Exploring the mediating role of attitude toward use in GenAI adoption for pre-service music teacher: insights from the UTAUT2 framework

Abstract

Introduction:

The rapid advancement of Generative AI (GenAI) has transformed educational practices, particularly in music education, offering new tools for enhancing teaching and learning experiences. However, research on its adoption by pre-service music teachers remains limited, particularly in understanding how cognitive and emotional factors influence their acceptance of these tools in real-world teaching contexts.

Methods:

Based on the UTAUT2 framework, this study examines the factors influencing the adoption of GenAI by 848 pre-service music teachers from eight universities in China, focusing on the mediating role of their attitude in the technology acceptance process to enhance its integration into music education. The data were analyzed using Covariance-Based Structural Equation Modeling to test the hypothesized relationships between constructs and evaluate the mediating role of attitude toward use in pre-service music teachers’ adoption of GenAI.

Results:

The hypothesis testing showed that most direct paths were significant, except for the direct effect of price value on behavioral intention. Mediation analysis found that ATU significantly mediated the relationships between predictors and BI. Adding ATU increased the model's explanatory power, with the R² for behavioral intention rising from 0.520 to 0.582.

Discussion:

The discussion highlights that attitude toward use plays a crucial role in mediating the adoption of GenAI by pre-service music teachers, emphasizing the importance of both cognitive and emotional factors in technology acceptance. The findings suggest practical implications for educational administrators and GenAI developers to design user-friendly tools, provide supportive resources, and encourage positive attitudes through peer collaboration and hands-on experiences.

1 Introduction

GenAI has rapidly emerged as a significant technological development in recent years, exerting a growing influence on teaching and learning in higher education (Luo, 2024; O'Dea, 2024). GenAI is increasingly transforming educational practices by reshaping learning and teaching approaches, enabling personalized learning experiences, and providing automated feedback to support students' progress (Jha and Atif, 2025; Laak and Aru, 2025; Lu et al., 2026). In music education, GenAI demonstrates considerable potential to assist music composition, analyze musical patterns, support improvisation, and provide personalized feedback, thereby enhancing both teaching efficiency and students’ learning experiences (Carnovalini et al., 2025; Peng et al., 2026). GenAI offers pre-service teachers powerful tools for creating dynamic learning materials, personalizing educational content, and simulating complex classroom scenarios to enhance pedagogical practice (Arantes, 2025).

Pre-service music teachers are widely regarded as future implementers of music education and therefore play a critical role in shaping how emerging technologies are integrated into classroom practice (Tondeur et al., 2012). During their professional preparation, challenges are frequently reported in areas such as lesson planning, resource integration, and addressing diverse student learning needs. Recent research suggests that GenAI provides new opportunities to support these tasks through intelligent content generation, instructional assistance, and creative tools for music teaching (Fang, 2026; Li et al., 2025). However, the effective integration of GenAI in educational contexts is largely dependent on teachers' willingness to adopt technology (Lu et al., 2024). In practice, pre-service music teachers often face challenges related to insufficient technological preparedness, cognitive burden in mastering GenAI tools, and concerns over perceived risks, which collectively influence their usage behavior in educational contexts (He and Ren, 2025). Consequently, a deeper understanding of the factors influencing pre-service music teachers' acceptance of GenAI is essential for promoting its effective integration into music education.

To understand the factors influencing individuals' adoption of emerging technologies, technology acceptance models have been widely employed in educational research (Buchanan et al., 2013; Granić, 2022). Among these models, the Unified Theory of Acceptance and Use of Technology (UTAUT) and its extended version, UTAUT2, have gained substantial attention for explaining users' behavioral intentions (BI) toward new technologies (Venkatesh et al., 2012). The UTAUT2 model includes several key determinants of technology adoption, including performance expectancy (PE), effort expectancy (EE), social influence (SI), facilitating conditions (FC), hedonic motivation (HM), price value (PV), and habit (HA). These constructs collectively explain individuals' behavioral intentions and technology usage behavior across various contexts.

Attitude Toward Use (ATU) has long been recognized as an important psychological mechanism in technology adoption research. In early technology acceptance studies, ATU was regarded as a key mediator linking cognitive beliefs and BI (Davis, 1989). Although later models such as UTAUT and UTAUT2 removed ATU to enhance model parsimony, subsequent studies have argued that ATU still plays a meaningful mediating role, particularly in emerging technology contexts where users' evaluations involve both cognitive judgments and affective responses (Chanda et al., 2024; Kong et al., 2024; Unal and Uzun, 2021). This issue could be particularly salient in the context of pre-service music teachers. Unlike many other disciplines, music education places strong emphasis on artistic authenticity, creativity, and personal expression (Humphreys, 2006). The emergence of GenAI has not only prompted instrumental evaluations regarding its efficiency and practical usefulness, but has also stimulated deeper cognitive and affective reflections within academia on whether GenAI-generated music in educational practice can genuinely embody human creativity, emotional expression, and artistic authenticity (Herington et al., 2026). Previous studies have noted that musicians and music educators often hold ambivalent attitudes toward GenAI-generated music, simultaneously recognizing its pedagogical potential while expressing concerns about its impact on artistic originality and musicianship (Herremans et al., 2017; Holster, 2024). Moreover, as pre-service teachers are still developing their professional identities and pedagogical beliefs, their evaluations of emerging technologies are likely to involve both cognitive assessments and affective reactions (Valtonen et al., 2021). In this sense, ATU function as a psychological bridge through which cognitive perceptions of GenAI technologies are translated into BI to integrate these tools into future music teaching practice.

It should be noted that He and Ren (2025) have already extended the UTAUT2 model by incorporating perceived compatibility and perceived risk and employed a Partial Least Squares Structural Equation Modeling (PLS-SEM) and artificial neural network (ANN) approach to examine pre-service music teachers' intention to adopt GenAI. In their study, He and Ren (2025) reported substantial explanatory power of the model, with the structural model explaining 68.2% of the variance in BI, indicating that the extended UTAUT2 model effectively explains pre-service music teachers’ adoption of GenAI. However, PLS-SEM and ANN are typically applied for prediction-oriented analysis, whereas Covariance-Based Structural Equation Modeling (CB-SEM) is commonly used for theory testing and the assessment of overall model fit within a specified theoretical framework. Therefore, employing CB-SEM in the present study enables a more rigorous examination of theoretical relationships and provides additional empirical evidence for validating the proposed research model.

Despite the growing body of research on GenAI adoption in education, several gaps remain. Existing studies have primarily focused on predictive modeling approaches and have paid limited attention to the mediating role of ATU in the technology adoption process, particularly in the context of music teacher education. Moreover, empirical evidence regarding pre-service music teachers' acceptance of GenAI remains relatively limited. To address these gaps, the present study extends the UTAUT2 framework by incorporating ATU as a mediating variable and examines the factors influencing pre-service music teachers' behavioral intentions to adopt GenAI in teaching. By employing CB-SEM, this study aims to provide a more rigorous validation of the proposed theoretical model and contribute additional empirical insights into the mechanisms underlying GenAI adoption in music education.

2 Literature review

2.1 The unified theory of acceptance and use of technology (UTAUT)

UTAUT, proposed by Venkatesh et al. (2003), represents a comprehensive framework that integrates eight prominent models of technology acceptance, including the Technology Acceptance Model (Davis, 1989), Theory of Planned Behavior (Ajzen, 1991), Innovation Diffusion Theory (Rogers et al., 2014), and Social Cognitive Theory (Bandura, 1986), among others, to explain individuals' behavioral intentions and technology usage behavior.

The original UTAUT model comprises four core constructs: PE, EE, SI, and FC, which together explain individuals' BI and technology use. In addition, the model incorporates four moderating variables, namely gender, age, experience, and voluntariness of use, to account for heterogeneity in technology adoption behavior. Empirical validation by Venkatesh et al. (2003) showed that UTAUT explains up to 69% of the variance in BI, substantially outperforming earlier models whose explanatory power ranged from 17% to 53%, thereby establishing UTAUT as one of the most influential frameworks in technology acceptance research. To enhance the model's applicability and explanatory power in consumer contexts, Venkatesh et al. (2012)extended the original framework to develop UTAUT2 by introducing three additional constructs: HM, PV, and HA. Empirical results based on a survey of 1,512 mobile Internet users showed that the extended model substantially improved its explanatory power, increasing the explained variance in behavioral intention from 56% to 74% and in technology use from 40% to 52% (Venkatesh et al., 2012).

Since its introduction, UTAUT2 has been widely applied to investigate the adoption of emerging technologies across various educational contexts, including mobile learning (Sitar-Taut and Mican, 2021), MOOCs (Tseng et al., 2022), VR Technology (Du and Liang, 2024), and GenAI-based educational tools (Wang and Liu, 2026). However, empirical studies applying the UTAUT2 framework to examine GenAI adoption in music education remain limited, particularly among pre-service teachers.

2.2 Determinants of behavioral intention in the UTAUT2

2.2.1 Performance expectancy (PE)

2.2.2 Effort expectancy (EE)

2.2.3 Social influence (SI)

2.2.4 Facilitating conditions (FC)

2.2.5 Hedonic motivation (HM)

2.2.6 Price value (PV)

2.2.7 Habit (HA)

2.3 Attitude toward use as a mediating variable

Attitude Toward Use (ATU) refers to an individual's overall evaluative disposition toward using a specific technology, capturing the degree to which a person holds a positive or negative affective judgment about engaging in that use behavior (Davis, 1989). This concept originates from the Theory of Reasoned Action (TRA), which posits that attitude is a critical psychological factor influencing BI, with attitudes serving as a key determinant of an individual's intention to perform a behavior (Fishbein and Ajzen, 1975). According to the TRA, both attitude and subjective norms jointly influence the formation of BI, thereby predicting the likelihood of a behavior occurring. TAM further extends this logic, positing that users' perceptions of ease of use and usefulness influence their attitude toward the technology, which in turn affects their BI (Davis, 1989). In numerous studies based on the TAM, ATU has consistently served as a key mediator in the relationship between external factors and BI to adopt new technologies, such as GenAI (Na et al., 2022), e-learning (Almulla, 2021), and mobile learning (Almaiah et al., 2022). These studies have demonstrated that ATU plays a crucial role in translating users' cognitive evaluations of technology into a concrete intention to use it. However, in many TAM studies, ATU has been deliberately removed because perceived usefulness and perceived ease of use were found to fully mediate the effects of external variables on behavioral intention, making a separate attitude variable unnecessary for explaining usage intention in those models (Tao, 2008).

In both the original UTAUT and its later extension UTAUT2, attitude is not included as a core construct (Venkatesh et al., 2003, 2012). Meta-analytic evidence indicates that although ATU has been added to extended versions of UTAUT in some studies, including ATU as a separate construct in the original UTAUT framework does not significantly enhance its explanatory power, which explains why the majority of UTAUT applications omit ATU (Or, 2023). However, some empirical research suggests that ATU still play a central role in predicting both BI and usage behavior in certain contexts. Recent meta-analytic research also indicates that ATU partially mediates the effects of PE, EE, FC, and SI on BI, and also has a direct influence on usage behavior (Dwivedi et al., 2019). This implies that in certain scenarios, especially in voluntary adoption contexts, including ATU in the UTAUT model could enhance its explanatory power and provide a more complete understanding of individual technology adoption behaviors (Dwivedi et al., 2019).

In the specific context of music teachers’ adoption of GenAI, the voluntary nature of technology use and the affective dimension of teaching innovation make ATU particularly relevant. In music education, teachers typically have the freedom to choose whether and how to adopt new instructional technologies, and their emotional evaluations and personal attitudes toward GenAI tools are likely to shape their intention to use them, beyond purely cognitive assessments. Most music teachers focus not only on whether a technology can enhance efficiency, but more importantly, on whether it can preserve or promote artistic integrity and creative expression (Lam, 2023). Therefore, the adoption of GenAI is not driven solely by rational assessments and is often influenced by teachers' emotional responses, especially concerning its impact on artistic creation. ATU plays a crucial role in this process, as it not only reflects their emotional disposition toward the technology but also influences whether they will integrate this new technology into their teaching practices.

Figure 1

3 Materials and methods

3.1 Research design

The participants in this study were recruited from eight universities located in Hubei and Jiangxi provinces in central China. These provinces were selected due to their educational diversity, with universities offering a range of academic strengths and student populations. The eight universities selected for this study include large, comprehensive institutions, specialized music academies, and smaller, regionally-focused universities. This selection ensures a broad representation of institutional types, allowing the study to capture various educational contexts and student demographics, which enhances the generalizability of the findings across China's higher education system.

In terms of sampling methods, a combination of judgment sampling and convenience sampling was employed. Judgment sampling was used to select participants who met the following criteria: (1) they were senior undergraduate students or final-year graduate students who were committed to becoming music teachers in the future, (2) they had prior music teaching practicum experience, and (3) they had used GenAI tool in designing and conducting music teaching activities. Convenience sampling was then used to select participants based on their availability and willingness to complete the survey, with the data collection process taking place at the selected universities.

The survey was conducted from July 10, 2025, to Feb 09, 2026. A total of 1,223 questionnaires were collected. After excluding 238 questionnaires with missing values and 137 questionnaires with overly consistent responses (where more than 90% of the scale items were answered with the same number, indicating a lack of careful consideration or random answering), 848 valid questionnaires were retained for further analysis, resulting in a response rate of approximately 69.33%.

3.2 Demographic and GenAI usage information

Demographic Information of the participants is shown in Table 1. In terms of gender, the majority were female (76.77%, n = 651), while 23.23% (n = 197) were male. Regarding age, most participants were aged between 20 and 24 years, comprising 69.81% (n = 592) of the sample. A smaller proportion (30.19%, n = 256) were over 25 years old. As for academic level, 73.35% (n = 622) of the participants were undergraduate students, and 26.65% (n = 226) were graduate students.

Table 2 presents the results regarding the usage of GenAI tools and the familiarity levels of pre-service music teachers in music education contexts. In the GenAI Usage in Music Learning and Teaching section, the most commonly used activity was searching and managing music material, with 79.95% (n = 678) of participants indicating they used GenAI for this purpose. The next most frequent activities were planning and designing music lessons (67.33%, n = 571) and studying music pedagogy and theory (66.04%, n = 560). Teaching and practicing music skills was less commonly reported, with 36.91% (n = 313) of participants using GenAI in this context. Regarding the frequently used GenAI tools, DeepSeek was the most popular tool, used by 88.68% (n = 752) of participants, followed by DouBao (65.45%, n = 555). ChatGPT (24.06%, n = 204) and Ernie Bot (17.10%, n = 145) were less frequently used, along with other tools (8.25%, n = 70). Finally, in terms of Familiarity with GenAI, the majority of participants identified as beginner or intermediate users. In music learning, 37.74% (n = 320) of participants were beginner users, while 23.70% (n = 201) were Intermediate. In Music Teaching, 37.97% (n = 322) were beginner users, with 22.41% (n = 190) being intermediate.

3.3 Research instruments

This study employed a structured questionnaire to examine the factors influencing pre-service music teachers' acceptance of GenAI. The questionnaire used a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree) to assess various constructs related to GenAI usage.

The second section of the questionnaire consists of an electronic informed consent form, where the researcher provides a detailed explanation of the study's purpose and the data to be collected. The form outlines that the study aims to explore the application of GenAI in music education and examine the factors influencing pre-service music teachers' acceptance of GenAI. Participants will provide demographic information, details about their teaching experiences, their frequency and context of GenAI usage, and their attitudes towards using GenAI in music teaching. The study guarantees that no sensitive information will be collected, ensuring confidentiality and the use of data exclusively for academic purposes. Participation is voluntary, with participants free to withdraw at any time without any negative consequences. By signing the informed consent form, participants acknowledge that they have read, understood, and agreed to participate in the study under the outlined terms.

The third section of the questionnaire consists of three demographic questions, designed to gather participants’ gender, age, and academic level. These questions aim to provide a profile of the sample population and help analyze any potential demographic influences on GenAI usage and acceptance in music education.

The fourth section contains four questions assessing participants' usage of and familiarity with GenAI in music education. These questions explore the frequency of GenAI usage in music teaching, participants' familiarity with different GenAI tools, and their experiences applying GenAI in music learning and instruction contexts.

The fifth section of the questionnaire consists of scales that were adapted from established English-language instruments. To ensure clarity and cultural relevance, the researchers hired a professional translation agency to translate the English questionnaire into Chinese, and subsequently, another professional translation agency was tasked with translating the Chinese version back into English. To ensure the accuracy and consistency of the translations, two Chinese professors specializing in English conducted a thorough comparison of the three versions of the scales. Based on their feedback, minor adjustments were made to ensure the questions were linguistically and culturally appropriate for the Chinese context while preserving the original meaning and intent of the scales.

The questionnaire items were adapted from established scales in the technology acceptance literature, including PE (3 items), PV (3 items), HA (3 items), FC (4 items), HM (3 items), and BI (3 items), all adapted from Venkatesh et al. (2012); EE (3 items), adapted from Mikalef et al. (2016); SI (4 items), adapted from Venkatesh et al. (2003); ATU (4 items), adapted from Taylor and Todd (1995). Before the formal administration of the questionnaire, a pilot test was conducted with a sample size of 50 participants. The results showed that all constructs had Cronbach's α coefficients above the recommended threshold of 0.70. The reliability coefficients for each construct were as follows: PE (α = 0.778), EE (α = 0.815), SI (α = 0.898), FC (α = 0.893), HM (α = 0.872), PV (α = 0.868), HA (α = 0.836), ATU (α = 0.905), and BI (α = 0.909). This demonstrates that the adapted questionnaire is reliable and suitable for use in the main study.

To detect potential common method bias, Harman's single-factor test was conducted by entering all scale items into an unrotated exploratory factor analysis using principal axis factoring; the result showed that a single factor accounted for 26.834% of the total variance, which is well below the commonly recommended threshold (< 50%), indicating that CMB is unlikely to be a serious concern in this study (Podsakoff et al., 2003).

3.4 Data analysis

For data analysis, descriptive statistics were first conducted to summarize the demographic information and GenAI usage patterns. Next, Confirmatory Factor Analysis (CFA) using CB-SEM was performed to assess the measurement model and test the hypothesized relationships between constructs. CB-SEM was chosen because it allows for a comprehensive evaluation of both measurement and structural models, making it suitable for testing complex relationships and ensuring validity. Model fit was evaluated using various fit indices, including CMIN/DF, CFI, TLI, and RMSEA. Convergent validity was assessed through standardized factor loadings, composite reliability, and average variance extracted. Discriminant validity was tested using the Fornell–Larcker criterion, ensuring that each construct shared more variance with its indicators than with other constructs in the model. Mediation effects were examined using the bootstrap method with 5,000 resamples and bias-corrected confidence intervals. Finally, the explanatory power of the structural model was assessed using R² values for the endogenous constructs.

4 Results

4.1 Descriptive analysis

Descriptive statistics for all scale items, including mean and standard deviation, are presented in Table 3. In addition to central tendency and dispersion, skewness and kurtosis were examined to assess the distributional normality of the observed variables. Although no formal normality test was conducted, the observed skewness values ranged from −0.480 to 0.038 and kurtosis values ranged from −0.957 to −0.214, all falling within the commonly accepted thresholds for SEM, with absolute skewness ≤ 1 and kurtosis ≤ 3 (Kim, 2013). These results indicate that none of the variables exhibit extreme asymmetry or peakedness, supporting approximate univariate normality suitable for maximum likelihood estimation in CB-SEM.

4.2 Measurement model testing

The measurement model was evaluated for two model specifications, one including the attitude construct and the other excluding it. As shown in Table 4, the results indicate that both measurement models demonstrate a satisfactory level of overall model fit according to commonly recommended SEM fit criteria (Carmines and McIver, 1981; Hu and Bentler, 1999). Overall, the fit indices consistently suggest that the hypothesized measurement structures adequately represent the observed data, providing empirical support for the validity of the measurement model under both specifications.

Convergent validity of the two measurement models was assessed using standardized factor loadings, composite reliability (CR), and average variance extracted (AVE), as summarized in Table 5. For the model including attitude, all standardized factor loadings exceeded the recommended threshold of 0.70. The CR values ranged from 0.804 to 0.900 and the AVE values ranged from 0.578 to 0.751. For the model excluding attitude, all factor loadings also exceeded 0.70, with CR values ranging from 0.804 to 0.900 and AVE values ranging from 0.579 to 0.750. These values exceed the commonly recommended thresholds of 0.70 for CR and 0.50 for AVE, indicating satisfactory internal consistency and adequate variance explained by the constructs (Fornell and Larcker, 1981; Hair et al., 2022). Overall, the results indicate that convergent validity is satisfactorily established for all constructs in both measurement models. In addition, Cronbach's alpha values ranged from 0.804 to 0.900, exceeding the recommended threshold of 0.70 and further confirming good internal consistency.

Discriminant validity was assessed using the Fornell–Larcker criterion. As shown in Table 6, all diagonal values exceed the correlations between the corresponding construct and all other constructs in the model, indicating that each latent variable shares more variance with its own indicators than with other constructs. These results provide evidence of satisfactory discriminant validity for all constructs in the measurement model.

4.3 Structural model testing

The structural model was assessed for two alternative specifications, namely a model incorporating ATU and a model excluding it. As presented in Table 7, the fit indices for both models satisfy the commonly accepted criteria for SEM model evaluation (Carmines and McIver, 1981; Hu and Bentler, 1999). These results indicate that the structural frameworks demonstrate adequate goodness of fit with the observed data. In general, the findings support the suitability of the proposed structural models in explaining the relationships among the latent variables under both specifications.

The hypothesis testing results for the structural model including ATU are presented in Table 8. PE significantly influenced ATU (β = 0.463, p < 0.001) and BI (β = 0.229, p < 0.001). EE had a significant effect on ATU (β = 0.119, p < 0.001) and BI (β = 0.049, p = 0.022). SI significantly affected both ATU (β = 0.302, p < 0.001) and BI (β = 0.284, p < 0.001). FC also showed significant effects on ATU (β = 0.066, p = 0.005) and BI (β = 0.091, p < 0.001). HM significantly influenced ATU (β = 0.246, p < 0.001) and BI (β = 0.233, p < 0.001). PV significantly affected ATU (β = 0.052, p = 0.012) but did not significantly influence BI (β = 0.003, p = 0.878). HA positively influenced ATU (β = 0.113, p < 0.001) and BI (β = 0.165, p < 0.001). Finally, ATU had a significant positive effect on BI (β = 0.366, p < 0.001).

The hypothesis testing results for the structural model excluding ATU are presented in Table 9. The results indicate that PE significantly influenced BI (β = 0.397, p < 0.001). EE also had a significant positive effect on BI (β = 0.093, p < 0.001). SI significantly affected BI (β = 0.397, p < 0.001), and FC showed a significant positive effect on BI (β = 0.116, p < 0.001). In addition, HM significantly influenced BI (β = 0.324, p < 0.001), while HA also had a significant positive effect on BI (β = 0.207, p < 0.001). However, PV did not significantly influence BI (β = 0.022, p = 0.290).

Mediation effects were examined using the bootstrap method with 5,000 resamples and bias-corrected confidence intervals at the 95% confidence level. As shown in Table 10, ATU was found to partially mediate the effects of several predictors on BI. Specifically, SI (β = 0.110, p < 0.001), HA (β = 0.041, p < 0.001), PV (β = 0.019, p = 0.010), HM (β = 0.090, p < 0.001), EE (β = 0.043, p < 0.001), and PE (β = 0.169, p < 0.001) had significant indirect effects. In contrast, the indirect effect of FC (β = 0.024, p = 0.006) was also significant. Indirect effects were considered significant if the 95% bias-corrected confidence interval did not include zero.

The explanatory power of the structural model was assessed using R² values for the endogenous constructs. In the model without ATU, BI was explained to a moderate degree, with R² = 0.520. After adding ATU, the explanatory power of the model improved slightly, with R² = 0.582. ATU was explained to moderate degree, with R² = 0.477. According to commonly used thresholds in SEM (Hair et al., 2022), R² values of 0.25, 0.50, and 0.75 correspond to weak, moderate, and substantial explanatory power, respectively. These results indicate that adding ATU to the model improves the proportion of variance explained, particularly for BI. Figure 2 presents the CB-SEM results with and without the ATU variable.

Figure 2

5 Discussions and implications

This study aimed to examine the factors influencing pre-service music teachers' acceptance of GenAI in music education by extending the UTAUT2 framework and incorporating ATU as a mediating variable. Using CB-SEM analysis of data collected from 848 pre-service music teachers in China, the findings revealed that several key determinants significantly influenced BI to use GenAI, while ATU played an important mediating role in the technology acceptance process. These results provide further insight into the mechanisms underlying technology acceptance in the context of emerging GenAI tools in music education. Overall, the findings suggest that the core explanatory logic of the UTAUT2 framework remains applicable to understanding generative GenAI adoption in music teacher education. The findings also highlight the theoretical importance of ATU, which significantly predicts BI and partially mediates the effects of several antecedent variables in the technology acceptance process. The increase in the explanatory power of the model after incorporating ATU, with the R² of BI rising from 0.520 to 0.582, further suggests that including ATU can enhance the explanatory capacity of technology acceptance models in the context of GenAI adoption in music teacher education. The following section discusses each variable based on the findings of this study and outlines the corresponding practical implications.

ATU emerged as the strongest predictor of BI in this study, indicating that pre-service music teachers' intention to adopt GenAI is strongly shaped by their overall evaluative disposition toward the technology. The mediation analysis further revealed that ATU significantly mediated the relationships between all antecedent variables and BI, indicating that cognitive and contextual perceptions of GenAI influence BI partly through users' attitudinal evaluations. This finding is consistent with prior technology acceptance research emphasizing the important role of attitude in shaping users' behavioral intention toward emerging technologies (Davis, 1989; Dwivedi et al., 2019). Music education places strong emphasis on creativity, artistic expression, and personal interpretation, which makes teachers particularly attentive to whether new technologies align with their pedagogical beliefs and artistic values before integrating them into teaching (Zhang and Wang, 2024). In addition, unlike conventional educational technologies that mainly support information delivery or task efficiency, GenAI directly participates in content creation and creative processes (Medel-Vera et al., 2025). As a result, teachers tend to carefully consider its role in creativity and artistic learning when deciding whether to adopt it in music teaching. These evaluative considerations are ultimately reflected in teachers’ attitudes toward GenAI, highlighting the central role of ATU in the technology acceptance process. Educational administrators should organize seminars and discussions on the role of GenAI in music creativity and pedagogy to help pre-service music teachers develop more open and reflective attitudes toward technology. GenAI developers should design tools that allow users to visualize and adjust AI-generated musical outputs, enabling teachers to maintain creative control and develop more positive attitudes toward using GenAI in music teaching. Pre-service music teachers should engage in critical reflection and guided practice when using GenAI tools in music-related tasks in order to develop balanced and informed attitudes toward technology.

PE was found to significantly influence both ATU and BI, and its effect on BI was also partially mediated by ATU. This finding is consistent with prior technology acceptance research (Altalhi, 2021; Dwivedi et al., 2019; He and Ren, 2025). One possible explanation lies in the perceived instructional benefits that GenAI can bring to music teaching. GenAI tools can assist pre-service music teachers in tasks such as lesson planning, generating musical materials, and providing creative ideas for classroom activities (Li et al., 2025). When teachers perceive that these tools can enhance teaching effectiveness, they tend to develop more positive attitudes toward the technology, which subsequently strengthens their intention to adopt it in music teaching (Dwivedi et al., 2019; He and Ren, 2025). This mechanism also explains why ATU mediates the relationship between PE and BI, as perceived performance benefits are first reflected in users' evaluative attitudes toward technology. Educational administrators should organize hands-on workshops where pre-service music teachers use GenAI to create lesson plans, generate accompaniment tracks, or design classroom activities so they can directly experience its value for music teaching. GenAI developers should design music-specific tools such as GenAI-assisted composition and accompaniment generation that clearly support classroom teaching tasks. Pre-service music teachers should regularly apply GenAI in micro-teaching or practicum activities to design exercises, generate practice materials, and explore creative teaching ideas.

EE significantly influenced both ATU and BI, with ATU partially mediating its effect on BI, indicating that perceiving GenAI as easy to use can promote more positive attitudes and stronger intentions to adopt the technology, consistent with prior research (Dwivedi et al., 2019; He and Ren, 2025; Shiferaw et al., 2021). Many pre-service teachers have limited technical training in GenAI-related tools, so technologies that are perceived as easier to operate are more likely to be accepted and integrated into their instructional practices (Scherer et al., 2019). Also, when GenAI tools are simple to learn and operate, teachers can focus more on their pedagogical and creative applications rather than technical difficulties, which leads them to form more favorable evaluations of the technology (Schoonenboom, 2014). Educational administrators should ensure pre-service music teachers have easy access to user-friendly GenAI tools, allowing them to focus on teaching tasks and fostering positive experiences with technology. GenAI developers should create intuitive, easy-to-use tools for music educators, enabling seamless integration into teaching and encouraging teachers to engage confidently with the technology. Pre-service music teachers should begin by using GenAI tools for simple, low-stakes tasks such as creating music exercises, and gradually progress to more advanced applications, enabling them to develop a deeper understanding and stronger confidence in using the technology.

SI was found to significantly impact both ATU and BI, with ATU partially mediating this effect, indicating that encouragement and support from peers and mentors help shape teachers' attitudes and adoption intentions toward GenAI, which aligns with prior research (He and Ren, 2025; Loveldy et al., 2021; Shiferaw et al., 2021). SI can shape teachers' attitudes toward GenAI by providing validation and encouragement from peers and mentors (Liu and Hu, 2025). Positive feedback from respected individuals can enhance teachers' confidence in using the technology, making them more open to adopting it in their teaching (Dahri et al., 2023). This social reinforcement fosters a more favorable disposition toward GenAI, ultimately influencing their intention to integrate it into their practices. Educational administrators should create peer-sharing opportunities such as teaching showcases where experienced users demonstrate how they apply GenAI in music lessons, allowing pre-service music teachers to observe successful practices and gain encouragement from respected peers. GenAI developers should build community features such as example libraries or shared prompt repositories where teachers can view and reuse GenAI-supported music teaching cases created by other educators. Pre-service music teachers should participate in peer collaboration activities such as jointly designing GenAI-assisted music lessons or sharing classroom experiences with GenAI during practicum discussions.

FC were found to significantly affect both ATU and BI, with ATU partially mediating this relationship, suggesting that sufficient resources and technical support can encourage teachers to adopt GenAI, consistent with previous research (He and Ren, 2025; Shiferaw et al., 2021; Teo, 2009). One possible explanation is that adequate resources and technical support reduce the practical barriers associated with using new technologies in teaching (Rogers, 2000). When teachers have access to stable platforms, reliable tools, and institutional support, they are more likely to perceive GenAI as manageable in classroom practice (Liu et al., 2025). These supportive conditions help create more favorable evaluations of the technology, which encourages its integration into music teaching. Educational administrators should provide reliable access to GenAI platforms and integrate them into music teaching labs or classrooms so that pre-service music teachers can easily use the tools during lesson preparation and practice teaching. GenAI developers should ensure that their platforms run stably on commonly used devices and provide ready-to-use templates for music lesson design, accompaniment generation, and classroom activities. Pre-service music teachers should take advantage of the available GenAI platforms and institutional resources during practicum and course assignments to explore how the tools can support music teaching tasks.

HM significantly influenced both ATU and BI, with ATU partially mediating this relationship, indicating that enjoyment derived from using GenAI can enhance teachers' evaluations and intentions to adopt the technology, consistent with previous research (Anand et al., 2019; He and Ren, 2025; Redda, 2020). GenAI tools often include interactive and creative features, such as generating musical compositions or personalized lesson plans, which make the teaching process more enjoyable and engaging (Peng et al., 2026). As teachers experience these enjoyable aspects, they are likely to form positive evaluations of the technology, seeing it as both useful and enjoyable for their teaching. These positive perceptions, in turn, increase their willingness to incorporate GenAI into their music teaching practices, as they associate the technology with a rewarding and creative teaching experience. Educational administrators should provide pre-service music teachers with easy access to engaging GenAI tools for creative tasks, helping them enjoy using the technology and develop positive attitudes toward it. GenAI developers should design tools with interactive, creative features that make using GenAI enjoyable, encouraging teachers to develop positive views of the technology. Pre-service music teachers should explore GenAI tools for creative tasks, such as generating music exercises, to enjoy the process and develop a positive connection with the technology.

PV significantly influenced ATU but did not directly affect BI, while its indirect effect on BI through ATU was significant, suggesting that cost-benefit evaluations shape adoption intention mainly by influencing teachers' evaluations of GenAI, which is partly consistent with previous research (Degirmenci and Breitner, 2017; He and Ren, 2025). Financial considerations are often less salient for pre-service music teachers when evaluating new educational technologies. Many GenAI tools are available through free versions, educational licenses, or institutional platforms, which reduces immediate cost concerns associated with their use (Ng et al., 2025). As a result, cost-benefit evaluations mainly influence how teachers perceive the overall value and practicality of the technology rather than directly determining their intention to use it. When GenAI is viewed as providing meaningful benefits for tasks such as lesson preparation, music material generation, or creative exploration relative to its cost, teachers are more likely to evaluate the technology favorably and consider integrating it into their music teaching practices. Educational administrators should provide institutional access to licensed GenAI tools for lesson preparation and music material generation, allowing pre-service music teachers to experience clear instructional benefits without personal cost. GenAI developers should offer affordable educational versions and clearly demonstrate how the tools support tasks such as accompaniment creation or music exercise design in teaching contexts. Pre-service music teachers should use available GenAI tools for lesson planning and music material creation, allowing them to experience its benefits and develop positive attitudes toward using the technology in teaching.

HA was found to significantly impact both ATU and BI, with ATU partially mediating this effect, indicating that frequent use of GenAI tools fosters more favorable evaluations and increases the likelihood of adoption, aligning with previous findings (Chu et al., 2022; He and Ren, 2025; Wijaya and Weinhandl, 2022). Frequent use of information technology tools likely leads to greater familiarity and comfort, allowing users to see the technology's value more clearly (Wu and Kuo, 2008). As users incorporate these tools into their routine teaching tasks, they become more confident in their ability to use them effectively, which enhances their overall evaluation of the technology (Xu et al., 2025). This regular engagement reinforces positive perceptions, making teachers more open to adopting GenAI as a regular part of their teaching practice. Educational administrators should create opportunities for pre-service music teachers to observe their peers using GenAI tools in real teaching tasks, fostering a supportive environment that encourages adoption through positive social reinforcement. GenAI developers should design tools with user-friendly interfaces and provide examples of successful classroom applications, making it easier for teachers to see the practical benefits of the technology and encouraging their willingness to use it. Pre-service music teachers should actively participate in collaborative activities with peers, such as joint lesson planning or teaching practice using GenAI, to gain confidence and develop a positive attitude toward integrating the technology into their future teaching.

6 Limitations

This study has several limitations. First, although the study used a relatively large sample size, the sample is still limited to students from specific universities in Hubei and Jiangxi provinces, which could introduce regional biases. Future studies could include a broader geographic scope, selecting participants from various regions and educational institutions, to enhance the representativeness of the sample. Second, the reliance on self-reported data may lead to social desirability bias, as participants may have provided responses they believe are more socially acceptable. Future research could incorporate objective measures or triangulate self-reported data with other data sources (e.g., classroom observations) to mitigate this bias. Third, while ATU has been shown to have a strong influence on BI and mediates all variables in this study, the decision to include ATU in the UTAUT framework may need to be context-specific. Future research should further examine the role of ATU across different technological and disciplinary contexts to determine whether its inclusion in the UTAUT framework provides consistent theoretical and explanatory value.

7 Conclusion

The purpose of this study is to explore the factors influencing pre-service music teachers' adoption of GenAI in music education. The findings show that ATU significantly influences BI to adopt GenAI, emphasizing the importance of both cognitive and emotional factors in technology acceptance. The practical implications suggest that educational administrators should organize workshops and peer collaborations to foster positive attitudes, while GenAI developers should focus on designing user-friendly tools for music education. Additionally, pre-service music teachers should actively engage in using GenAI tools to develop balanced and informed attitudes, ultimately promoting the adoption of the technology in their teaching practices. This research contributes to the existing literature by extending the UTAUT2 framework, demonstrating the significance of ATU in the context of GenAI adoption for music teachers. However, the study is limited by regional biases and reliance on self-reported data, which can be addressed in future research. Further studies could expand the sample scope, explore other disciplines, and examine the role of ATU in different technological contexts.

References

• 1

  AjzenI. (1991). The theory of planned behavior. Organ. Behav. Hum. Decis. Process.50 (2), 179–211. 10.1016/0749-5978(91)90020-T

  • CrossRef
  • Google Scholar

• 2

  AlmaiahM. A.Al-OtaibiS.LutfiA.AlmomaniO.AwajanA.AlsaaidahA.et al (2022). Employing the TAM model to investigate the readiness of M-learning system usage using SEM technique. Electronics (Basel)11 (8), 1259. 10.3390/electronics11081259

  • CrossRef
  • Google Scholar

• 3

  AlmullaM. A. (2021). Technology acceptance model (TAM) and E-learning system use for education sustainability. Acad. Strat. Manag. J.20.

  • Google Scholar

• 4

  AltalhiM. (2021). Toward a model for acceptance of MOOCs in higher education: the modified UTAUT model for Saudi Arabia. Educ. Inf. Technol.26 (2), 1589–1605. 10.1007/s10639-020-10317-x

  • CrossRef
  • Google Scholar

• 5

  AnandT.RamachandranJ.SambasivanM.BatraG. S. (2019). Impact of hedonic motivation on consumer satisfaction towards online shopping: evidence from Malaysia. e-Service J.11 (1), 56–88. 10.2979/eservicej.11.1.03

  • CrossRef
  • Google Scholar

• 6

  ArantesJ. (2025). Understanding intersections between GenAI and pre-service teacher education: what do we need to understand about the changing face of truth in science education?J. Sci. Educ. Technol.34 (6), 1311–1322. 10.1007/s10956-024-10189-7

  • CrossRef
  • Google Scholar

• 7

  BanduraA. (1986). Social Foundations of Thought and Action: A Social Cognitive Theory.Englewood Cliffs, NJ: Prentice-Hall.

  • Google Scholar

• 8

  BuchananT.SainterP.SaundersG. (2013). Factors affecting faculty use of learning technologies: implications for models of technology adoption. J. Comput. High. Educ.25 (1), 1–11. 10.1007/s12528-013-9066-6

  • CrossRef
  • Google Scholar

• 9

  CarminesE. G.McIverJ. P. (1981). “Analyzing Models with Unobserved Variables: analysis of Covariance Structures,” in Social Measurement: Current Issues, eds. BohrnstedtG. W.BorgattaE. F. (Beverly Hills: Sage Publications, Inc), 65–115.

  • Google Scholar

• 10

  CarnovaliniF.SantoL. E.WigginsG. A. (2025). Personalized music education: a systematic review of AI generation methods. IEEE. Access.13, 207433–207447. 10.1109/ACCESS.2025.3639898

  • CrossRef
  • Google Scholar

• 11

  ChandaR. S.PabalkarV.SharmaS. (2024). Attitude and behavioral intention for using metaverse in education: learner’s perspective. J. Appl. Res. High. Educ.16 (5), 2168–2184. 10.1108/jarhe-07-2023-0307

  • CrossRef
  • Google Scholar

• 12

  ChuT.-H.ChaoC.-M.LiuH.-H.ChenD.-F. (2022). Developing an extended theory of UTAUT 2 model to explore factors influencing Taiwanese consumer adoption of intelligent elevators. Sage. Open.12 (4), 21582440221142209. 10.1177/21582440221142209

  • CrossRef
  • Google Scholar

• 13

  DahriN. A.Al-RahmiW. M.AlmogrenA. S.YahayaN.VighioM. S.Al-maatuokQ.et al (2023). Acceptance of mobile learning technology by teachers: influencing mobile self-efficacy and 21st-century skills-based training. Sustainability15 (11), 8514. 10.3390/su15118514

  • CrossRef
  • Google Scholar

• 14

  DavisF. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS. Q.13 (3), 319–340. 10.2307/249008

  • CrossRef
  • Google Scholar

• 15

  DegirmenciK.BreitnerM. H. (2017). Consumer purchase intentions for electric vehicles: is green more important than price and range?Transp. Res. D-Transp. Environ.51, 250–260. 10.1016/j.trd.2017.01.001

  • CrossRef
  • Google Scholar

• 16

  DuW.LiangR.-y. (2024). Teachers’ continued VR technology usage intention: an application of the UTAUT2 model. Sage. Open.14 (1), 21582440231220112. 10.1177/21582440231220112

  • CrossRef
  • Google Scholar

• 17

  DwivediY. K.RanaN. P.JeyarajA.ClementM.WilliamsM. D. (2019). Re-examining the unified theory of acceptance and use of technology (UTAUT): towards a revised theoretical model. Inf. Syst. Front.21 (3), 719–734. 10.1007/s10796-017-9774-y

  • CrossRef
  • Google Scholar

• 18

  FangZ. (2026). Integrating generative AI in higher art education: a systematic review of tools, pedagogies, and practices. SN Comput. Sci.7 (2), 215. 10.1007/s42979-026-04788-x

  • CrossRef
  • Google Scholar

• 19

  FishbeinM.AjzenI. (1975). Belief, Attitude, Intention and Behaviour: An Introduction to Theory and Research. Reading, MA: Addison-Wesley, 290.

  • Google Scholar

• 20

  FornellC.LarckerD. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. J. Mark. Res.18 (1), 39–50. 10.2307/3151312

  • CrossRef
  • Google Scholar

• 21

  GranićA. (2022). Educational technology adoption: a systematic review. Educ Inf Technol (Dordr)27 (7), 9725–9744. 10.1007/s10639-022-10951-7

  • CrossRef
  • Google Scholar

• 22

  HairJ.HultG. T. M.RingleC.SarstedtM. (2022). A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM) (3rd ed.). Thousand Oaks, CA: Sage.

  • Google Scholar

• 23

  HeS.RenY. (2025). Exploring pre-service music teachers’ acceptance of generative artificial intelligence: a PLS-SEM-ANN approach. Front. Psychol.16, 1571279. 10.3389/fpsyg.2025.1571279

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 24

  HeringtonJ.BorasiR.GuerreroB. J.MillerD. E.KoernerB.HanY. J.et al (2026). Musicians’ ethical concerns about AI: an interview study. AI. Soc.41 (2), 1075–1088. 10.1007/s00146-025-02601-6

  • CrossRef
  • Google Scholar

• 25

  HerremansD.ChuanC.-H.ChewE. (2017). A functional taxonomy of music generation systems. ACM Comput. Surv.50 (5), 69. 10.1145/3108242

  • CrossRef
  • Google Scholar

• 26

  HolsterJ. (2024). Augmenting music education through AI: practical applications of ChatGPT. Music Educ. J.110 (4), 36–42. 10.1177/00274321241255938

  • CrossRef
  • Google Scholar

• 27

  HuL.-t.BentlerP. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Struct. Equ. Model. Multidiscip. J.6 (1), 1–55. 10.1080/10705519909540118

  • CrossRef
  • Google Scholar

• 28

  HumphreysJ. T. (2006). Toward a reconstruction of ‘creativity’ in music education. Br. J. Music Educ.23, 351–361. 10.1017/S0265051706007029

  • CrossRef
  • Google Scholar

• 29

  JhaM.AtifA. (2025). Reimagining pedagogy for the GenAI era: frameworks, challenges and institutional strategies. Australas. J. Educ. Technol.41 (5), 56–73. 10.14742/ajet.10645

  • CrossRef
  • Google Scholar

• 30

  KimH. Y. (2013). Statistical notes for clinical researchers: assessing normal distribution (2) using skewness and kurtosis. Restor. Dent. Endod.38 (1), 52–54. 10.5395/rde.2013.38.1.52

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 31

  KongS. C.YangY.HouC. (2024). Examining teachers’ behavioural intention of using generative artificial intelligence tools for teaching and learning based on the extended technology acceptance model. Comput. Educ.: Artif. Intell.7, 100328. 10.1016/j.caeai.2024.100328

  • CrossRef
  • Google Scholar

• 32

  LaakK.-J.AruJ. (2025). AI and personalized learning: bridging the gap with modern educational goals. Educ. Technol. Soc.28 (4), 133–150.

  • Google Scholar

• 33

  LamC. (2023). Technology-enhanced creativity in K-12 music education: a scoping review. Int. J. Music Educ.42 (4), 691–703. 10.1177/02557614231194073

  • CrossRef
  • Google Scholar

• 34

  LiH.ChaiC. S.WangX. (2025). Promoting pre-service music teachers’ TPACK with generative AI: an intervention through designing with AI. Comput. Educ.: Artif. Intell.9, 100525. 10.1016/j.caeai.2025.100525

  • CrossRef
  • Google Scholar

• 35

  LiuW.HuK. (2025). How social and organizational support influence online teaching experience in China? Exploring the mediating role of technological efficacy and innovation awareness. Sage. Open.15 (2), 21582440251336822. 10.1177/21582440251336822

  • CrossRef
  • Google Scholar

• 36

  LiuY.WangQ.LeiJ. (2025). Adopting generative AI in future classrooms: a study of preservice teachers’ intentions and influencing factors. Behav. Sci.15 (8), 1040. 10.3390/bs15081040

  • CrossRef
  • Google Scholar

• 37

  LoveldyZ. A. C.Tuan IsmailT. A.JubaedahS. S. (2021). Examining the attitude-behavior gap and adoption intention of SHS technology: the role of social influence. Int. J. Appl. Bus. Res.3 (1), 14–24. 10.35313/ijabr.v3i1.119

  • CrossRef
  • Google Scholar

• 38

  LuH.HeL.YuH.PanT.FuK. (2024). A study on teachers’ willingness to use generative AI technology and its influencing factors: based on an integrated model. Sustainability16 (16), 7216. 10.3390/su16167216

  • CrossRef
  • Google Scholar

• 39

  LuQ.YaoY.XiaoL.ZhuX.YinH. (2026). Exploring the impact of a GenAI-supported feedback practice on student feedback literacy in L2 writing. Comput. Assist. Lang. Learn., 1–30. 10.1080/09588221.2025.2605541

  • CrossRef
  • Google Scholar

• 40

  LuoJ. (2024). A critical review of GenAI policies in higher education assessment: a call to reconsider the “originality” of students’ work. Assess. Eval. High. Educ.49 (5), 651–664. 10.1080/02602938.2024.2309963

  • CrossRef
  • Google Scholar

• 41

  Medel-VeraC.BrittonS.GatesW. F. (2025). An exploration of the role of generative AI in fostering creativity in architectural learning environments. Comput. Educ.: Artif. Intell.9, 100501. 10.1016/j.caeai.2025.100501

  • CrossRef
  • Google Scholar

• 42

  MikalefP.PappasI. O.GiannakosM. (2016). An integrative adoption model of video-based learning. Int. J. Inf. Learn. Technol.33 (4), 219–235. 10.1108/ijilt-01-2016-0007

  • CrossRef
  • Google Scholar

• 43

  NaS.HeoS.HanS.ShinY.RohY. (2022). Acceptance model of artificial intelligence (AI)-based technologies in construction firms: applying the technology acceptance model (TAM) in combination with the technology–organisation–environment (TOE) framework. Buildings12 (2), 90. 10.3390/buildings12020090

  • CrossRef
  • Google Scholar

• 44

  NgD. T. K.ChanE. K. C.LoC. K. (2025). Opportunities, challenges and school strategies for integrating generative AI in education. Comput. Educ.: Artif. Intell.8, 100373. 10.1016/j.caeai.2025.100373

  • CrossRef
  • Google Scholar

• 45

  O’DeaX. (2024). Generative AI: is it a paradigm shift for higher education?Stud. High. Educ.49 (5), 811–816. 10.1080/03075079.2024.2332944

  • CrossRef
  • Google Scholar

• 46

  OrC. (2023). The role of attitude in the unified theory of acceptance and use of technology: a meta-analytic structural equation modelling study. Int. J. Technol. Educ. Sci.7, 552. 10.46328/ijtes.504

  • CrossRef
  • Google Scholar

• 47

  PengX.SunK.ShanX.ZhangJ. (2026). Empowerment or dependency? A systematic review of the impacts of intelligent assessment and generative AI on learners’ self-beliefs and cognitive agency in music education. Front. Psychol.17, 1776445. 10.3389/fpsyg.2026.1776445

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 48

  PodsakoffP. M.MacKenzieS. B.LeeJ. Y.PodsakoffN. P. (2003). Common method biases in behavioral research: a critical review of the literature and recommended remedies. J. Appl. Psychol.88 (5), 879–903. 10.1037/0021-9010.88.5.879

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 49

  ReddaE. H. (2020). The influence of utilitarian and hedonic consumption values on consumer attitude towards online shopping and purchasing intentions. Journal of Reviews on Global Economics9, 331–342. 10.6000/1929-7092.2020.09.32

  • CrossRef
  • Google Scholar

• 50

  RogersE. M.SinghalA.QuinlanM. M. (2014). “Diffusion of innovations,” in An Integrated Approach to Communication Theory and Research, eds StacksD. W.SalwenM. B. (New York, NY: Routledge), 432–448.

  • Google Scholar

• 51

  RogersP. L. (2000). Barriers to adopting emerging technologies in education. J. Educ. Comput. Res.22 (4), 455–472. 10.2190/4uje-b6vw-a30n-mce5

  • CrossRef
  • Google Scholar

• 52

  SchererR.SiddiqF.TondeurJ. (2019). The technology acceptance model (TAM): a meta-analytic structural equation modeling approach to explaining teachers’ adoption of digital technology in education. Comput. Educ.128, 13–35. 10.1016/j.compedu.2018.09.009

  • CrossRef
  • Google Scholar

• 53

  SchoonenboomJ. (2014). Using an adapted, task-level technology acceptance model to explain why instructors in higher education intend to use some learning management system tools more than others. Comput. Educ.71, 247–256. 10.1016/j.compedu.2013.09.016

  • CrossRef
  • Google Scholar

• 54

  ShiferawK. B.MengisteS. A.GullslettM. K.ZelekeA. A.TilahunB.TebejeT.et al (2021). Healthcare providers’ acceptance of telemedicine and preference of modalities during COVID-19 pandemics in a low-resource setting: an extended UTAUT model. PLoS One16 (4), e0250220. 10.1371/journal.pone.0250220

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 55

  Sitar-TautD.-A.MicanD. (2021). Mobile learning acceptance and use in higher education during social distancing circumstances: an expansion and customization of UTAUT2. Online Inf. Rev.45 (5), 1000–1019. 10.1108/OIR-01-2021-0017

  • CrossRef
  • Google Scholar

• 56

  TaoD. (2008). Understanding intention to use electronic information resources: a theoretical extension of the technology acceptance model (TAM). AMIA. Annu. Symp. Proc.2008, 717–721.

  • Pubmed Abstract
  • Google Scholar

• 57

  TaylorS.ToddP. A. (1995). Understanding information technology usage: a test of competing models. Inf. Syst. Res.6 (2), 144–176. 10.1287/isre.6.2.144

  • CrossRef
  • Google Scholar

• 58

  TeoT. (2009). The impact of subjective norm and facilitating conditions on pre-service teachers’ attitude toward computer use: a structural equation modeling of an extended technology acceptance model. J. Educ. Comput. Res.40 (1), 89–109. 10.2190/EC.40.1.d

  • CrossRef
  • Google Scholar

• 59

  TondeurJ.van BraakJ.SangG.VoogtJ.FisserP.Ottenbreit-LeftwichA. (2012). Preparing pre-service teachers to integrate technology in education: a synthesis of qualitative evidence. Comput. Educ.59 (1), 134–144. 10.1016/j.compedu.2011.10.009

  • CrossRef
  • Google Scholar

• 60

  TsengT. H.LinS.WangY.-S.LiuH.-X. (2022). Investigating teachers’ adoption of MOOCs: the perspective of UTAUT2. Interact. Learn. Environ.30 (4), 635–650. 10.1080/10494820.2019.1674888

  • CrossRef
  • Google Scholar

• 61

  UnalE.UzunA. M. (2021). Understanding university students’ behavioral intention to use edmodo through the lens of an extended technology acceptance model. Br. J. Educ. Technol.52 (2), 619–637. 10.1111/bjet.13046

  • CrossRef
  • Google Scholar

• 62

  ValtonenT.HoangN.SointuE.NäykkiP.VirtanenA.Pöysä-TarhonenJ.et al (2021). How pre-service teachers perceive their 21st-century skills and dispositions: a longitudinal perspective. Comput. Human. Behav.116, 106643. 10.1016/j.chb.2020.106643

  • CrossRef
  • Google Scholar

• 63

  VenkateshV.MorrisM.DavisG.DavisF. (2003). User acceptance of information technology: toward a unified View1. MIS. Q.27, 425–478. 10.2307/30036540

  • CrossRef
  • Google Scholar

• 64

  VenkateshV.ThongJ.XuX. (2012). Consumer acceptance and use of information technology: extending the unified theory of acceptance and use of technology. MIS. Q.36, 157–178. 10.2307/41410412

  • CrossRef
  • Google Scholar

• 65

  WangH.LiuT. (2026). Psychological determinants of GenAI adoption for foreign language education: an extended UTAUT2 model and sentiment analysis approach. Front. Psychol.16, 1622926. 10.3389/fpsyg.2025.1622926

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 66

  WijayaT. T.WeinhandlR. (2022). Factors influencing students’ continuous intentions for using micro-lectures in the post-COVID-19 period: a modification of the UTAUT-2 approach. Electronics (Basel)11 (13), 1924. 10.3390/electronics11131924

  • CrossRef
  • Google Scholar

• 67

  WuM.-C.KuoF.-Y. (2008). An empirical investigation of habitual usage and past usage on technology acceptance evaluations and continuance intention. Data Base Adv. Inf. Syst.39, 48–73. 10.1145/1453794.1453801

  • CrossRef
  • Google Scholar

• 68

  XuW.ZhuK.ZhouD.WangC.WenC. (2025). Examining the effects of habit and self-efficacy on users’ acceptance of a map-based online learning system via an extended TAM. Educ. Sci.15 (7), 828. 10.3390/educsci15070828

  • CrossRef
  • Google Scholar

• 69

  YalçınY.Bakır-YalçınE.İslamoğluH.UrsavaşÖF (2025). Investigating the role of emotions in pre-service teachers’ acceptance and use of learning management systems. Australas. J. Educ. Technol.41 (4), 32–47. 10.14742/ajet.10201

  • CrossRef
  • Google Scholar

• 70

  ZhangR.WangH.-P. (2024). The impact of Chinese university music teachers’ teaching beliefs on creative teaching behaviors: the mediating role of technological acceptance. Front. Educ.9. 10.3389/feduc.2024.1404541

  • CrossRef
  • Google Scholar