Learning Time Research Articles

Learning from interactive video: the influence of self-explanations, navigation, and cognitive load

AbstractRecent research has shown that enhancing instructional videos with questions, such as self-explanation prompts, and thus shifting the process from receptive to constructive learning, is beneficial to learning. However, the inclusion of questions is often confounded with the implementation of learner pacing through navigation features. Furthermore, previous studies have often not controlled for learning time. To address these shortcomings, an experiment (N = 128) was conducted. Participants watched an instructional video about cloud formation and lightning, with learning time controlled. In a 2 × 2 between-subjects design, navigation features (learner pacing vs. system pacing) and self-explanation prompts (prompts vs. no prompts) were manipulated. The results showed no effects of navigation features and self-explanation prompts on learning performance. While navigation features did not affect cognitive load, self-explanation prompts increased both intrinsic and extraneous cognitive load. Overall, the quality of responses to prompts was low but positively related to comprehension. The results are discussed in terms of the Interactive-Constructive-Active-Passive framework and Cognitive Load Theory. They highlight the importance of boundary conditions when investigating the effects of interactive features in instructional videos.

Working students’ academic performance on flexible learning

The purpose of this study was to determine the working students’ academic performance on flexible learning. It looked into the relationship between the working students’ academic performance and flexible learning. It also explored the difference between the students’ academic performance and their profile. This explanatory-sequential study involved 166 working students of Baliwag Polytechnic College. Findings revealed that highly significant correlations were found between respondents’ perceptions on flexible learning and their academic performance. Further, student-respondents affirmed that their self-disciplined was developed in self-regulated learning in this time of flexible learning and they believed that they get to save money from transportation and be able to use it for other school-related dues. Also, significant differences existed between the respondents’ gender, task performance, and general weighted average. Further analysis revealed that highly significant differences existed between the respondents’ track, year level, and general weighted average. Findings also disclosed that no significant differences were found among the respondents’ academic performance, sibling position and hours worked per week. Hence, highly significant differences were revealed between the respondents’ strand, monthly family income, and general weighted average.

Accelerated Transfer Learning for Cooperative Transportation Formation Change via SDPA-MAPPO (Scaled Dot Product Attention-Multi-Agent Proximal Policy Optimization)

A method for cooperative transportation, which required formation change in a traveling environment, is gaining interest. Deep reinforcement learning is used in formation changes for multi-robot cases. The MADDPG (Multi-Agent Deep Deterministic Policy Gradient) method is popularly used for recognized environments. On the other hand, re-learning may be required in unrecognized circumstances by using the MADDPG method. Although the development of MADDPG using model-based learning and imitation learning has been applied to reduce learning time, it is unclear how the learning results are transferred when the number of robots changes. For example, in the GASIL-MADDPG (Generative adversarial self-imitation learning and Multi-agent Deep Deterministic Policy Gradient) method, how the results of three robot training can be transferred to the four robots’ neural networks is uncertain. Nowadays, Scaled Dot Product Attention (SDPA) has attracted attention and is highly impactful for its speed and accuracy in natural language processing. When transfer learning is combined with fast computation, the efficiency of edge-level re-learning is improved. This paper proposes a formation change algorithm that allows easy and fast multi-robot knowledge transfer using SDPA combined with MAPPO (Multi-Agent Proximal Policy Optimization), compared to other methods. This algorithm applies SDPA to multi-robot formation learning and performs fast learning by transferring the acquired knowledge of formation changes to a certain number of robots. The proposed algorithm is verified by simulating the robot formation change and was able to achieve dramatic high-speed learning capabilities. The proposed SDPA-MAPPO (Scaled Dot Product Attention-Multi-Agent Proximal Policy Optimization) learned 20.83 times faster than the Deep Dyna-Q method. Furthermore, using transfer learning from a three-robot to five-robot case, the method shows that the learning time can be reduced by about 56.57 percent. A scenario of three-robot to five-robot is chosen based on the number of robots often used in cooperative robots.

Research on Product Classification and Demand Prediction Based on ALSTM-CNN

The rapid development of e-commerce platforms has prompted a large number of retailers to join and store goods through logistics warehouses provided by the platform, which are then centrally managed by the platform. In this context, the rise of artificial intelligence and machine learning technology has brought revolutionary impact to the e-commerce industry. Under this framework, accurate sales forecasting has become the core link. This article aims to integrate machine learning models and time series analysis, aiming to improve the accuracy of sales data prediction. This article is based on the K-Means algorithm for unsupervised multi-level clustering, determining the optimal K value through elbow rules and contour coefficients. After labeling the classification data with class labels, objectively evaluate the clustering effect by combining the Calinski Harabasz index and visualizing SKUs after each clustering. Finally, using evaluation metric 1-wmape, the performance of ARIMA-ANN, BiLSTM, GRU, CNN-LSTM, Stack LSTM, and three tree models in predicting SKU sales was compared. We compared the accuracy of 15 day recursive prediction and 15 day one-time output, and selected sliding window sizes of 3, 5, 7, and 14. After comparison, it was determined that the BiLSTM model performed best in predicting 15 day sales volume and a 3-day sliding time window under multiple sample categories; ARIMA-ANN performs the best in recursive prediction of 15 day sales volume in a few sample categories. By adjusting the ARIMA parameters through ACF and PACF, the maximum predictive performance is ensured.

A Shortest Distance Priority UAV Path Planning Algorithm for Precision Agriculture

Unmanned aerial vehicles (UAVs) have made significant advances in autonomous sensing, particularly in the field of precision agriculture. Effective path planning is critical for autonomous navigation in large orchards to ensure that UAVs are able to recognize the optimal route between the start and end points. When UAVs perform tasks such as crop protection, monitoring, and data collection in orchard environments, they must be able to adapt to dynamic conditions. To address these challenges, this study proposes an enhanced Q-learning algorithm designed to optimize UAV path planning by combining static and dynamic obstacle avoidance features. A shortest distance priority (SDP) strategy is integrated into the learning process to minimize the distance the UAV must travel to reach the target. In addition, the root mean square propagation (RMSP) method is used to dynamically adjust the learning rate according to gradient changes, which accelerates the learning process and improves path planning efficiency. In this study, firstly, the proposed method was compared with state-of-the-art path planning techniques (including A-star, Dijkstra, and traditional Q-learning) in terms of learning time and path length through a grid-based 2D simulation environment. The results showed that the proposed method significantly improved performance compared to existing methods. In addition, 3D simulation experiments were conducted in the AirSim virtual environment. Due to the complexity of the 3D state, a deep neural network was used to calculate the Q-value based on the proposed algorithm. The results indicate that the proposed method can achieve the shortest path planning and obstacle avoidance operations in an orchard 3D simulation environment. Therefore, drones equipped with this algorithm are expected to make outstanding contributions to the development of precision agriculture through intelligent navigation and obstacle avoidance.

Intelligence-assisted college English teaching: The application of artificial intelligence technology in personalized learning path design

English learning is an integrated learning process in which listening, speaking, reading, writing and other parts are interrelated. Because of this feature of English learning, higher requirements are put forward for the design of English learning assistance systems. On the other hand, the wide application of advanced science and technology such as automation and cloud computing in all walks of life has promoted the development of society into the era of artificial intelligence. Under such a large development background, how to build a multifunctional integrated English learning system to realize the personalized learning of learners with different levels and different needs has become a focus of current English teaching research. With the increasing abundance of English learning materials, it has become very important to help users find suitable English learning resources from the massive materials, which directly affects the cost of learning time and interest of users. Based on this, this paper introduces a multi-similarity ranking model to achieve personalized design, and then uses artificial intelligence technology to design and implement a personalized intelligent assistant system for college English teaching.

Teaching for transfer of learning in health professions education: AMEE Guide No. 176

Transfer of learning occurs when past learning is applied to new situations, and also at a varying time from the initial time of learning. Importantly, research in both academic and clinical areas of health professions education has highlighted that transfer of learning often does not successfully occur. Successful transfer is multi-dimensional and occurs when the learner has the required motivation, mental model, metacognitive processes relevant to the task, and the opportunity to transfer their learning to different situations. An essential aspect of successful transfer is the educator. This Guide provides an overview of an integrated model of transfer that can inform a variety of practical teaching strategies in both academic and clinical areas of health professions education.

Teaching analysis of fortified monuments in a time of remote learning

First COVID-19 then the war in Ukraine forced educational institutions to adapt to online and hybrid teaching methods in new realms. Despite the challenges presented by this, some educators have found innovative ways to teach in hybrid mode, such as using new technologies or creative teaching methods. This article shows the method used by Olha Tikhonova and Oresta Remeshylo-Rybchynska to engage students in difficult situations with different kinds of restrictions and challenges, both during the COVID-19 restrictions and then after Russia’s invasion of Ukraine from 24 February 2022. The method is rooted in Olha Tikhonova’s comprehensive approach to analyzing case studies focused on castles, and the teaching methodology is built around this approach. The method is suitable for use in a partially or entirely virtual teaching environment. It is easily adaptable to interactive communication with participants through web meetings.

Function Placement for In-network Federated Learning

Federated learning (FL), particularly when data is distributed across multiple clients, helps reducing the learning time by avoiding training on a massive pile-up of data. Nonetheless, low computation capacities or poor network conditions can worsen the convergence time, therefore decreasing accuracy and learning performance. In this paper, we propose a framework to deploy FL clients in a network, while compensating end-to-end time variation due to heterogeneous network setting. We present a new distributed learning control scheme, named In-network Federated Learning Control (IFLC), to support the operations of distributed federated learning functions in geographically distributed networks, and designed to mitigate the stragglers with lower deployment costs. IFLC adapts the allocation of distributed hardware accelerators to modulate the importance of local training latency in the end-to-end delay of federated learning applications, considering both deterministic and stochastic delay scenarios. By extensive simulation on realistic instances of an in-network anomaly detection application, we show that the absence of hardware accelerators can strongly impair the learning efficiency. Additionally, we show that providing hardware accelerators at only 50% of the nodes, can reduce the number of stragglers by at least 50% and up to 100% with respect to a baseline FIRST-FIT algorithm, while also lowering the deployment cost by up to 30% with respect to the case without hardware accelerators. Finally, we explore the effect of topology changes on IFLC across both hierarchical and flat topologies.

Fracture mode classification of UHPC based on deep learning and wavelet time–frequency spectrum derived from acoustic emission waveform

Fracture mode categorization is essential in identifying the failure stage, evaluating the damage characteristics, and providing references for structural maintenance. Acoustic emission (AE), as a passive nondestructive technique, presents great potential in damage characterization of cementitious materials and structures. In this study, a convolutional neural network (CNN) and gated recurrent unit (GRU) integrated model for tensile, shear, and mixed cracking mode classification of ultra-high-performance concrete (UHPC) was proposed and verified based on AE waveform features. First, four-point bending and direct shear tests for UHPC specimens were carried out, respectively, to generate AE waveform signals corresponding to different cracking modes (tensile, shear, and mixed ones). Subsequently, the collected one-dimensional AE signals were converted into two-dimensional time–frequency spectrum by continuous wavelet transform. And an AE dataset was constructed after manual labeling based on the corresponding relationships between cracking modes and wavelet time–frequency spectra. Then, an automatic CNN-GRU automatic classification model was established using wavelet time–frequency spectrum images as input variables, where the CNN extracts spatial image features, and the GRU screens time dependencies and realizes final classification. The proposed CNN-GRU model achieves an accuracy of 96.17% for cracking mode classification of UHPC, which is superior both in computational accuracy and efficiency to lightweight CNN and AlexNet. Classification outcomes for UHPC specimen under four-point bending and direct shear tests revealed the proportion of tensile cracks progressively decreased with the increase of damage, while shear cracks exhibited the opposite trend. The percentage of mixed mode cracks maintained relatively stable throughout the failure stages. The model proposed in this study presents favorable performance to reveal the underlying damage evolution mechanism during UHPC fracture, which can also provide references for the cracking mode classification of other cementitious materials and structures.

Analysis of the Effectiveness of the Teaching at the Right Level (TaRL) Approach in Improving Students' Literacy Skills through the 15-Minute Literacy Program at Primary School

This research aims to analyze the effectiveness of the Teaching at the Right Level (TaRL) approach in improving students' literacy skills through a 15-minute literacy program. This approach is based on grouping students according to their ability levels, allowing teaching strategies to be tailored to students' needs. The method used in this study is descriptive with a qualitative approach, where data is collected through observation, interviews, and documentation. Students are grouped based on their reading abilities: beginner, letters, words, paragraphs, and stories. The instruments used in this study are initial diagnostic and developmental tests conducted orally at the beginning and during the research process. These tests measure the overall development of students' reading skills over time. Data analysis is conducted to evaluate students' engagement and their responses to short but intensive literacy activities. The results indicate a significant improvement in students' literacy skills at all reading ability levels, both in comprehension and reading fluency. This improvement suggests that the TaRL approach can effectively enhance elementary school literacy skills. This research is expected to significantly contribute to developing more effective literacy strategies in elementary schools to optimize short yet effective learning time.

The evolution of medical education in the era of Covid-19 and beyond: a longitudinal study

BackgroundThe COVID-19 pandemic has brought about profound transformations in nearly all aspects of life, leaving its impact on the global community as a whole. Nowhere has this transformation been more pronounced than in the sphere of education, including medical education. Healthcare professionals and educators faced the daunting task of preparing the next generation of practicing physicians amid the ongoing public health crisis.MethodsThe study involved students from three different Chinese medical universities (groups A, B, and C, respectively). The research employed two instruments: a Scale of Satisfaction with Simulation-Based Education and a Scale of Satisfaction with Telemedicine Competency Education. Data were collected at three different time points: 2020–2021 (online/distance learning), 2021–2022 (hybrid learning), and 2022–2023 (traditional face-to-face learning).ResultsIt was revealed that students demonstrated the highest level of satisfaction during the hybrid learning period (Time Point 2), while online/distance learning (Time Point 1) received the lowest ratings. Statistical analysis indicated significant differences in satisfaction levels across the various time points.ConclusionsHybrid learning emerged as the preferred method among students, yielding the highest level of satisfaction. Online/distance learning during the 2020–2021 academic year and traditional face-to-face instruction in 2022–2023 exhibited lower satisfaction levels. The research findings underscore the practical significance of hybrid learning for students in medical educational programs, suggesting its potential for optimizing educational curricula and resources within academic institutions.

Understanding faculty perspectives on health equity curriculum implementation in graduate medical education: a qualitative study

BackgroundDespite increased awareness of persistent healthcare disparities, integrating health equity education into medical residency curricula needs urgent expansion. The Accreditation Council for Graduate Medical Education (ACGME) mandates addressing healthcare disparities and social determinants of health, but implementation remains insufficient due to pervasive barriers.ObjectiveTo explore residency faculty perspectives at one institution surrounding implementing a health equity curriculum within their programs.MethodsBetween November 2022 and April 2023, qualitative interviews and focus groups were conducted with residency faculty participants (n = 13) from a large academic medical institution. Participants were recruited via an internal email list directory of program directors. Semi-structured sessions explored faculty perspectives on health equity curriculum implementation. Data analysis utilized template analysis and thematic coding.ResultsFaculty perceived barriers included limited time for teaching health equity, challenges in prioritizing health equity content within the curriculum, lack of formal training and expertise, discomfort in leading discussions, and institutional barriers such as resource allocation and lack of buy-in.ConclusionsFindings underscored the need for institutionally supported faculty development and protected learning time for health equity education. Despite participant recruitment and representation limitations, the study provides valuable insights for addressing curricular challenges. This study highlights critical areas for improving health equity curriculum implementation in graduate medical education. Addressing these faculty barriers can enhance resident competency in identifying and addressing healthcare disparities.Trial registrationThe study received approval from the Thomas Jefferson Institutional Review Board (IRB #22E.762).

Research on Joint Forecasting Technology of Cold, Heat, and Electricity Loads Based on Multi-Task Learning

The cooperative optimization and dispatch operation of the integrated energy system (IES) depends on accurate load forecasts. A multivariate load, joint prediction model, based on the combination of multi-task learning (MTL) and dynamic time warping (DTW), is proposed to address the issue of the prediction model’s limited accuracy caused by the fragmentation of the multivariate load coupling relationship and the absence of future time series information. Firstly, the MTL model, based on the bidirectional long short-term memory (BiLSTM) neural network, extracts the coupling information among the multivariate loads and performs the preliminary prediction; secondly, the DTW algorithm clusters and splices the load data that are similar to the target value as the input features of the model; finally, the BiLSTM-attention model is used for secondary prediction, and the improved Bayesian optimization algorithm is applied for adaptive selection of optimal hyperparameters. Based on the game-theoretic view of Shapley’s additive interpretation (SHAP), a model interpretation technique is introduced to determine the validity of the liquidity indicator and the asynchronous relationship between the significance of the indicator and its actual contribution. The prediction results show that the joint prediction model proposed in this paper has higher training speed and prediction accuracy than the traditional single-load prediction model.

FORECASTING FIRE RISK VALUES USING MACHINE LEARNING AND TIME SERIES ANALYSIS

The probability of occurrence of emergencies and their consequences can be minimized by predicting events that contribute to the emergence of such a situation. An optimal and useful tool for such prediction and forecasting can be modeling, which can be carried out on the basis of information technology using computer programs.The methods of modeling and forecasting emergencies related to fires are the same as for other emergencies, but there are some specifics. The forecast of emergencies related to fires is aimed at three main stages: 1. forecasting the probability of occurrence with the determination of the place of the highest probability of fire occurrence, time and factors contributing to it; 2. forecasting the development of the situation and the spread of the fire (direction, speed) when it occurs, the process itself; 3. forecasting the consequences of a fire at a specific facility.The forecast is made based on the analysis of all possible unfavorable factors, their fields of action, features of the facility, its fire hazard and other data.The use of machine learning methods and time series analysis for forecasting fire risk values is substantiated. The possibility of using statistical methods of mathematical modeling of fire occurrence as the basis of the machine learning model is noted. This approach enables the machine learning model to process and analyze a large amount of statistical information for specific time periods and issue forecasts.

Exploring the role of primary care pharmacists in NHS Lanarkshire following the implementation of the pharmacotherapy hub: a service evaluation

Abstract Introduction The General Medical Service (GMS) contract was produced in Scotland in an attempt to tackle General Practitioner (GP) workforce issues. The pharmacotherapy work stream is part of this and provides a framework for pharmacy teams, dividing services into level 1 (core), level 2 (advanced) and level 3 (specialist).1 A statement by the Royal Pharmaceutical Society (RPS) and the British Medical Association (BMA) discussed the introduction of pharmacotherapy hubs to complete level 1 work for multiple practices, centralising the medicines reconciliation process. Theoretically, this would release pharmacists to carry out clinical and more advanced/specialist services.2 There are currently 8 pharmacotherapy hubs within NHS Lanarkshire (NHSL), incorporating level 2 and 3 services into job plans for General Practice Clinical Pharmacists (GPCPs). No work has been undertaken to explore changes in practice since the implementation of these hubs. The qualitative study explored drivers and motivators for pharmacists in practice, shaping future pharmacotherapy services in NHSL and Scotland. Aim To explore the views and attitudes of primary care pharmacists towards their practice role in NHSL following the implementation of the pharmacotherapy hub. Methods Ethical approval was granted from the School of Pharmacy and Life Sciences Ethics Committee (Robert Gordon University). Pharmacists working within the 8 hub localities were invited to participate via email correspondence if they had over 6 months experience in practice pre-hub implementation and then 6 months post-hub implementation. Participants attended a one-to-one interview to share their views towards their practice role since the hub was implemented. Interviews were recorded, transcribed verbatim and thematically analysed. Results Twelve pharmacists across six localities participated. The five key themes identified were job role, managing time, professional development, professional belonging and patient care. The implementation of the hub was an enabler for pharmacists to partake in advanced pharmacotherapy tasks. Pharmacists negatively perceived level 1 core pharmacotherapy tasks in comparison to level 3 clinical patient-facing time. Participants linked the hubs impact on practice with role development, an increase in autonomy, confidence and job satisfaction. Discussion/Conclusion The hub has enabled in-practice pharmacists to move to a more sustainable balance between core and advanced/specialist tasks. Collaboration is required between pharmacy teams and practices to understand the pharmacist role in pharmacotherapy. Further recommendations include continuing the roll-out of pharmacotherapy hubs, standardisation of pharmacist clinical time balanced with core tasks and implementation of protected learning time for pharmacy teams. The study was open to response bias but this was addressed by using semi-structure interview guides in the one-to-one interviews. Researcher bias was minimised by two researchers analysing the data and practicing reflexivity. The study was open to recall bias due to participant’s retrospective thinking. In NHSL prioritising the hub roll-out and collaboration with stakeholders are essential. This will ensure the balance between efficient pharmacotherapy services and maintaining attractive roles, contributing to a sustainable workforce.

Optimizing LLM Strategies for Playing Mendikot using Prompt Engineering

This study investigates the integration of a large language model (LLM) enhanced by prompt engineering and game theory to effectively engage in the strategic card game Mendikot. By refining complex prompts and leveraging a tailored visual understanding of game dynamics, we significantly bolster the decision-making prowess of the LLM. Our methodology involved the systematic simplification of game prompts to facilitate deeper learning and faster response times, coupled with the implementation of a visual recognition system to interpret and react to game states dynamically. The results illustrate that the adapted LLM outperforms traditional AI approaches in strategic decision-making tasks, underscoring a substantial improvement in both the accuracy and efficiency of game-play. This research not only demonstrates a viable model for enhancing AI interaction in recreational gaming but also opens avenues for deploying advanced AI strategies in complex strategic environments, offering insights into the broader application of AI in leisure and competitive arenas. The findings suggest that AI can transcend conventional gaming roles, potentially transforming strategic gameplay in digital and physical platforms.

Game-Based Learning and Gamification, Effective Innovation in Emergency Medicine Training Program

Background: Game-based learning (GBL) and gamification have recently been applied in medical education. Both approaches utilize methods to teach specific skills or knowledge in a more motivating manner than traditional teaching. Objectives: This study aimed to evaluate the impact of GBL and gamification in an emergency medicine (EM) residency training program by comparing education through e-learning with GBL and gamification. Methods: We randomly enrolled EM residents into two groups. Over the course of one educational year, one group of residents participated in a GBL and gamification course (10 designed GBL and gamification sessions), while the other group received e-learning materials (10 designed podcasts and videos). Traditional training programs continued as routine in both groups. After the intervention, we assessed learners' educational achievements (exam scores) in both groups. Additionally, EM residents' viewpoints (in the first group) on the innovative program were collected using the model for the evaluation of educational games (MEEGA). Finally, the short version of the postgraduate hospital educational environment measure (PHEEM) was administered to evaluate learners' general feedback (in the first group) about the new educational environment after the intervention. Results: Game-based learning and gamification significantly improved scores in written, oral, and electronic exams assessing knowledge, skills, clinical reasoning, and judgment among EM residents (P-value < 0.05). All MEEGA items under “Usability” and “Player Experience” had mean scores > 3 (indicating agree and totally agree) (median score = 3 - 4). All PHEEM items under “Teacher Support and Skills” and “Work Conditions and Allocated Time for Learning” also had mean scores > 3 (indicating agree and totally agree) (median score = 4). Conclusions: Game-based learning and gamification enhanced EM residents’ knowledge, skills, clinical reasoning, and judgment. Learners largely agreed on the positive impact of the intervention on their residency training program.

Evidence for a general cognitive structure in pigeons (Columba livia)

A well replicated result in humans is that performance, whether good or bad, is consistent across a wide variety of cognitive tasks. Factor analysis extracts one factor that can account for approximately half of the variance in performance. This factor is termed g and almost all cognitive tasks positively load onto this factor. While some neurobiological correlates of g have been identified in humans, causal experiments are only feasible in animals. When mice and some avian species are assessed with cognitive test batteries, performance positively correlates, and the first component extracted has similar properties to g. There are some limitations to the species tested thus far, including comparability in the cognitive domains assessed. The pigeon is an ideal subject to overcome these issues since pigeons, humans, and other primates are frequently given similar tasks and many neural correlates of performance have been identified in the pigeon. We created a test battery that assessed different domains, including associative learning, memory, cognitive flexibility, and reaction time. When all tasks were included, there was evidence for a two-component structure that was influenced by subjects’ age. When the reaction time task was excluded, there was a g-like component. The implications for these results when constructing future test batteries and comparing across species are discussed.

대학생의 생성형 AI 이용경험에 관한 현상학적 연구

Objectives This study aims to explore the direction of Generative AI literacy education through the exploration of college students' experiences using Generative AI. Methods For this purpose, written and face-to-face in-depth interviews were conducted with 12 university stu-dents (5 male students and 7 female students) and analyzed by applying Colaizzi's phenomenological method, which consists of seven steps to explore the nature of common experiences rather than the individual participants. Results As a result of analyzing the interview data, 5 semantic themes and 14 sub-themes were derived under the essence of the experience called ‘co-evolution’. The five semantic themes consist of ‘getting to know new media’, ‘useful but dangerous existence’, ‘sharing roles with Generative AI’, ‘adapting to changes in the learning environment’, and ‘seeking a life that coexists with Generative AI’. The 14 semantic themes consisted of ‘matching up with Generative AI’, ‘creating my own method of using prompts’, ‘limitation of inanimate object’, ‘repeating unnecessary learning activities’, ‘incomplete learning assistant’, ‘my own tutor’, ‘reducing learning time’, ‘setting my own scope of use’, ‘another team member’, ‘passive new media user’, ‘recognition of reality and acceptance of new media’, ‘improving the competence to use Generative AI’, ‘selecting Generative AI according to the purpose of use’, and ‘competition with the human-specific domain’. Conclusions The discussion points on the direction of Generative AI literacy education for university students are as follows. First, it provides education in which prompts can be effectively input. Second, it provides guidelines so that the Generative AI can be used as an auxiliary tool rather than the main tool for learning. Third, it provides education on the characteristics of the Generative AI. Fourth, it enhances education on usage ethics.