186/s12909-024-05131-9\u003c/li\u003e\n\u003cli\u003eLisana, L., Dinata, H., \u0026amp; Valencia Tanudjaja, G. (2025). Playing to learn: Game-based approach to financial literacy for generation Z. Entertainment Computing, 52, 100896. https://doi.org/10.1016/J.ENTCOM.2024.100896\u003c/li\u003e\n\u003cli\u003eLiu, T.-Y., Jiang, Y.-H., Wei, Y., Wang, X., Huang, S., \u0026amp; Dai, L. (2024). Educational Practices and Algorithmic Framework for Promoting Sustainable Development in Education by Identifying Real-World Learning Paths. Sustainability (Switzerland), 16(16). https://doi.org/10.3390/su16166871\u003c/li\u003e\n\u003cli\u003eLučić, A., \u0026amp; Uzelac, M. (2024). Investigating alternative avenues for financial behaviour change: moving beyond the traditional approach. Young Consumers. https://doi.org/10.1108/YC-05-2023-1748\u003c/li\u003e\n\u003cli\u003eMabwe, K., Aminu, N., Ivanov, S. H., \u0026amp; Dimov, D. (2025). Generative artificial intelligence chatbots in investment decision-making: a phantom menace or a new hope? Foresight, 27(4), 820 \u0026ndash; 863. https://doi.org/10.1108/FS-06-2024-0122\u003c/li\u003e\n\u003cli\u003eMalik, A. (2023). Assessing the Effectiveness of Financial Literacy Mobile Apps Using the Content Analysis Approach. International Journal of Interactive Mobile Technologies, 17(23), 68\u0026ndash;84. https://doi.org/10.3991/IJIM.V17I23.42213\u003c/li\u003e\n\u003cli\u003eMarques, M. M., \u0026amp; Pombo, L. (2023). User Experience of a Mobile App in a City Tour Game for International Doctoral Students. Education Sciences, 13(12). https://doi.org/10.3390/educsci13121221\u003c/li\u003e\n\u003cli\u003eMayo-Alvarez, L., Del-Aguila-Arcentales, S., Alvarez-Risco, A., Chandra Sekar, M., Davies, N. M., \u0026amp; Y\u0026aacute;\u0026ntilde;ez, J. A. (2024). Innovation by integration of Drum-Buffer-Rope (DBR) method with Scrum-Kanban and use of Monte Carlo simulation for maximizing throughput in agile project management. Journal of Open Innovation: Technology, Market, and Complexity, 10(1). https://doi.org/10.1016/j.joitmc.2024.100228\u003c/li\u003e\n\u003cli\u003ePitthan, F., \u0026amp; Witte, K. De. (2025). Game over or continue? How gamification can improve completion rate in adaptive learning. Education and Information Technologies, 30(3), 2757\u0026ndash;2786. https://doi.org/10.1007/S10639-024-12928-0\u003c/li\u003e\n\u003cli\u003ePlatz, L., \u0026amp; Zauner, M. (2025). Financial Literacy Games\u0026mdash;Increasing Utility Value by Instructional Design in Upper Secondary Education. Education Sciences, 15(2). https://doi.org/10.3390/educsci15020227\u003c/li\u003e\n\u003cli\u003eQatawneh, A. M., Lutfi, A., \u0026amp; Al Barrak, T. (2024). Effect of Artificial Intelligence (AI) on Financial Decision-Making: Mediating Role of Financial Technologies (Fin-Tech). HighTech and Innovation Journal, 5(3), 759\u0026ndash;773. https://doi.org/10.28991/HIJ-2024-05-03-015\u003c/li\u003e\n\u003cli\u003eRamli, R. Z., Sahari Ashaari, N., Mat Noor, S. F., Noor, M. M., Yadegaridehkordi, E., Abd Majid, N. A., Dahlan, H. A., \u0026amp; Abdul Wahab, A. N. (2024). Designing a mobile learning application model by integrating augmented reality and game elements to improve student learning experience. Education and Information Technologies, 29(2), 1981\u0026ndash;2008. https://doi.org/10.1007/s10639-023-11874-7\u003c/li\u003e\n\u003cli\u003eRazak, R. N. I. A., \u0026amp; Senan, N. (2022). Mobile Learning for Manually Coded Malay Sign Language Using Augmented Reality. Journal of Soft Computing and Data Mining, 3(1), 86\u0026ndash;94. https://doi.org/10.30880/jscdm.2022.03.01.009\u003c/li\u003e\n\u003cli\u003eReisdorfer-da-Silva, R. C., Becker, K. L., \u0026amp; Vieira, K. M. (2025). The impact of board games on the financial literacy of public-school students. Journal of Behavioral and Experimental Economics, 114, 102331. https://doi.org/10.1016/J.SOCEC.2024.102331\u003c/li\u003e\n\u003cli\u003eRobb, C. A., \u0026amp; Chy, S. (2023). Undergraduate financial knowledge, attitudes, and behaviors: The impact of financial life skills course on college students. Financial Planning Review, 6(1). https://doi.org/10.1002/cfp2.1155\u003c/li\u003e\n\u003cli\u003eRuiz, A. N., Yauricasa, J. A., \u0026amp; Morales, J. C. (2024). Design of a Mobile Learning App for Financial Literacy in Young People Using Gamification. IJACSA) International Journal of Advanced Computer Science and Applications, 15(12). www.ijacsa.thesai.org\u003c/li\u003e\n\u003cli\u003eSchrepp, M., Hinderks, A., \u0026amp; Thomaschewski, J. (2017). Construction of a Benchmark for the User Experience Questionnaire (UEQ). International Journal of Interactive Multimedia and Artificial Intelligence, 4, 40\u0026ndash;44. https://doi.org/10.9781/ijimai.2017.445\u003c/li\u003e\n\u003cli\u003eSinnewe, E., \u0026amp; Nicholson, G. (2023). Healthy financial habits in young adults: An exploratory study of the relationship between subjective financial literacy, engagement with finances, and financial decision-making. Journal of Consumer Affairs, 57(1), 564 \u0026ndash; 592. https://doi.org/10.1111/joca.12512\u003c/li\u003e\n\u003cli\u003eSujith, A. V. L. N., Qureshi, N. I., Dornadula, V. H. R., Rath, A., Prakash, K. B., \u0026amp; Singh, S. K. (2022). A Comparative Analysis of Business Machine Learning in Making Effective Financial Decisions Using Structural Equation Model (SEM). Journal of Food Quality, 2022. https://doi.org/10.1155/2022/6382839\u003c/li\u003e\n\u003cli\u003eTorma, J., Barbić, D., \u0026amp; Ivanov, M. (2023). Analyzing the Effects of Financial Education on Financial Literacy and Financial Behaviour: A Randomized Field Experiment in Croatia. South East European Journal of Economics and Business, 18(2), 63\u0026ndash;86. https://doi.org/10.2478/jeb-2023-0019\u003c/li\u003e\n\u003cli\u003eY\u0026aacute;nez-P\u0026eacute;rez, I., Toma, R. B., \u0026amp; Meneses-Villagr\u0026aacute;, J. \u0026Aacute;. (2024). Design and usability of IndagApp: an app for inquiry-based science education | Dise\u0026ntilde;o y usabilidad de IndagApp: una app para la ense\u0026ntilde;anza de las ciencias por indagaci\u0026oacute;n. RIED-Revista Iberoamericana de Educacion a Distancia, 27(2). https://doi.org/10.5944/ried.27.2.39109\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Universidad Peruana de Ciencias Aplicadas","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Mobile Learning, Gamification, Financial Literacy, Artificial Intelligence, Virtual Assistant, Personalized learning, Educational Technology","lastPublishedDoi":"10.21203/rs.3.rs-7170120/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7170120/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study presents the implementation of a mobile application that integrates gamification and artificial intelligence to strengthen financial literacy in college students. The proposed solution combines a user-centered approach, educational and structured content, and interactive challenges, integrating a virtual assistant powered by Gemini AI. The application is composed of three stages: an initial diagnostic assessment, gamified challenges, and a final assessment. The system offers personalized feedback, financial topics with levels of difficulty, and a system of achievements and rewards, which seeks to enhance engagement and learning outcomes. The development stage was carried out under the Scrum framework over eight sprints, where Unity and FastAPI were used for frontend and backend development, respectively. On the other hand, usability and user experience were evaluated using two instruments: the System Usability Scale (SUS) and the User Experience Questionnaire (UEQ). The results obtained from the tests with 50 university students show a high usability (SUS\u0026thinsp;=\u0026thinsp;91.9) and a positive experience in all the dimensions evaluated by the UEQ, which confirms the effectiveness of the application. This work contributes to the field of educational technologies by providing a scalable and validated architecture for financial education through mobile learning.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e","manuscriptTitle":"Development of a Personalized Gamified Mobile Learning App Using an AI-Based Virtual Assistant for Financial Literacy","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-24 09:15:27","doi":"10.21203/rs.3.rs-7170120/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"06fbdc15-a8e4-42b8-a621-45cbc0a73d5a","owner":[],"postedDate":"July 24th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":51811062,"name":"Special Education"},{"id":51811063,"name":"Finance"}],"tags":[],"updatedAt":"2025-07-24T09:15:28+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-24 09:15:27","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7170120","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7170120","identity":"rs-7170120","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}