Real Classroom Research Articles

Digital ambitions vs. classroom reality in Norwegian lower secondary schools: What digital competencies are students developing over time?

Despite high political ambitions, the increased use of digital technology in lower secondary school has resulted in little indication of the advanced use of digital skills. Drawing on video-recorded lessons (N = 29) from five schools in Norway over time and across two national curricula, this study mapped the opportunities students had to develop digital skills in English classrooms. The findings provide nuanced insight into the instruction of digital skills in highly digitalised schools. Despite a slight increase in digital skills after the 2020 reform, our findings indicate that digital ambitions in the curriculum were not reflected in classroom reality.

The influence of emotions on science teaching: a case study with three early childhood education teachers

Teaching science in early childhood education (ECE) is crucial for fostering essential scientific thinking skills, which are vital for young learners’ development and formative trajectories. While this educational level presents exciting opportunities for integrating science teaching, it also poses challenges related to teachers ́ cognitive, pedagogical, and emotional constraints. Recognizing the pivotal role of teachers’ emotions in shaping their practices, we aim to specifically examine the influence of teachers’ emotions on science teaching in ECE, an area that remains underresearched. We conducted a multiple case study involving three ECE teachers in La Plata, Argentina, who demonstrated a strong interest in improving their pedagogical practices despite acknowledging their difficulties and insecurities in teaching science. We explored their emotions and how they relate to science lesson planning and implementation through surveys, in-depth interviews, and observations. Data collection was designed in three stages: pre-active, interactive, and post-active teaching. First, we surveyed the range of negative and positive emotions that teachers experienced toward sciences, rooted in their personal and formative prior experiences, and examined how these emotions influenced their planning (pre-active stage). Subsequently, we investigated the emotions associated with science teaching in the interactive stage and recorded classroom observations to document how they implemented their plans. Finally, we conducted in-depth interviews with the teachers while they watched the recorded class videos, allowing them to reflect on their decision-making and implementation characteristics (post-active stage). Our findings reveal a dichotomy: negative emotions derived from past experiences reduce engagement, while positive emotions enhance motivation and enrich the learning environment. Negative emotions arising from the implementation of science activities in real classroom contexts may lead to canceled classes, whereas positive emotions foster persistence and optimism. We discuss the significance of emotional awareness in teaching practices, emphasizing the importance of understanding and managing negative emotions through teacher training and professional development programs. This approach can enhance teacher confidence and satisfaction, ultimately contributing to improved early childhood science education.

A Machine Learning Framework for Classroom EEG Recording Classification: Unveiling Learning-Style Patterns

Classroom EEG recordings classification has the capacity to significantly enhance comprehension and learning by revealing complex neural patterns linked to various cognitive processes. Electroencephalography (EEG) in academic settings allows researchers to study brain activity while students are in class, revealing learning preferences. The purpose of this study was to develop a machine learning framework to automatically classify different learning-style EEG patterns in real classroom environments. Method: In this study, a set of EEG features was investigated, including statistical features, fractal dimension, higher-order spectra, entropy, and a combination of all sets. Three different machine learning classifiers, random forest (RF), K-nearest neighbor (KNN), and multilayer perceptron (MLP), were used to evaluate the performance. The proposed framework was evaluated on the real classroom EEG dataset, involving EEG recordings featuring different teaching blocks: reading, discussion, lecture, and video. Results: The findings revealed that statistical features are the most sensitive feature metric in distinguishing learning patterns from EEG. The statistical features and RF classifier method tested in this study achieved an overall best average accuracy of 78.45% when estimated by fivefold cross-validation. Conclusions: Our results suggest that EEG time domain statistics have a substantial role and are more reliable for internal state classification. This study might be used to highlight the importance of using EEG signals in the education context, opening the path for educational automation research and development.

A Student Facial Expression Recognition Model Based on Multi-Scale and Deep Fine-Grained Feature Attention Enhancement.

In smart classroom environments, accurately recognizing students' facial expressions is crucial for teachers to efficiently assess students' learning states, timely adjust teaching strategies, and enhance teaching quality and effectiveness. In this paper, we propose a student facial expression recognition model based on multi-scale and deep fine-grained feature attention enhancement (SFER-MDFAE) to address the issues of inaccurate facial feature extraction and poor robustness of facial expression recognition in smart classroom scenarios. Firstly, we construct a novel multi-scale dual-pooling feature aggregation module to capture and fuse facial information at different scales, thereby obtaining a comprehensive representation of key facial features; secondly, we design a key region-oriented attention mechanism to focus more on the nuances of facial expressions, further enhancing the representation of multi-scale deep fine-grained feature; finally, the fusion of multi-scale and deep fine-grained attention-enhanced features is used to obtain richer and more accurate facial key information and realize accurate facial expression recognition. The experimental results demonstrate that the proposed SFER-MDFAE outperforms the existing state-of-the-art methods, achieving an accuracy of 76.18% on FER2013, 92.75% on FERPlus, 92.93% on RAF-DB, 67.86% on AffectNet, and 93.74% on the real smart classroom facial expression dataset (SCFED). These results validate the effectiveness of the proposed method.

Enhancing AR‐based learning environments for STEM education: A design‐based study on design features, kinematics learning and mathematics self‐efficacy

AbstractThis design‐based research project explored how various design features of AR‐based learning environments (ARLE) influence students' mathematics self‐efficacy and learning of kinematics. Specifically, five ARLEs with different design features were developed and implemented with 136 seventh‐grade students in two rounds. Data were gathered from pre‐ and post‐mathematics self‐efficacy tests, interviews, classroom observation records and reflective journals, to identify which design features promote or hinder students' learning and self‐efficacy. Results revealed that adopting a well‐designed ARLE is a viable way to promote students' learning and mathematics self‐efficacy, and the success of ARLE lies in (1) coherent tasks with proper time for self‐exploration, progressive levels, useful problem contexts and a certain degree of open‐endedness, and (2) sufficient technological affordances with strengthened authenticity, clarity and integrated media presentation, as well as engaging interactions with decent flexibility, embodied actions and feedback. These results validated the applicability of the proposed checklist for ARLE designers and practitioners in their design and implementation of ARLEs, within the domain of mathematics, physics and beyond. Practitioner notesWhat is already known about this topic AR‐based learning environments (ARLEs) have demonstrated substantial advantages in STEM education, but ARLEs can be a drawback if they are not adequately designed. ARLE may facilitate students' mathematics self‐efficacy, and relevant design features are underexplored. Design‐based research sustains interventions combining empirical research and theory‐driven design of innovative learning environments, for example, ARLEs. What this paper adds Demonstrates the potential of ARLEs that can be used in classrooms to support STEM learning. Identifies a collection of effective and ineffective design features of ARLEs that promote or hinder students' STEM learning and mathematics self‐efficacy. Tackles the challenges for the design of ARLEs for classrooms by taking AR technological affordances and the pedagogical perspective together. Implications for practice and/or policy Five ARLEs for STEM topics were developed and implemented in secondary schools. Evidence of positive impacts of ARLEs on students' learning and mathematics self‐efficacy. Our research produces useful guidelines for designers and practitioners in their development and implementation of ARLEs in real classrooms.

Predicting quality of English language teaching through augmented reality competencies and TPACK model components among Kuwaiti undergraduates

<b>Background: </b>Augmented reality is among the emerging technologies that hold greater potential in the context of foreign language learning. No research has been done to date to investigate pre-service teachers’ competencies in augmented reality and their association with quality of teaching English and technological and pedagogical content knowledge (TPACK) model components in the state of Kuwait.<br /> <b>Aim: </b>This study aimed to assess the utility of using augmented reality competencies and English as a foreign language (EFL) TPACK model components to predict the quality of English language teaching of pre-service undergraduates.<br /> <b>Method: </b>A total of 317 students enrolled in college of education at Kuwait university were recruited and responded to three online questionnaires measuring EFL TPACK, teachers’ augmented reality competencies, and quality of teaching English skills (QELT).<br /> <b>Results: </b>Results indicated a significant positive association among all variables at 0.01 level. Teacher’s augmented reality competencies (TARC), TPACK, technological knowledge (TK), and technological content knowledge (TCK) were significant predictors of QELT. One-way ANOVA revealed that there was no significant effect of gender on the TARC, TPACK, TK, TCK, and QELT. The cut-off-criteria of the mean scores indicated that all participants strongly believe that they acquire the essential competencies of augmented reality in EFL classrooms and possess a high level of proficiency in TPACK. Descriptive statistics showed that more than (70%) of pre-service teachers selected “strongly agree” and “agree”, 13% or less selected “strongly disagree” and “disagree” while 26% or less selected “neutral” response. Linear regression analysis revealed that TARC, TPACK, TK, and TCK were significant predictors of QELT.

Explaining preservice teachers' intention and behavior to use technology‐enabled learning in China: A multi‐group analysis across experiences

AbstractThis study examines the development of preservice teachers' intention and behavior to use of Technology‐Enabled Learning (TEL) within the Theory of Planned Behavior (TPB) framework. Despite growing interest, the effective integration of TEL among preservice teachers remains a challenge. Engaging 1023 participants ‐ senior undergraduates with practical TEL experience in real classroom settings ‐ the study employed structural equation modeling (SEM) to analyze the relationships between key factors ‐ experience, subjective norms, attitudes, and perceived behavioral control ‐ and their influence on the intention and behavior to use of TEL. The findings revealed that behavioral intention did not significantly influence the actual use of TEL, whereas perceived behavioral control acted both as a determinant of TEL use and as a mediator in the relationship between subjective norms and TEL usage. Additionally, while subjective norms showed minimal impact on behavioral intention, attitudes and perceived behavioral control emerged as crucial antecedent variables and mediators. The multi‐group analysis using SEM revealed that preservice teachers' experience with TEL moderated the first half of the mediating pathway between subjective norms, perceived behavioral control and the actual use of TEL. Notably, multi‐group analysis indicated that as preservice teachers gained more experience with TEL, the influence of subjective norms on perceived behavioral control weakened. This study contributes to a deeper understanding of the TPB framework and its extended models in the context of TEL in teacher education, offering valuable insights for developing more effective TEL training and guidance strategies, ultimately aiming to enhance preservice teachers' proficiency and engagement with TEL.

Challenges in Teaching English as a Lingua Franca: Japanese In-service Teachers’ Attitudes toward ELF Teaching and Classroom Realities

With the global spread of English, there are growing calls for English as a lingua franca (ELF) teaching (e.g., exposure to different varieties of English and opportunities to practice communication strategies) to help learners communicate in intercultural settings successfully. However, despite the widely acknowledged importance of ELF teaching, it is implemented only occasionally in classroom settings. While some previous research has explored challenges in ELF teaching, this study uniquely focuses on the Japanese context, which has not been fully explored yet. Therefore, this research explored Japanese teachers of English’s (JTEs’) attitudes toward ELF teaching and identified barriers preventing them from implementing ELF. The results from a questionnaire completed by 22 JTEs suggested that most of them demonstrated supportive attitudes, while there was a wide range of opinions toward ELF teaching. However, most participants rarely implemented ELF-related activities, primarily because of a lack of teaching time and a lack of familiarity with ELF. Based on an analysis of the results, the current study suggested that developing approachable ELF-related activities and introducing ELF content in teacher training programs were critical to effectively implementing ELF teaching in classrooms.

More than tough luck

Abstract Applied Cognitive Linguistics (ACL, henceforth) offers a vision of language closely tied to our experiences and views of the world, showing great promise for foreign language teaching and learning. Building upon this meaning-centered perspective, this paper discusses an interventional, quasi-experimental study assessing the effectiveness of incorporating animated images into materials for teaching comparative constructions in L2 Spanish. Conducted in three phases with varying participation rates (69 students in the first, 18 in the second, and 27 in the third), the study faced numerous factors that hindered data collection process and thereby influenced the results. Although the findings were not statistically significant, they suggest a slight improvement in post-test performance in both the experimental and comparison groups. While the potential of ACL in second language teaching and learning is undeniable, so are the challenges faced in L2 classrooms. This underscores the need for further empirical research to bridge the gap between theoretical principles and their practical implementation in real classroom settings.

Evaluation of an English as a Foreign Language (EFL) Teacher Preparation Program

The responsibility of a teacher preparation program (TPP) goes beyond teaching future educators the knowledge and skills of the profession. It needs to provide experiences that help them succeed in the real classroom settings. (Garza-Rodriguez, 2019). However, there is evidence that disconnection between what is taught in TPPs and what is needed in the practical work exists. Therefore, there is an urgent need to improve the quality of these programs by constantly evaluating their quality and effectiveness. This paper reports a pilot study on pre-service teachers’ perspectives about their English as a foreign language teacher (EFL) teacher preparation program in Mexico. The aim of the study was to reveal how pre-service teachers evaluated the strengths and weaknesses of their EFL teacher preparation program. Following a mixed-method, data were collected from 17 participants using an online Likert scale, adapted from Peacock (2009), and three open questions to elicit strengths and weaknesses. Findings were divided in both quantitative and qualitative. Quantitative data indicated that students were undecided or to some extent in agreement with most of the items associated with the features of the program; however, the lowest mean was shown in the areas of meeting students’ needs, the balance between theory and practice and learning some practical teaching skills. On the other hand, the qualitative results showed that students’ preparation as linguists and self-reflective learners as well as some subjects from the program were perceived as strengths, while the lack of teaching experience, the content, and low English language proficiency as the main weaknesses. In conclusion, though students’ perspectives were still uncertain, they provided useful information to improve the curriculum.

Helping teacher education students’ understanding of self-regulated learning and how to promote self-regulated learning in the classroom

This research investigated the details and effects of a short online Professional Learning Program designed to develop teacher education students’ knowledge about how to promote self-regulated learning (SRL) in the classroom. The Program was based on a new framework for how teachers can promote SRL, the SRL Teacher Promotion Framework (SRL-TPF), which focused on the promotion of SRL strategies, students’ knowledge about learning, and students’ metacognition. It consisted of seven modules describing the different SRL promotion types and SRL capabilities and ways to promote them through teacher talk and action. Modules included written information and video examples taken from observations of real classrooms, which were used to illustrate the transfer of SRL theory to instructional practice. Each module concluded with several assessment items. During the Program the participants, 91 teacher education students, were asked to use a simplified scoring system based on the SRL-TPF to code lesson transcripts taken from classroom observations. The results showed that by the end of the program over 85% of the participants were able to provide teacher instructions that included explicit SRL promotion and/or promoted students’ SRL knowledge. Our study contributes to research findings on teacher education students’ knowledge of SRL, their promotion of SRL to students, and the contribution of short duration SRL professional development.

Video-Based Mathematics Teacher Training: A Comparison of Real Classroom Videos and Virtual Reality

Video-Based Mathematics Teacher Training: A Comparison of Real Classroom Videos and Virtual Reality

Development of instructional design principles for using ICT in resource-limited learning environments: a case of Bangladesh

The integration of ICT in education in developing countries has become a significant focus in various ODA projects and educational research. With the increasing demand for teachers to incorporate ICT in their classrooms, the precise methods for effectively integrating ICT resources have not been thoroughly explored. Therefore, this research aimed to develop the most optimal instructional design principles that may systematically guide teachers in planning their lessons using ICT, specifically in the context of Bangladesh. By employing the design and development research method, this research developed an initial set of instructional design principles based on literature review and field research, then elaborated on the principles through three sets of internal validation tests and a usability test by conducting in-depth interviews to confirm the applicability of the principles in real classroom settings in Bangladesh. The final version of instructional design principles is composed of five components, ten principles, and 22 specific guidelines. Two distinct features of the principles are discussed for further research. In discussion, this research underscores the necessity of providing practical instructional design principles to guide teachers in effectively utilizing ICT, particularly within a resource-limited learning environment like Bangladesh.

The Neurophysiological Costs of Learning in a Noisy Classroom: An Ecological Virtual Reality Study.

Many real-life situations can be extremely noisy, which makes it difficult to understand what people say. Here, we introduce a novel audiovisual virtual reality experimental platform to study the behavioral and neurophysiological consequences of background noise on processing continuous speech in highly realistic environments. We focus on a context where the ability to understand speech is particularly important: the classroom. Participants (n = 32) experienced sitting in a virtual reality classroom and were told to pay attention to a virtual teacher giving a lecture. Trials were either quiet or contained background construction noise, emitted from outside the classroom window. Two realistic types of noise were used: continuous drilling and intermittent air hammers. Alongside behavioral outcomes, we measured several neurophysiological metrics, including neural activity (EEG), eye-gaze and skin conductance (galvanic skin response). Our results confirm the detrimental effect of background noise. Construction noise, and particularly intermittent noise, was associated with reduced behavioral performance, reduced neural tracking of the teacher's speech and an increase in skin conductance, although it did not have a significant effect on alpha-band oscillations or eye-gaze patterns. These results demonstrate the neurophysiological costs of learning in noisy environments and emphasize the role of temporal dynamics in speech-in-noise perception. The finding that intermittent noise was more disruptive than continuous noise supports a "habituation" rather than "glimpsing" hypothesis of speech-in-noise processing. These results also underscore the importance of increasing the ecologically relevance of neuroscientific research and considering acoustic, temporal, and semantic features of realistic stimuli as well as the cognitive demands of real-life environments.

SLBDetection-Net: Towards closed-set and open-set student learning behavior detection in smart classroom of K-12 education

Through effective analysis of K-12 education students’ learning behaviors in the classroom, it is possible to greatly improve the interaction between teaching and learning, thereby enhancing the quality of education. However, current traditional analysis of student classroom behavior focuses on closed-set behavior detection in a single scenario. For complex and open real classroom environments, the challenge lies in obtaining meaningful representations of behaviors in small and densely populated complex scenes, while also achieving good performance in both closed-set and open-set environments. To address these challenges, this study introduces a novel method for detecting student learning behavior in both closed-set and open-set scenarios, termed SLBDetection-Net. This method focuses on accurately capturing learning behavior representation, with a specific emphasis on Multi-scale Focusing Key Information (MFKI). The study begins with designing a Learning Behavior-aware Attention (LBA) mechanism, dedicated to extracting key features of learning behaviors and capturing the complex characteristics of targets across different scales. Building on this attention mechanism, a backbone network feature encoder, LBA-Swin Transformer Block, is constructed to form the comprehensive SLBDetection-Net. The effectiveness of SLBDetection-Net is validated through rigorous testing and evaluation on real classroom scenario data of K-12 education, comparing its performance with the State-of-the-Art (SOTA) methods. The results demonstrate that SLBDetection-Net achieves a mean Average Precision (mAP) of 96.4% on the ClaBehavior dataset and 85.9% on the SCB dataset. These findings underscore the method’s significant advantages in enhancing detection precision and efficiency in both closed-set and open-set scenarios, thereby expanding the application scope of educational assessment frameworks. The source code of this study is publicly available at https://github.com/CCNUZFW/SLBDetection-Net.

Going farther, together: completing literacy professional development as peers

PurposeIn the Fall of 2019, Marietta College and Marietta City Schools, in Marietta, Ohio, piloted a program to improve literacy knowledge and pedagogy through completing LETRS training, a two-year literacy professional development based on the science of reading, as peers.Design/methodology/approachThis project was aligned to effective professional development research that states better trainings are content-specific, allow support and collaboration and are ongoing throughout the school year (Blank and de las Alas, 2009; Darling-Hammond, Hyler, and Gardener, 2017) and respect participants work/life schedules (Bigsby and Firestone, 2017).FindingsSome benefits of this collaboration were that teachers, college faculty and teacher candidates were able to communicate more effectively about literacy because we all had shared background knowledge and spoke the same language when it came to literacy. We were also able to make more meaningful clinical experiences for our teacher candidates because we created a stronger connection between the knowledge and pedagogy taught in students’ literacy courses and the practices they saw in real classrooms. Inservice teachers saw the college faculty and teacher candidates as strong partners, which greatly strengthened our clinical preparation.Practical implicationsThe paper includes implications for the development of stronger partnerships between teacher preparation program faculty and school partnership faculty and more authentic and meaningful clinical experiences for teacher candidates.Originality/valueThis project shows how meaningful partnerships and clinical experiences can be created when partnership faculty are seen as peers.

Relationship between teacher's ability model and students' behavior based on emotion-behavior relevance theory and artificial intelligence technology under the background of curriculum ideological and political education

PurposeThis paper aims to fully understand the relationship between students' classroom emotions and teachers' behaviors in the teaching environment, provide personalized teaching support, stimulate teachers' motivation, and enhance the emotional connection between teachers and students. This paper designs a teacher's ability model based on emotion-behavior relevance theory and artificial intelligence (AI) technology. MethodsFirstly, this paper expounds on the integration of emotion-behavior relevance theory, AI, and teacher's ability model under the background of ideological and political education. Subsequently, the artificial neural network (ANN) model is deeply analyzed in relation to the development of psychology and AI technology. Guided by the theory of emotion-behavior relevance, an ANN model is used to optimize the classroom emotion recognition module in the teacher's ability model. The accuracy of the optimized teacher's ability model in recognizing attention and resistance is verified by experiments. The relationship between the teacher's classroom emotions, students' classroom behavior, and the teacher's teaching ability is analyzed. ResultsThe experimental results show that the optimized model can effectively identify students' classroom emotions, and the accuracy of identifying attention and resistance reaches 93.6 % and 92 %, respectively, significantly exceeding the traditional model. For other emotions, the accuracy of the experimental group ranges from 80.6 % to 87.9 %, while that of the control group is only 61.1 %. The optimized model shows a better effect on the emotional recognition of multiple students. This proves the effectiveness of the proposed optimization model. In addition, by analyzing teachers' emotions in real classroom videos, people can observe that teachers' psychological emotions and behaviors change with the changes in students' classroom emotions. Under the students' positive emotions, teachers scored high in psychological emotions, behaviors, and teaching achievements, with the lowest scores of 87, 80, and 85, respectively. Under students' negative emotions, teachers' related scores are low, with the highest scores of 40, 40, and 42. This highlights the critical influence of the emotional attitude and values of the main characters on teachers' teaching ability in the classroom environment from the psychological perspective. ConclusionStudents' positive emotions, such as concentration and devotion, will arouse teachers' satisfaction and happiness, motivate them to teach more diligently, and improve the teaching effect. On the contrary, students' negative emotions, such as resistance and doubt, may cause teachers to feel depressed and discouraged. This proves the importance and effectiveness of integrating the emotion-behavior relevance theory into teaching. Students' emotions impact teachers' psychological emotions and behaviors and affect the teaching effect. In this case, teachers need to use their emotional adjustment ability to adjust their feelings in time to better guide students and improve teaching efficiency. This paper provides valuable insights for optimizing the teacher's ability model.

Comparing contrasting instructional approaches: a way for research to develop insights about backward transfer

ABSTRACT Backward transfer, defined as the influence that learning about new concepts has on individuals' reasoning about previously learned concepts, is a relatively new topic in mathematics education. Our exploratory study compared the backward transfer produced in a mathematics enrichment programme that emphasised covariational reasoning to that produced in real algebra classrooms that did not emphasise a particular type of reasoning. Quadratic and linear functions were the new concepts being learned about and the concepts previously learned about, respectively. One theme that emerged from our comparisons was that quadratic instruction that emphasised covariational reasoning had less, but more consistent, backward transfer than quadratic instruction that did not have a strong emphasis on a particular type of reasoning. This and other findings led us to new insights about backward transfer, as well as tentative new ideas about backward transfer more generally, that can be tested in future studies..

Exploring Pre-Service Teacher Professional Learning on Their Preparedness for Teaching Practicum

The teaching practicum is essential in pre-service teacher education, allowing aspiring teachers to apply theoretical knowledge in real classrooms. It helps them to develop practical skills, build communication abilities, and connect with clients, colleagues, and mentors. This study investigates pre-service teachers' perceptions of their readiness for teaching practicums, the influencing factors, and the contribution of professional learning programs on their preparedness. Using a mixed-method approach with an explanatory sequential design, the research was conducted in Metro City, Lampung. Data were collected through a close-ended questionnaire from 80 pre-service teachers and semi-structured interviews with 9 pre-service teachers who were ready as the volunteers. Descriptive statistics and qualitative analysis were used to analyze the data. Results indicated that professional learning programs significantly enhance preparedness in pedagogical training, content knowledge, classroom management, understanding diverse student needs, and technology integration. These programs boost confidence, competence, and the ability to create effective learning environments. Informal learning, such as roles as teaching assistants and using social media, also contributes significantly to innovative teaching. Integrating formal and informal learning experiences is crucial for the holistic development of pre-service teachers

A Study of Classroom Behavior Recognition Incorporating Super-Resolution and Target Detection.

With the development of educational technology, machine learning and deep learning provide technical support for traditional classroom observation assessment. However, in real classroom scenarios, the technique faces challenges such as lack of clarity of raw images, complexity of datasets, multi-target detection errors, and complexity of character interactions. Based on the above problems, a student classroom behavior recognition network incorporating super-resolution and target detection is proposed. To cope with the problem of unclear original images in the classroom scenario, SRGAN (Super Resolution Generative Adversarial Network for Images) is used to improve the image resolution and thus the recognition accuracy. To address the dataset complexity and multi-targeting problems, feature extraction is optimized, and multi-scale feature recognition is enhanced by introducing AKConv and LASK attention mechanisms into the Backbone module of the YOLOv8s algorithm. To improve the character interaction complexity problem, the CBAM attention mechanism is integrated to enhance the recognition of important feature channels and spatial regions. Experiments show that it can detect six behaviors of students-raising their hands, reading, writing, playing on their cell phones, looking down, and leaning on the table-in high-definition images. And the accuracy and robustness of this network is verified. Compared with small-object detection algorithms such as Faster R-CNN, YOLOv5, and YOLOv8s, this network demonstrates good detection performance on low-resolution small objects, complex datasets with numerous targets, occlusion, and overlapping students.