From Appropriate Use to Dependence: How AI Attitudes and Use Motivations Shape University Students’ AI Dependence

From Appropriate Use to Dependence: How AI Attitudes and Use Motivations Shape University Students’ AI Dependence | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article From Appropriate Use to Dependence: How AI Attitudes and Use Motivations Shape University Students’ AI Dependence Jianfeng Yin, Yueying Zhang, Runjie Jiang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8954144/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 13 You are reading this latest preprint version Abstract Whether students develop dependence on generative artificial intelligence (AI) technologies largely depends on their attitudes toward using such tools. Existing studies have primarily explained the formation of AI dependence from the perspectives of usage attitudes and technological characteristics, while paying limited attention to the psychological mechanisms through which attitudes exert their influence. Grounded in Self-Determination Theory, the present study examines how different types of use motivation mediate the relationship between students’ AI attitudes and AI dependence. A total of 405 university students completed a comprehensive questionnaire measuring AI attitudes, use motivations, and AI dependence. The results indicate that: (i) AI attitudes do not directly predict AI dependence; instead, their effects are transmitted through multiple motivational pathways, with intrinsic motivation inhibiting AI dependence, whereas identified regulation facilitates its development; (ii) three distinct AI dependence profiles were identified: dependent, moderate, and conservative users. These profiles exhibited clear stratification, with dependent users showing the highest scores across all dimensions and conservative users the lowest; and (iii) individuals with stronger motivations are more likely to be classified into the dependent user profile, with external motivation playing a particularly salient role. These findings enrich the literature on AI technology dependence and offer practical implications for higher education administrators, tool developers, and educational researchers. Business and commerce/Information systems and information technology Biological sciences/Psychology Social science/Psychology Social science/Science technology and society AI attitude AI dependence AI use motivation Self-determination theory(SDT) generative artificial intelligence (AI) Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Generative artificial intelligence can dynamically adopt to individual learners’ needs (Chiu, 2024 ; Barrett & Pack, 2023 ) and plays an important role in shaping learning strategies, enhancing academic performance, and fostering reflective practice (Hsu et al., 2023 ; Wang & Lin, 2023 ; Xia et al., 2022 ). It has thus been shown to support students’ personalized learning and competence development (Kasneci et al., 2023 ; Wang et al., 2023 ; Chiu, 2023). However, despite its empowering potential, growing concerns persist regarding the possible erosion of the core education values. Excessive reliance on technology may undermine students’ agency, weaken sustained attention, and give rise to academic integrity issues (Annamalai et al., 2025 ; Chiu et al.,2024; Chiu & Rospigliosi, 2025 ). Empirical evidence further indicates that AI overreliance can result in cognitive offloading and the diminished critical thinking (Zhai et al., 2024 ; Lee et al., 2025 ; Gerlich, 2025 ), thereby hindering students’ self-regulated learning and cognitive development (Crawford et al., 2024 ). Accordingly, clarifying the mechanisms underlying the formation of AI dependence is crucial for harnessing technology potential while safeguarding students’ learning and development. Individuals’ attitudes toward external objects shape their behavioral intentions, with positive attitudes strengthening intentions and negative attitudes weakening them (Ajzen, 2001 ). Research indicates that perceived usefulness and ease of use of AI enhance their acceptance, thereby constituting a psychological foundation for technological dependence. Moreover, reliance on highly anthropomorphic and emotionally expressive AI interactions (Moussawi & Koufaris, 2019 ) may erode socio-emotional competencies. This erosion may be reflected in diminished conflict management skills and a narrowed empathic scope (Yang & Xie, 2024 ). Self-Determination Theory provides a framework for understanding ho w AI attitudes translate into AI dependence. It posits that satisfaction of basic psychological needs facilitates intrinsic motivation and the internalization of extrinsic motivation (Ryan & Deci, 2017, 2020). Previous studies further indicates that attitudesinfluence behavioral patterns and quality only after being internalized into different types of motivation (Deci & Ryan, 1985 ; Ryan & Deci, 2000 ; Vansteenkiste et al., 2010 ). For example, Li et al. (2025) found that intrinsic motivation, identified regulation, and introjected regulation were significantly associated with behavioral engagement in AI-supported learning, with intrinsic motivation exerting stronger effects on affective and cognitive engagement. Other SDT-based research further indicate that autonomy is the most robust predictor of sustained technology use motivation (Annamalai et al., 2025 ). Existing studies have provided valuable insights into the relationships among AI attitudes, AI use motivation, and AI dependence (Li et al., 2025a ; Annamalai et al., 2025 ; Mohamed et al., 2025 ; Chiu et al., 2024 ). Despite these contributions, several important gaps remain unaddressed. First, Research has largely centered on technology acceptance theories and related models (e.g. Strzelecki, 2024 ; Menon & Shilpa, 2023; Nawaz et al., 2024 ; Liu & Ma, 2023; Saif et al., 2024 ). SDT-based studies, by contrast, have largely examined how AI supports students’ learning, with limited attention to the mechanisms underlying the formation or prevention of dependence (Annamalai et al., 2025 ; Li et al., 2025b ; Chiu et al.,2024; Xia et al., 2022 ). Second, research on technological dependence has mainly focused to traditional instrumental forms, such as smartphone use (Zhang et al., 2022 ). When AI dependence is considered, it is often treated as a background variable in the studies of learning outcomes (Jia et al., 2025 ), leaving its psychological mechanisms unexplored. Finally, most studies adopt a variable-centered approach to examine the mechanisms of AI dependence (Zhang et al., 2024 ; Zhong et al., 2024 ), which may obscure internal heterogeneity. In contrast, person-centered approaches, such as latent profile analysis, allow for more nuanced identification of dependence patterns and their predictors across different user groups (Bray & Dziak, 2018 ; Lubke & Muthén, 2005 ). Literature Review AI attitudes and AI dependence Behavioral attitude refers to the extent to which an individual evaluates a behavior positively or negatively. In the Technology Acceptance Model, usage attitude is defined as an affective evaluation of using a specific technological system. These frameworks, However, were developed before the AI era. No consensus yet exists on how to conceptualize attitudes toward AI. Some measurement instruments distinguish between positive and negative AI attitudes (Arce et al., 2025 ; Kaya et al., 2024 ; Schepman & Rodway, 2020 ; Bergdahl et al., 2023 ), Others adopt a tripartite framework encompassing cognitive, affective, and behavioral components (Suh & Ahn, 2022 ). Existing research thus provides both theoretical and practical insights. Building on this literature, the present study conceptualizes AI attitude as an integrated configuration of cognition, emotion, and behavior formed through individuals’ engagement with AI. Such attitudes may shape students’ modes, frequency, and intensity of technology use, as well as their willingness to engage in learning (Fidan, 2025 ; Arce et al., 2025 ; Suh & Ahn, 2022 ; Ajlouni et al., 2023 ). These processes may, in turn, contribute to dependence-related experiences. Previous studies indicate that attitudes technology-related attitudes, including positive and negative attitudes, fear of missing out, and task-switching tendencies, predict dependence outcomes (Cocoradă et al., 2018 ). Fidan ( 2025 ) further highlight the dual nature of AI attitudes: positive attitudes enhance engagement but may elevate the risk of technological dependence. Similarly, Arce et al. ( 2025 ) report that both positive and negative attitudes predict AI dependence. The relationship between AI usage attitudes and AI dependence therefore warrants further clarification. Individual motivation and AI dependence Self-Determination theory argues that behavioral outcomes are shaped not o n ly by explicit attitudes or technological characteristics, but by the degreeo f motivational internalization (Deci & Ryan, 1985 ; Ryan & Deci, 2000 ). In traditional contexts, motivation is strongly associated with dependence-related behaviors, These include exercise dependence (González-Cutre & Sicilia, 2012 ), nicotine dependence, substance addiction (Kalivas & Volkow, 2005 ), internet dependence (Sun et al.,2008), and smartphone addiction (Zhang et al., 2014 ). Motivation may facilitate openness to novel technologies. However, it may also produce maladaptive outcomes, including dependence. Consistent with this view, Frenkenberg and Hochman ( 2025 ) found that the intensity of AI use varies across motivational orientations. Within SDT, motivation is classified as intrinsic motivation, extrinsic motivation, and intermediate forms. Identified regulation represents a relatively autonomous form of extrinsic motivation grounded in the personal endorsement of a behavior’s value. Because of its proximity to intrinsic motivation, it is often viewed as a relatively neutral motivational orientation (Hagger et al., 2014 ). It emphasizes the perceived importance of engaging in a behavior. Empirical studies show that intrinsic motivation and identified regulation promote active learning (Harding et al., 2007 ). Individuals with stronger intrinsic motivation are more likely to translate personal interests into action (De Bilde et al., 2011 ), which facilitating technology acceptance (Zhang et al., 2008 ). However, their relationship with dependence remains ambiguous. Some evidence suggests that students may develop reliance on AI tools to satisfy intrinsic needs for competence and relatedness (Salah et al., 2024 ). In contrast, other studies indicate that intrinsic motivation may alleviates technological addiction.It may also reduce ChatGPT dependence indirectly by lowering procrastination. External motivation refers to behavior driven by external incentives (Barkoukis et al., 2008 ; Ryan & Deci, 2000 ). Its sources lies in outcomes separate from the activity itself (De Bilde et al.,2011). Prior studies shows that external motivation supports successful remote learning and enhaces engagement (Namestovski et al., 2018 ). It also occupies a central position in AI motivation systems (Li et al., 2025b ) and may directly facilitate AI dependence. Identified regulation, by contrast, involves internalizing the value of a goal into their self-concept, enabling autonomous action (Guay et al.,2017). Students with higher identified regulation demonstrate greater autonomy, flexibility, and creativity. Although higher identified regulation has been associated with lower social media dependence, AI-related research often subsumes it under general extrinsic motivation, overlooking its relative autonomy. AI attitudes an d learning motivation Attitudes are object-specific, whereas motivation is goal-oriented. The two constructs are nevertheless closely related. Motivation is partly shaped by attitudes (Pham, 2021 ). In educational settings, attitudes influence learners’ general orientations toward learning and constitute a foundation for motivation (Masgoret & Gardner, 2003 ). Positive attitudes enhance motivation, whereas negative attitudes weaken it. Importantly, interventions can transform negative attitudes into positive ones,thereby restoring motivation (Oroujlou & Vahedi, 2011 ). Positive attitudes also promote the sustained use of deep learning strategies, which reinforce intrinsic motivation. In this sense, attitudes function as a driving force throughout the learning process (Oroujlou & Vahedi, 2011 ). In the AI context, positive usage attitudes are associated with stronger learning motivation (Tiwari et al., 2024 ) and greater creativity (He, 2023 ). By contrast, negative or ambivalent attitudes may frustrate basic psychological needs and foster passive AI use. This pattern may increase the risk of dependence. For example, high AI anxiety can undermine self-efficacy and competence satisfaction (Bewersdorff et al., 2024). Students may then adopt externally regulated behaviors to avoid failure (Zhang et al., 2024 ). Even when students acknowledge the instrumental value of AI tools, their motivation may remain passive. Under academic pressure, this orientation can increase vulnerability to problematic use (Zhang et al., 2024 ; Zhong et al., 2024 ). Passive AI use is often linked to lower self-reported cognitive effort (Lee et al., 2025 ) and may further intensify dependence through need frustration (Zhong et al., 2024 ). The present study To address these research gaps, this study aims to examine (i) the effects of AI attitudes on AI dependence and (ii) the mediating role of individual motivation, and (iii) the latent profiles of AI dependence and their predictors. Accordingly, the following research questions are proposed: RQ1: How do university students’ AI use attitudes influence their AI dependence through AI use motivation? RQ2: What latent profiles of AI dependence can be identified? RQ3: How do AI use attitudes and use motivations predict profile membership? The conceptual model is shown in Fig. 1 and Fig. 2. Methods This study adopted both variable-centered and person-centered perspectives to explore the mediating mechanism of university students' AI attitudes and use motivations on AI dependence, as well as the latent profiles of AI dependence and the predictive factors of profile membership. The overall research design and procedures are as follows. Participants A total of 405 undergraduate students from a typical comprehensive public university in China participated in this study. Among them, 115 were freshmen (28.4%), 81 were sophomores (20.0%), 69 were juniors (17.0%), and 140 were seniors and above (34.6%). In terms of gender, 352 were female (86.9%) and 53 were male (13.1%). Instruments AI attitude scale The AI attitude of students was assessed using a revised version of the University Students’ AI Attitude Scale developed by Katsantonis (2024) and Bewersdorff et al. (2024), which evaluated university students’ AI attitudes from three dimensions: cognitive, affective, and behavioral. The cognitive dimension included 1 item (Do you believe that AI tools play an important role in higher education?). The affective dimension contained 1 item (Do you support using AI tools to assist in learning?). The behavioral dimension was measured with 4 items (e.g., When using AI tools, do you attempt to explore different functions to complete tasks?). Confirmatory Factor Analysis (CFA) confirmed the strong structural validity of the scale, with the model fit indices showing a good fit: χ ²(0) = 0.00, CFI = 1.00, TLI = 1.00, RMSEA = 0.00, SRMR = 0.00. AI use motivations scale Students’ AI use motivations were assessed using a revised version of the Academic Motivation Scale by Barkoukis et al. (2008). The scale consisted of 9 items divided into three dimensions: intrinsic motivation, extrinsic motivation, and identified regulation. The intrinsic motivation dimension included three items (e.g., “I use AI learning tools because I find the learning process itself to be enjoyable”; α = 0.81). The extrinsic motivation dimension was measured with three items (e.g., “I use AI learning tools because I want to achieve good grades in the course”; α = 0.81).The identified regulation dimension comprised three items (e.g., “I believe that using AI learning tools is important for my academic growth and future career development”; α = 0.71).. CFA verified the strong structural validity of the scale, with the model fit indices indicating a good alignment: χ ²(24) = 114.73, CFI = 0.94, TLI = 0.90, RMSEA = 0.09, SRMR = 0.04. AI dependency scale Students’ AI dependence was assessed using a revised version of the Facebook Addiction Scale by Andreassen et al. (2012). The revision referred to the AI Dependency Scale adapted by Zhang et al. (2024), as well as the University Student AI Dependency Scale developed and validated by Morales-García et al. (2024). The scale included 5 items (e.g., I feel less confident when completing academic tasks without the help of AI; α = 0.89). CFA confirmed the strong structural validity of the scale, with the model fit indices showing a good fit: χ ²(5) = 29.54, CFI = 0.98, TLI = 0.96, RMSEA = 0.11, SRMR = 0.02. Research procedure All scales were translated and adapted to the Chinese context to ensure their applicability in Chinese higher education. Three members of the research team independently translated the scales from English to Chinese and made adaptations based on the learning context of Chinese university students. A third team member mediated to resolve discrepancies in the translations. The final version of the questionnaire was digitized and uploaded to the online survey platform Wenjuanxing. The sample size was determined a priori using the power analysis tool G*Power 3.1. The results showed that a sample of 405 participants was sufficient to detect a small effect size (0.1) with a 95% power and a 5% level of statistical significance (Faul et al., 2007). The online questionnaire was distributed to university students through WeChat groups, a widely used communication platform in China, to ensure the demographic diversity of the sample and maximize the heterogeneity of the data. Analytic strategy The survey data were analyzed using SPSS 27.0 and Mplus 8.3. Descriptive statistics and correlation analysis were performed with SPSS. Subsequently, a structural equation model (SEM) was employed to examine the influence of AI attitude on AI dependency, and to further test the multiple chain mediating effect of AI uses intrinsic motivation, AI uses identity regulation, AI uses external motivation. We also employed the person-centered latent profile analysis (LPA) in Mplus 8.3. This approach is designed to identify subgroups of students who demonstrate distinct patterns of variable responses (Lubke & Muthén,2005). Compared with traditional clustering methods (e.g. K-means clustering), LPA is more accurate because it models the measurement errors and provides a combination of goodness-of-fit indices for model comparison and selection (Bray & Dziak,2018). LPA enables researchers to (1) identify subgroups of students characterized by distinct combinations of key variables, and (2) investigate factors that predict one’s group or profile membership. To answer RQ2, LPA was used to identify the AI dependency profiles. A combination of goodness-of-fit indices was used to determine the optimal number of profiles (Nylund, Asparouhov & Muthén,2007). Specifically, a lower value of Akaike Information Criteria (AIC), Bayesian Information Criteria (BIC), and sample size adjusted BIC (aBIC) indicate better model fit. A value of entropy higher than 0.8 demonstrates an accurate class separation. A significant p-value of Lo-Mendel-Rubin’s Likelihood ratio test (LMR) and Bootstrap Likelihood ratio test (BLRT) suggested that the K class model fits better than the K-1 class model. To answer RQ3, AI attitude, AI uses identity regulation, AI uses intrinsic motivation, and AI uses external motivation variables were used as independent variables to explore their relationship with AI dependency profile membership. The Mplus automated three-step procedures were used (Asparouhov & Muthén,2014). The odds ratio (OR) value could show to what degree the teaching and learning environment could predict the students’ socio-emotional profile. OR values greater than 1 indicate an increased likelihood of membership in a specific profile compared with the reference profile.4.5 Common Method Bias Testing Common method bias The Harman single-factor test (Podsakoff et al., 2003) was used to conduct an unrotated exploratory factor analysis on all measurement items of AI attitude, AI use motivations (intrinsic, identified regulation, extrinsic), and AI dependence. The results showed that four common factors with eigenvalues greater than 1 were extracted, and the first factor accounted for 33.42% of the total variance, which was below the critical threshold of 40%. This indicated that there was no serious common method bias in this study. Results Descriptive statistics and bivariate correlations Descriptive statistics and bivariate correlation coefficients are shown in Table 1 . Student’ AI attitude showed a positive correlation with AI uses identity regulation ( r = .59, p < 0.001), AI uses intrinsic motivation ( r = .45, p < 0.001), AI uses external motivation ( r = .40, p < 0.001), and AI dependency ( r = .22, p < 0.001). AI uses identity regulation showed a positive correlation with AI uses intrinsic motivation ( r = .59, p < 0.001), AI uses external motivation ( r = .42, p < 0.001), and AI dependency ( r = .19, p < 0.001). AI uses intrinsic motivation showed a positive correlation with AI uses external motivation ( r = .40, p < 0.001), and AI dependency ( r = .12, p = 0.014). AI uses external motivation showed a positive correlation with AI dependency ( r = .44, p < 0.001). Table 1 Descriptive statistics: bivariate correlations, means, deviations (N = 405) 1.Al attitude 1 2 3 4 5 — 2.Al uses identity regulation .59 *** — 3.Al uses intrinsic motivation .45 *** .59 *** — 4.Al uses external motivation .40 *** .42 *** .40 *** — 5.AI dependency .22 *** .19 *** .12 * .44 *** — M 3.67 3.93 3.57 3.30 3.07 SD 0.60 0.70 0.81 0.75 0.85 Note : * p < 0.05; ** p < 0.01; *** p < 0.001. Multiple chain mediation effect of AI use motivations on the relationship between AI attitudes and AI dependence The research model demonstrated a good fit to the data: χ ²/ df = 3.25, RMSEA = .07 (< .08), SRMR = .05 ( .90), CFI = .92 (> .90). The results of the measure model are shown in Fig. 3 and Table 2 . AI attitude positively predicts AI uses identity regulation ( β = .47, p < 0.001), AI uses intrinsic motivation ( β = .51, p < 0.001), AI uses external motivation ( β = .67, p < 0.001), but does not significantly predict AI dependency ( β = .38, p = 0.164). AI uses identity regulation positively predicts AI uses intrinsic motivation ( β = .29, p = 0.035), AI uses external motivation ( β = .23, p = 0.048), and AI dependency ( β = .65, p < 0.001). AI uses intrinsic motivation ( β = − .33, p = 0.014) and AI uses external motivation ( β = − .20, p = 0.375) negatively predicts AI dependency. Figure 3 Path diagram of the multiple chain mediation model Note Measurement model and covariates were omitted from the path diagram due to brevity. p < 0.05; ** p < 0.01; *** p < 0.001. Table 2 Path Analysis Results Independent variables Est. SE. P AI Dependency AI attitude→AI Dependency .38 .28 0.164 AI uses identity regulation→AI Dependency .65 .12 < 0.001 AI uses intrinsic motivation→AI Dependency − .33 .13 0.014 AI uses external motivation→AI Dependency − .20 .23 0.375 AI uses intrinsic motivation AI attitude→AI uses intrinsic motivation .51 .13 < 0.001 AI uses identity regulation→AI uses intrinsic motivation .29 .14 0.035 AI uses external motivation AI attitude→AI uses external motivation .67 .11 < 0.001 AI uses identity regulation→AI uses external motivation .23 .11 0.048 AI uses identity regulation AI attitude→AI uses identity regulation .47 .09 < 0.001 R 2 AI Dependency .368 AI uses identity regulation .218 AI uses intrinsic motivation .488 AI uses external motivation .634 Note : * p < 0.05; ** p < 0.01; *** p < 0.001. Mediation analysis indicated that 5 indirect paths contributed to the overall effect (see Table 3 ). AI uses intrinsic motivation negatively mediated the associations between AI attitude and AI Dependency ( β = − .17, 95% CI [-0.424, -0.056]). AI uses identity regulation positively mediated the associations between AI attitude and AI Dependency ( β = .30, 95% CI [0.178, 0.482]). Likewise, the mediation chain of AI uses intrinsic motivation and AI uses identity regulation negatively mediated the associations between AI attitude and AI Dependency ( β = − .05, 95% CI [-0.148, -0.006]). However the mediating effect of AI uses external motivation ( β = − .14, 95% CI [-0.493, 0.014]) and the mediation chain of AI uses external motivation and AI uses identity regulation ( β = − .02, 95% CI [-0.095, 0.002]) between AI attitude and AI Dependency were nonsignificant, as their confidence intervals included zero. Table 3 Mediation effect analysis Path Est. SE 95%CI BootLLCI BootULCI AI attitude→AI uses intrinsic motivation→AI Dependency − .17 .10 -0.424 -0.056 AI attitude→AI uses external motivation→AI Dependency − .14 .21 -0.493 0.014 AI attitude→AI uses identity regulation→AI Dependency .30 .08 0.178 0.482 AI attitude→AI uses intrinsic motivation→AI uses identity regulation→AI Dependency − .05 .04 -0.148 -0.006 AI attitude→AI uses external motivation→AI uses identity regulation→AI Dependency − .02 .03 -0.095 0.002 Note : * p < 0.05; ** p < 0.01; *** p < 0.001. Latent profile analysis of university students’ AI dependence Latent Profile Analysis was performed in Mplus 8.3 with the 5 items of the AI Dependency Scale as indicators. The optimal number of profiles was determined based on multiple goodness-of-fit indices including AIC, BIC, aBIC, entropy, LMR, and BLRT (Nylund, Asparouhov & Muthén, 2007 ), and the fit indices of models with 1 to 7 profiles are summarized in Table 4 . Table 4 LPA model fit statistics for AI dependency indicators. Number of Profiles K AIC BIC aBIC Minimum Class Size Entropy aLRT 1 10 5852.59 5892.63 5860.90 2 16 5157.19 5221.25 5170.48 39.75% 0.788 688.30 3 22 4897.99 4986.08 4916.27 26.91% 0.850 263.87 *** 4 28 4817.95 4930.05 4841.21 5.19% 0.868 89.56 5 34 4718.33 4854.46 4746.58 4.44% 0.972 64.88 6 40 4695.12 4855.27 4728.35 1.98% 0.973 34.26 7 46 4744.85 4929.03 4783.07 0.49% 0.905 8.93 The results showed that: the AIC, BIC, and aBIC values decreased with the increase in the number of profiles; the three-profile model had an entropy value of 0.850 (> 0.8), and the LMR and BLRT tests were both significant, indicating good class separation of the model; although the fit indices of the four-profile and above models were slightly improved, the minimum class size was less than 10%, which indicated overclassification. Therefore, the three-profile model was identified as the optimal model. Combined with the AI dependence level of each profile, the three profiles were named: Conservative Users, Moderate Users, and Dependent Users. The basic characteristics and descriptive statistics of each profile are presented in Table 5 and Fig. 4 :The first profile was labeled dependent user profile (29.4% of participants, N = 119) due to having the highest levels of AI dependency. The second profile was labeled moderate user profile (43.7% of participants, N = 177) with moderate values in the five socio-emotional skills. The third profile was labeled conservative user profile (26.9% of participants, N = 109) due to relatively low levels of AI dependency. Table 5 Comparisons of mean differences in indicator variables across the three profiles. Conservative user ( N = 109) Moderate user ( N = 177) Dependent user ( N = 119) ANOVA F (2,402) η ² AI attitude 3.59(0.70) 3.60(0.55) 3.83(0.54) 6.31 ** 0.03 AI uses identity regulation 3.93(0.73) 3.80(0.68) 4.11(0.67) 7.25 ** 0.04 AI uses intrinsic motivation 3.62(0.90) 3.47(0.72) 3.68(0.84) 2.80 0.01 AI uses external motivation 3.02(0.71) 3.19(0.70) 3.73(0.66) 33.99 *** 0.15 The role of AI attitude, AI uses identity regulation, AI uses intrinsic motivation, and AI uses external motivation in predicting profile membership Table 6 summarizes the relationship between AI attitude, AI uses identity regulation, AI uses intrinsic motivation, AI uses external motivation and profile membership. Compared with the moderate user profiles, students were characterized by lower levels of AI attitude (OR = 2.08, p = 0.001), AI uses intrinsic motivation (OR = 1.46, p = 0.031), AI uses identity regulation (OR = 2.19, p = 0.001), and higher AI uses external motivation (OR = 3.75, p < 0.001) were more likely to be members of the Dependent user profile than the Moderate user profiles. Likewise, Compared with the conservative user profiles, students were characterized by lower levels of AI attitude (OR = 2.09, p = 0.005), and higher AI uses external motivation (OR = 5.23, p < 0.001) were more likely to be members of the Dependent user profile than the conservative user profiles. Table 6 Logistic Regression Results of Gender and SDT with AI on AI dependency profile Moderate user vs Conservative user Dependent user vs Conservative user Dependent user vs Moderate user Coef. OR P Coef. OR P Coef. OR P AI attitude 0.008 1.008 0.975 0.738 2.092 0.005 0.730 2.076 0.001 AI uses intrinsic motivation -0.277 0.758 0.150 0.104 1.110 0.600 0.381 1.464 0.031 AI uses identity regulation -0.345 0.709 0.124 0.440 1.553 0.063 0.785 2.191 0.001 AI uses external motivation 0.334 1.396 0.111 1.655 5.233 < 0.001 1.321 3.748 < 0.001 Discussion Empirical findings First, this study finds that attitudes toward AI do not directly predict dependence on generative artificial intelligence; instead, their influence is primarily transmitted indirectly through motivational mechanisms (RQ1). This finding is consistent with prior research suggesting that positive AI attitudes may be associated with a higher risk of dependence (Zhai et al., 2024 ), while further revealing the complexity of the underlying mechanisms. Specifically, attitudes per se do not exert a direct effect on AI dependence; rather, they shape individuals’ usage motivations, which in turn regulate or amplify dependence-related outcomes. This result echoes the conclusions of Tiwari et al. ( 2024 ), who demonstrated that positive attitudes toward technology mainly influence learning behaviors by enhancing learning motivation. Taken together, these findings indicate that the relationship between AI attitudes and AI dependence is mediated rather than direct, thereby refining existing understandings of how these two constructs are linked. The second empirical finding highlights heterogeneity across motivational types in their influence on generative AI dependence. The critical distinction lies not in the strength of motivation, but in whether the motivation sustains learners’ autonomy (RQ1). Specifically, intrinsic motivation exhibits a significant inhibitory effect on AI dependence. This result aligns with prior studies showing that intrinsic motivation promotes higher-order cognitive processing and autonomous learning (Martín et al., 2023; Mohamed et al., 2025 ), suggesting that AI use driven by interest, exploration, and understanding supports critical and reflective engagement rather than fostering dependence. In contrast, identified regulation significantly promotes AI dependence. This finding implies that when students internalize AI use as a valuable and necessary learning strategy, their reliance on the tool is more easily legitimized and normalized as a dependency-oriented practice. Meanwhile, the effect of external motivation is not statistically significant. This does not negate the role of external motivation; rather, it suggests that AI use primarily driven by grades or task completion often lacks sustained psychological support and is therefore insufficient, on its own, to produce stable patterns of dependence (Li et al., 2025a ). Its influence may instead depend on specific experiential pathways or contextual conditions (Ye et al., 2025 ). The third empirical finding identifies three distinct profiles of generative AI use among university students, namely, dependent, moderate, and conservative users. Individuals holding more positive attitudes toward AI are more likely to be classified into the dependent user group, and those with stronger overall motivation are particularly prone to membership in this group, with external motivation playing an especially salient role. On the one hand, these results indicate that AI dependence is not merely a matter of continuous variation in degree, but rather reflects meaningful latent group differentiation, thereby complementing prior research that has predominantly relied on variable-centered approaches (Zhang et al., 2024 ; Zhong et al., 2024 ). On the other hand, the profile analysis further demonstrates that external motivation serves as a key discriminating factor between dependent and non-dependent users, with dependent students exhibiting significantly higher levels of external motivation. This finding resonates with previous studies linking avoidance-oriented or instrumental motives to elevated risks of dependence (Huang et al., 2024 ; Ye et al., 2025 ). Theoretical contributions The first and second empirical findings elucidate how AI attitudes influence AI dependence through distinct motivational pathways, thereby providing empirical evidence for the relationship between AI attitudes and dependence and underscoring the critical role of student motivation in appropriate AI use. These findings offer empirical support for self-determination theory (SDT) in the context of generative AI. Most SDT-based studies have focused on how need satisfaction or motivational regulation enhances students’ technology acceptance or learning engagement (Chiu, 2021 ; Li & Chiu, 2025 ; Shen et al., 2025 ). In contrast, the present study adopts a different perspective by examining the association between specific SDT motivational types and generative AI dependence, offering both theoretical and empirical justification for more nuanced distinctions among motivational forms and their potential consequences in AI-supported learning contexts. The third empirical finding further reveals that students with more positive AI attitudes and stronger usage motivation are not necessarily protected from dependence; rather, they may be more likely to fall into the AI-dependent category. This conclusion diverges from traditional variable-centered findings and aligns with the inverted U-shaped hypothesis of motivation, suggesting that AI use is influenced not only by motivational type but also by motivational intensity. Moderate levels of motivation may enable students to effectively leverage AI to support learning while avoiding the motivational trap of overreliance. This insight provides a theoretical reference for guiding students toward balanced and autonomous AI use in the era of generative AI. Practical implications The findings of this study yield three practical implications for educational AI developers, higher education administrators, and university instructors. First, the quality of customized generative conversational AI currently relies heavily on users’ prompts, highlighting the urgent need for task- and learner-adaptive design. As demonstrated in this study, generative AI dependence is not directly driven by usage attitudes, but primarily shaped through motivational mechanisms, particularly the buffering role of intrinsic motivation and the risk-enhancing effect of identified regulation. For educational AI developers, tool design should therefore extend beyond improving efficiency and convenience to actively support motivational regulation. For example, visualizing AI reasoning processes and reducing the opacity of algorithmic decision-making may help learners engage cognitively during AI interaction. Such designs can stimulate intrinsic motivation while weakening implicit messages that frame AI as an indispensable learning tool, thereby encouraging learners to use AI while maintaining autonomous cognitive processing and avoiding the internalization of AI as an irreplaceable learning strategy. Second, higher education administrators should remain vigilant regarding the reinforcing effect of performance-oriented evaluation systems on students’ external motivation when formulating policies and governance frameworks related to generative AI. The results indicate that external motivation plays a decisive role in distinguishing dependent from non-dependent user groups, suggesting that institutional designs emphasizing grades, efficiency, or task completion may structurally amplify dependence risks. As Xu and Gao ( 2025 ) argue, AI governance should prioritize human-centered leadership across all levels, aiming to empower humans and safeguard human agency while preventing technological alienation. Accordingly, universities may proactively regulate technology dependence by revising assessment systems, for instance, by incorporating process-oriented evaluation, reflective assignments, or staged comprehension checks, to shift students’ focus toward learning processes and depth of understanding rather than treating generative AI as the optimal tool for assessment performance. Such institutional arrangements can attenuate the role of external motivation in driving dependence at the organizational level. While existing studies have largely focused on universal AI ethics guidelines or usage norms (Hagendorff, 2020 ; Morley et al., 2020 ; Contractor et al., 2022 ), future research should further explore differentiated AI usage strategies across user groups to prevent dependence. Third, university instructors play a critical role in reducing dependence risk by intentionally guiding students to reflect on their generative AI use during instruction. Teachers’ behaviors are a key determinant of students’ motivation (Ahmadi et al., 2023 ), making it essential to understand how instructional guidance and support shape motivational regulation in AI-supported learning (Chiu et al., 2024 ). The findings demonstrate that intrinsic motivation effectively inhibits generative AI dependence and that motivational autonomy is more influential than attitudes alone. Accordingly, instructors may integrate AI-use reflection into routine teaching practices, for example, by asking students to articulate their purposes, processes, and perceived learning impacts of AI use in assignments or learning journals. Such practices encourage students to become aware of their underlying motivations and avoid unreflective substitution of AI for thinking. Through this approach, generative AI can be positioned as a tool that supports understanding and exploration rather than a shortcut that replaces cognition, thereby reducing the likelihood of dependence at the pedagogical level. This implication is consistent with calls for higher education instructors to develop essential competencies for the AI era by optimizing instructional strategies and helping students appropriately position the role of AI in learning (Chiu, 2024 ). Conclusion, limitations, and future research Using a mixed-methods approach, this study examined the effects of AI attitudes and usage motivation on generative AI dependence. The results indicate that university students’ attitudes toward AI do not exert a direct influence on AI dependence; rather, their effects are fully mediated by usage motivation. Specifically, intrinsic motivation significantly inhibits dependence, identified regulation significantly promotes dependence, and the direct effect of external motivation is unstable. Moreover, students’ AI dependence can be categorized into three distinct profiles: dependent, moderate, and conservative users. Both positive AI attitudes and high levels of usage motivation increase the likelihood of belonging to the dependent group, with external motivation emerging as a key predictor of profile membership. By integrating variable-centered and person-centered perspectives, this study advances understanding of the mechanisms underlying AI dependence and provides theoretical support and practical insights for guiding students toward appropriate AI use in higher education. Despite its theoretical and practical contributions, this study has several limitations. First, the reliance on self-report measures makes it difficult to capture the dynamic interactions between learners and AI in authentic learning contexts, which may constrain deeper understanding of the mechanisms underlying AI dependence. Future research may employ qualitative approaches, such as interviews or case studies, to explore the complex interplay among these factors in specific contexts. Second, the use of cross-sectional data limits causal inference among variables. Longitudinal designs are therefore recommended to more clearly identify causal relationships. Finally, factors such as task type, disciplinary background, and task difficulty may influence motivational structures and dependence patterns. Future studies could incorporate diverse learning contexts into analytical models to enhance the external validity of the findings. Declarations Data Availability Statement The datasets used during the current study are available from the corresponding author on reasonable request. Competing Interests The authors declare no competing interests. Ethical Statements Ethical Approval Ethical approval for this study was obtained from the Ethics Committee of our University on 08 August 2025. All procedures involving participants were conducted in accordance with the relevant guidelines and regulations, and strictly adhered to the ethical principles of the Declaration of Helsinki. The ethical approval covered the entire research project, including the research protocol, participant recruitment, questionnaire-based data collection, data analysis, and the publication of anonymised research results. The Ethics Committee confirmed that the legitimate rights and interests of the research participants were adequately protected and formally approved the overall research plan. Informed Consent This study was a non-interventional questionnaire survey. Participants received verbal notification between October 9 and December 8, 2025, before taking part. The questionnaire cover page explained the study objectives, procedures, voluntary participation, and data use. Participants were informed that their responses would remain confidential and anonymous, data would be used for academic purposes only, and participation posed no foreseeable risks. They also agreed to publication of the results. Consent covered participation, use of collected data for analysis, and publication of anonymized findings. All participants completed the survey voluntarily, with no obligation. References Ahmadi A, Noetel M, Parker P, Ryan RM, Ntoumanis N, Reeve J, Lonsdale C (2023) A classification system for teachers’ motivational behaviors recommended in self-determination theory interventions. J Educ Psychol 115(8):1158. ttps://doi.org/10.1037/edu0000783 Ajlouni AO, Wahba FA-A, Almahaireh AS (2023) Students’ attitudes towards using ChatGPT as a learning tool: The case of the University of Jordan. Int J Interact Mob Technol 17(18):99–117. ttps://doi.org/10.3991/ijim.v17i18.41753 Ajzen I (2001) Nature and operation of attitudes. Ann Rev Psychol 52:27–58. ttps://doi.org/10.1146/annurev.psych.52.1.27 Andreassen CS, Torsheim T, Brunborg GS, Pallesen S (2012) Development of a Facebook Addiction Scale. Psychol Rep 110(2):501–517. ttps://doi.org/10.2466/02.09.18.PR0.110.2.501-517 Annamalai N, Bervell B, Mireku D, Andoh R, Debo D (2025) Artificial Intelligence in Higher Education: Modelling Students’ Motivation for Continuous Use of ChatGPT Based on a Modified Self-Determination Theory. Computers Educ Artif Intell 8(1):100346. ttps://doi.org/10.1016/j.caeai.2024.100346 Arce CM, Gavilanes JC, Arce EM, Haro EM, Bonilla-Jurado D (2025) Artificial Intelligence in Higher Education: Predictive Analysis of Attitudes and Dependency Among Ecuadorian University Students. Sustainability 17(17):7741. ttps://doi.org/10.3390/su17177741 Asparouhov T, Muthén B (2014) Auxiliary variables in mixture modeling: Three-step approaches using Mplus. Struct Equ Model 21(3):329–341. ttps://doi.org/10.1080/10705511.2014.915181 Barkoukis V, Tsorbatzoudis H, Grouios G, Sideridis G (2008) The assessment of intrinsic and extrinsic motivation and amotivation: Validity and reliability of the Greek version of the Academic Motivation Scale. Assess Education: Principles Policy Pract 15(1):39–55. ttps://doi.org/10.1080/09695940701876128 Barrett A, Pack A (2023) Not quite eye to A.I.: student and teacher perspectives on the use of generative artificial intelligence in the writing process. Int J Educational Technol High Educ 20:59. ttps://doi.org/10.1186/s41239-023-00427-0 Bergdahl J, Latikka R, Celuch M, Savolainen I, Mantere ES, Savela N, Oksanen A (2023) Self-determination and attitudes toward artificial intelligence: Cross-national and longitudinal perspectives. Telematics Inform 82:102013. ttps://doi.org/10.1016/j.tele.2023.102013 Bray BC, Dziak JJ (2018) Commentary on latent class, latent profile, and latent transition analysis for characterizing individual differences in learning. Learn Individual Differences 66:105–110. ttps://doi.org/10.1016/j.lindif.2018.06.001 Chiu TKF (2021) Applying the self-determination theory (SDT) to explain student engagement in online learning during the COVID-19 pandemic. J Res Technol Educ 54(sup1):S14–S30. ttps://doi.org/10.1080/15391523.2021.1891998 Chiu TKF (2024) The impact of Generative AI (AI) on practices, policies and research direction in education: a case of ChatGPT and Midjourney. Interact Learn Environ 32(10):6187–6203. ttps://doi.org/10.1080/10494820.2023.2253861 Chiu TKF, Moorhouse BL, Chai CS, Ismailov M (2024) Teacher support and student motivation to learn with Artificial Intelligence (AI) based chatbot. Interact Learn Environ 32(7):3240–3256. ttps://doi.org/10.1080/10494820.2023.2172044 Chiu TKF, Rospigliosi PA (2025) Encouraging human-AI collaboration in interactive learning environments. Interact Learn Environ 33(2):921–924. ttps://doi.org/10.1080/10494820.2025.2471199 Cocoradă E, Maican CI, Cazan A-M, Maican MA (2018) Assessing the smartphone addiction risk and its associations with personality traits among adolescents. Child Youth Serv Rev 93:345–354. ttps://doi.org/10.1016/j.childyouth.2018.08.006 Contractor D, McDuff D, Haines JK, Lee J, Hines C, Hecht B, Li H (2022) Behavioral use licensing for responsible AI. In Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency (pp. 778–788). ttps://doi.org/10.1145/3531146.3533143 Crawford J, Allen KA, Pani B, Cowling M (2024) When artificial intelligence substitutes humans in higher education: the cost of loneliness, student success, and retention. Stud High Educ 49(5):883–897. ttps://doi.org/10.1080/03075079.2024.2326956 De Bilde J, Vansteenkiste M, Lens W (2011) Understanding the association between future time perspective and self-regulated learning through the lens of self-determination theory. Learn Instruction 21(3):332–344. ttps://doi.org/10.1016/j.learninstruc.2010.03.002 Deci EL, Ryan RM (1985) Intrinsic motivation and self-determination in human behavior. Plenum. ttps://doi.org/10.1007/978-1-4899-2271-7 Faul F, Erdfelder E, Lang AG, Buchner A (2007) G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 39(2):175–191. ttps://doi.org/10.3758/bf03193146 Fidan Ü (2025) Individual Privacy Perception in the Digital Age: The Interaction of Artificial Intelligence Attitude and Dependency. OPUS J Soc Res 22(5):869–881. ttps://doi.org/10.26466/opusjsr.1725180 Frenkenberg A, Hochman G (2025) It’s Scary to Use It, It’s Scary to Refuse It: The Psychological Dimensions of AI Adoption-Anxiety, Motives, and Dependency. Systems 13(2):82. ttps://doi.org/10.3390/systems13020082 Gerlich M (2025) AI tools in society: Impacts on cognitive offloading and the future of critical thinking. Societies 15(1):6. ttps://doi.org/10.3390/soc15010006 González-Cutre D, Sicilia A (2012) Motivation and exercise dependence: a study based on self-determination theory. Res Q Exerc Sport 83(2):318–329. ttps://doi.org/10.1080/02701367.2012.10599863 Guay F, Denault A-S, Renauld S (2017) School attachment and relatedness with parents, friends and teachers as predictors of students’ intrinsic and identified regulation. Contemp Educ Psychol 51:416–428. ttps://doi.org/10.1016/j.cedpsych.2017.10.001 Hagendorff T (2020) The Ethics of AI Ethics: An Evaluation of Guidelines. Minds Machines 30:99–120. ttps://doi.org/10.1007/s11023-020-09517-8 Hagger MS, Hardcastle SJ, Chater A, Mallett C, Pal S, Chatzisarantis NL (2014) Autonomous and controlled motivational regulations for multiple health-related behaviors: between- and within-participants analyses. Health Psychol Behav Med 2(1):565–601. ttps://doi.org/10.1080/21642850.2014.912945 Harding TS, Vanasupa L, Savage RN, Stolk JD (2007) Work-in-progress - Self-directed learning and motivation in a project-based learning environment. In 2007 37th Annual Frontiers In Education Conference - Global Engineering: Knowledge Without Borders, Opportunities Without Passports (pp. F2G-3–F2G-4). IEEE. ttps://doi.org/10.1109/FIE.2007.4418013 He WJ (2023) Positive and negative affect facilitate creativity motivation: Findings on the effects of habitual mood and experimentally induced emotion. Front Psychol 14:1014612. ttps://doi.org/10.3389/fpsyg.2023.1014612 Hsu TC, Huang HL, Hwang GJ, Chen MS (2023) Effects of incorporating an expert decision-making mechanism into chatbots on students’ achievement, enjoyment, and anxiety. Educational Technol Soc 26(1):218–231. https://www.jstor.org/stable/48707978 Huang S, Lai X, Ke L, Li Y, Wang H, Zhao X, Dai X, Wang Y (2024) AI Technology panic-is AI Dependence Bad for Mental Health? A Cross-Lagged Panel Model and the Mediating Roles of Motivations for AI Use Among Adolescents. Psychol Res Behav Manage 17:1087–1102. ttps://doi.org/10.2147/PRBM.S440889 Jia W, Pan L, Neary S (2025) Effect of AI Dependency on University Students' Academic Achievement: The Mediating Role of Self-Efficacy and Moderating Role of Perceived Teacher Caring. Behav Sci 15(10):1348. ttps://doi.org/10.3390/bs15101348 Kalivas PW, Volkow ND (2005) The neural basis of addiction: a pathology of motivation and choice. Am J Psychiatry 162(8):1403–1413. ttps://doi.org/10.1176/appi.ajp.162.8.1403 Kasneci E, Seßler K, Küchemann S, Bannert M, Dementieva D, Fischer F, Kasneci G (2023) ChatGPT for good? On opportunities and challenges of large language models for education. Learn individual differences 103:102274. ttps://doi.org/10.1016/j.lindif.2023.102274 Katsantonis A, Katsantonis I (2024) University students’ attitudes toward artificial intelligence: An exploratory study of the cognitive, emotional, and behavioural dimensions of AI attitudes. Educ Sci 14(9):988. ttps://doi.org/10.3390/educsci14090988 Kaya F, Aydin F, Schepman A, Rodway P, Yetişensoy O, Kaya D, M (2024) The Roles of Personality Traits, AI Anxiety, and Demographic Factors in Attitudes toward Artificial Intelligence. Int J Hum Comput Interact 40(2):497–514. ttps://doi.org/10.1080/10447318.2022.2151730 Lee H-P, Sarkar A, Tankelevitch L, Drosos I, Rintel S, Banks R, Wilson N (2025) The impact of generative AI on critical thinking: Self-reported reductions in cognitive effort and confidence effects from a survey of knowledge workers. In CHI’25: Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems (p. 1121, pp. 1–22). ttps://doi.org/10.1145/3706598.3713778 Li J, King RB, Chai CS, Zhai X, Lee VWY (2025a) The AI Motivation Scale (AIMS): a self-determination theory perspective. J Res Technol Educ 1–22. ttps://doi.org/10.1080/15391523.2025.2478424 Li J, Zhang J, Chai CS, Lee VWY, Zhai X, Wang X, King RB (2025b) Analyzing the network structure of students’ motivation to learn AI: A self-determination theory perspective. npj Sci Learn 10(48). ttps://doi.org/10.1038/s41539-025-00339-w Li Y, Chiu TKF (2025) The mediating effects of needs satisfaction on the relationship between teacher support and student engagement with generative artificial intelligence (AI) chatbots from a self-determination theory (SDT) perspective. Educational Technol Soc 30:20051–20070. ttps://doi.org/10.1007/s10639-025-13574-w Liu G, Ma C (2024) Measuring EFL learners’ use of ChatGPT in informal digital learning of English based on the technology acceptance model. Innov Lang Learn Teach 18(2):125–138. ttps://doi.org/10.1080/17501229.2023.2240316 Lubke GH, Muthén B (2005) Investigating population heterogeneity with factor mixture models. Psychol Methods 10(1):21–39. ttps://doi.org/10.1037/1082-989X.10.1.21 Martín Núñez JL, Ar AY, Pérez Fernández R, Abbas A, Radovanovic D (2023) Does intrinsic motivation mediate? Perceived artificial intelligence (AI) learning and computational thinking of students during the COVID-19 pandemic. Computers Educ Artif Intell 4:100128. ttps://doi.org/10.1016/j.caeai.2023.100128 Masgoret A-M, Gardner RC (2003) Attitudes, motivation, and second language learning: A meta-analysis of studies conducted by Gardner and Associates. Lang Learn 53(1):123–163. ttps://doi.org/10.1111/1467-9922.00212 Mohamed AM, Shaaban TS, Bakry SH, Guillén-Gámez FD, Strzelecki A (2025) Empowering the faculty of education students: Applying AI’s potential for motivating and enhancing learning. Innov High Educ 50(2):587–609. ttps://doi.org/10.1007/s10755-024-09747-z Morales-García WC, Sairitupa-Sanchez LZ, Morales-García SB, Morales-García M (2024) Development and validation of a scale for dependence on artificial intelligence in university students. Front Educ 9:1323898. ttps://doi.org/10.3389/feduc.2024.1323898 Morley J, Floridi L, Kinsey L, Elhalal A (2020) From What to How: An Initial Review of Publicly Available AI Ethics Tools, Methods and Research to Translate Principles into Practices. Sci Eng Ethics 26(4):2141–2168. ttps://doi.org/10.1007/s11948-019-00165-5 Moussawi S, Koufaris M (2019) Perceived intelligence and perceived anthropomorphism of personal intelligent agents: Scale development and validation. In Hawaii International Conference on System Sciences. ttps://doi.org/10.24251/HICSS.2019.015 Namestovski Ž, Major L, Molnár G, Szűts Z, Esztelecki P, Kőrösi G (2018) External motivation, the key to success in the MOOCs framework. Acta Polytech Hungarica 15(6):125–142. ttps://doi.org/10.12700/APH.15.6.2018.6.7 Nawaz N, Arunachalam H, Pathi BK, Gajenderan V (2024) The adoption of artificial intelligence in human resources management practices. Int J Inform Manage Data Insights 4(1):100208. ttps://doi.org/10.1016/j.jjimei.2023.100208 Nylund KL, Asparouhov T, Muthén BO (2007) Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Struct Equ Model 14(4):535–569. ttps://doi.org/10.1080/10705510701575396 Oroujlou N, Vahedi M (2011) Motivation, attitude, and language learning. Procedia-Social Behav Sci 29:994–1000. ttps://doi.org/10.1016/j.sbspro.2011.11.333 Pham T (2021) Attitude and motivation in language learning: A review. J Engl Lang Teach Appl Linguistics 3(5):64–72. ttps://doi.org/10.32996/jeltal.2021.3.5.7 Ryan RM, Deci EL (2000) Self-Determination Theory and the Facilitation of Intrinsic Motivation, Social Development, and Well-Being. Am Psychol 55:68–78. ttps://doi.org/10.1037/0003-066x.55.1.68 Salah M, Abdelfattah F, Alhalbusi H et al (2024) Me and my AI bot: Exploring the 'AIholic' phenomenon and university students' dependency on generative AI chatbots - Is this the new academic addiction? [Preprint]. Research Square. ttps://doi.org/10.21203/rs.3.rs-3508563/v2 Saif N, Khan SU, Shaheen I, Alotaibi FA, Alnfiai MM, Arif M (2024) Chat-GPT; validating Technology Acceptance Model (TAM) in education sector via ubiquitous learning mechanism. Comput Hum Behav 154:108097. ttps://doi.org/10.1016/j.chb.2023.108097 Schepman A, Rodway P (2020) Initial validation of the general attitudes towards Artificial Intelligence Scale. Computers Hum Behav Rep 1:100014. ttps://doi.org/10.1016/j.chbr.2020.100014 Shen X, Mo X, Xia T (2025) Exploring the attitude and use of AI-image among art and design college students based on TAM and SDT. Interact Learn Environ 33(2):1198–1215. ttps://doi.org/10.1080/10494820.2024.2365959 Strzelecki A (2024) Students’ Acceptance of ChatGPT in Higher Education: An Extended Unified Theory of Acceptance and Use of Technology. Innov High Educ 49:223–245. ttps://doi.org/10.1007/s10755-023-09686-1 Suh W, Ahn S (2022) Development and validation of a scale measuring student attitudes toward artificial intelligence. SAGE Open 12(2):21582440221100463. ttps://doi.org/10.1177/21582440221100463 Sun S, Rubin AM, Haridakis PM (2008) The Role of Motivation and Media Involvement in Explaining Internet Dependency. J Broadcast Electron Media 52(3):408–431. ttps://doi.org/10.1080/08838150802205595 Tiwari CK, Bhat MA, Khan ST, Subramaniam R, Khan MAI (2024) What drives students toward ChatGPT? An investigation of the factors influencing adoption and usage of ChatGPT. Interact Technol Smart Educ 21(3):333–355. ttps://doi.org/10.1108/ITSE-04-2023-0061 Vansteenkiste M, Smeets S, Soenens B, Lens W, Matos L, Deci EL (2010) Autonomous and controlled regulation of performance-approach goals: Their relations to perfectionism and educational outcomes. Motivation Emot 34(4):333–353. ttps://doi.org/10.1007/s11031-010-9188-3 Wang CY, Lin JJ (2023) Utilizing artificial intelligence to support analyzing self-regulated learning: A preliminary mixed-methods evaluation from a human-centered perspective. Comput Hum Behav 144:107721. ttps://doi.org/10.1016/j.chb.2023.107721 Wang H, Fu T, Du Y et al (2023) Scientific discovery in the age of artificial intelligence. Nature 620:47–60. ttps://doi.org/10.1038/s41586-023-06221-2 Xia Q, Chiu TKF, Lee M, Sanusi IT, Dai Y, Chai CS (2022) A self-determination theory (SDT) design approach for inclusive and diverse artificial intelligence (AI) education. Comput Educ 189:104582. ttps://doi.org/10.1016/j.compedu.2022.104582 Xu W, Gao Z (2025) An Intelligent Sociotechnical Systems (iSTS) Framework: Enabling a Hierarchical Human-Centered AI (hHCAI) Approach. IEEE Trans Technol Soc 6(1):31–46. ttps://doi.org/10.1109/TTS.2024.3486254 Yang W, Xie Y (2024) Can robots elicit empathy? The effects of social robots’ appearance on emotional contagion. Computers Hum Behavior: Artif Hum 2(1):100049. ttps://doi.org/10.1016/j.chbah.2024.100049 Ye JH, Zhang M, Nong W et al (2025) The relationship between inert thinking and ChatGPT dependence: An I-PACE model perspective. Educational Technol Soc 30:3885–3909. ttps://doi.org/10.1007/s10639-024-12966-8 Zhai C, Wibowo S, Li LD (2024) The effects of over-reliance on AI dialogue systems on students’ cognitive abilities: a systematic review. Smart Learn Environ 11:28. ttps://doi.org/10.1186/s40561-024-00316-7 Zhang KZK, Chen C, Lee MKO (2014) Understanding the role of motives in smartphone addiction. In PACIS 2014 Proceedings (p. 131). https://aisel.aisnet.org/pacis2014/131 Zhang S, Zhao J, Tan W (2008) Extending TAM for online learning systems: An intrinsic motivation perspective. Tsinghua Sci Technol 13(3):312–317. ttps://doi.org/10.1016/S1007-0214(08)70050-6 Zhang S, Zhao X, Zhou T, Kim JH (2024) Do you have AI dependency? The roles of academic self-efficacy, academic stress, and performance expectations on problematic AI usage behavior. Int J Educational Technol High Educ 21:34. ttps://doi.org/10.1186/s41239-024-00467-0 Zhang Y, Ding Y, Huang H, Peng Q, Wan X, Lu G, Chen C (2022) Relationship between insecure attachment and mobile phone addiction: A meta-analysis. Addict Behav 131:107317. ttps://doi.org/10.1016/j.addbeh.2022.107317 Zhong W, Luo J, Lyu Y (2024) How Do Personal Attributes Shape AI Dependency in Chinese Higher Education Context? Insights from Needs Frustration Perspective. PLoS ONE 19(11):e0313314. ttps://doi.org/10.1371/journal.pone.0313314 Additional Declarations No competing interests reported. Supplementary Files AIresearch.sav Analyticalresources.docx Questionnairesurveyguidelinescollectionprocedures.docx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 17 Apr, 2026 Reviews received at journal 14 Apr, 2026 Reviews received at journal 07 Apr, 2026 Reviewers agreed at journal 23 Mar, 2026 Reviews received at journal 20 Mar, 2026 Reviewers agreed at journal 20 Mar, 2026 Reviewers agreed at journal 19 Mar, 2026 Reviewers agreed at journal 19 Mar, 2026 Reviewers invited by journal 18 Mar, 2026 Editor assigned by journal 18 Mar, 2026 Editor invited by journal 18 Mar, 2026 Submission checks completed at journal 12 Mar, 2026 First submitted to journal 11 Mar, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8954144","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":609387466,"identity":"6aa253b1-7d9b-4af0-b5da-20660f687c36","order_by":0,"name":"Jianfeng Yin","email":"","orcid":"","institution":"Qingdao University","correspondingAuthor":false,"prefix":"","firstName":"Jianfeng","middleName":"","lastName":"Yin","suffix":""},{"id":609387467,"identity":"8765a293-4774-4322-b00b-a04fc49be4a3","order_by":1,"name":"Yueying Zhang","email":"","orcid":"","institution":"Qingdao University","correspondingAuthor":false,"prefix":"","firstName":"Yueying","middleName":"","lastName":"Zhang","suffix":""},{"id":609387468,"identity":"6c29a598-9695-4bf3-995f-ccef7c8384e4","order_by":2,"name":"Runjie Jiang","email":"data:image/png;base64,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","orcid":"","institution":"Nanjing University","correspondingAuthor":true,"prefix":"","firstName":"Runjie","middleName":"","lastName":"Jiang","suffix":""}],"badges":[],"createdAt":"2026-02-24 07:39:46","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8954144/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8954144/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105564687,"identity":"c997a86c-68cd-4272-b90e-ae8eabf69912","added_by":"auto","created_at":"2026-03-27 12:50:31","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1095386,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eVariable-centered research conceptual model\u003c/em\u003e\u003c/p\u003e","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8954144/v1/b08ee46c578b832022ef0268.jpg"},{"id":105233400,"identity":"286faf04-9292-4d28-b41e-05cf2c5f196b","added_by":"auto","created_at":"2026-03-23 19:00:23","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":879787,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003ePerson-centered research conceptual model\u003c/em\u003e\u003c/p\u003e","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8954144/v1/45d203bc2f7dce3fd7c85ce2.jpg"},{"id":105233395,"identity":"3a29d5c1-3f90-4e8e-9242-f7341861b9e3","added_by":"auto","created_at":"2026-03-23 19:00:23","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1179914,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003ePath diagram of the multiple chain mediation model\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNote: \u003c/em\u003eMeasurement model and covariates were omitted from the path diagram due to brevity. \u003cem\u003ep\u0026lt;\u003c/em\u003e0.05; \u003csup\u003e**\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e\u0026lt;0.01; \u003csup\u003e***\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e\u0026lt;0.001.\u003c/p\u003e","description":"","filename":"Figure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8954144/v1/d7d7d0ab406bcfa4b04b28ef.jpg"},{"id":105233397,"identity":"c8e7d9ae-b562-4980-9723-16741ed02cc5","added_by":"auto","created_at":"2026-03-23 19:00:23","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":105611,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eAI dependency latent profile analysis.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"Figure4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8954144/v1/21d4e1507698d535c9c6a8b0.jpg"},{"id":105569374,"identity":"87d13af0-60a0-4388-b5f6-c475b022e73f","added_by":"auto","created_at":"2026-03-27 13:12:13","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4491580,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8954144/v1/423644b0-03b1-4ee8-87f9-6d64ae924b38.pdf"},{"id":105233396,"identity":"490e3943-552b-4337-bd24-54de2925cfa6","added_by":"auto","created_at":"2026-03-23 19:00:23","extension":"sav","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":24711,"visible":true,"origin":"","legend":"","description":"","filename":"AIresearch.sav","url":"https://assets-eu.researchsquare.com/files/rs-8954144/v1/06a4a72a1a71f28927e88557.sav"},{"id":105233398,"identity":"6785c2c0-e068-4023-be25-0733fa1ca2bb","added_by":"auto","created_at":"2026-03-23 19:00:23","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":12088,"visible":true,"origin":"","legend":"","description":"","filename":"Analyticalresources.docx","url":"https://assets-eu.researchsquare.com/files/rs-8954144/v1/60adf00139fdb4aaf0ff9dba.docx"},{"id":105564472,"identity":"3c4bbad9-b059-49ff-a53e-297eab497048","added_by":"auto","created_at":"2026-03-27 12:49:42","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":20129,"visible":true,"origin":"","legend":"","description":"","filename":"Questionnairesurveyguidelinescollectionprocedures.docx","url":"https://assets-eu.researchsquare.com/files/rs-8954144/v1/fedae2b36fb6f69796ad843d.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"From Appropriate Use to Dependence: How AI Attitudes and Use Motivations Shape University Students’ AI Dependence","fulltext":[{"header":"Introduction","content":"\u003cp\u003eGenerative artificial intelligence can dynamically adopt to individual learners\u0026rsquo; needs (Chiu, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Barrett \u0026amp; Pack, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) and plays an important role in shaping learning strategies, enhancing academic performance, and fostering reflective practice (Hsu et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Wang \u0026amp; Lin, \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Xia et al., \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). It has thus been shown to support students\u0026rsquo; personalized learning and competence development (Kasneci et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Wang et al., \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Chiu, 2023). However, despite its empowering potential, growing concerns persist regarding the possible erosion of the core education values.\u003c/p\u003e \u003cp\u003eExcessive reliance on technology may undermine students\u0026rsquo; agency, weaken sustained attention, and give rise to academic integrity issues (Annamalai et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Chiu et al.,2024; Chiu \u0026amp; Rospigliosi, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Empirical evidence further indicates that AI overreliance can result in cognitive offloading and the diminished critical thinking (Zhai et al., \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Lee et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Gerlich, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), thereby hindering students\u0026rsquo; self-regulated learning and cognitive development (Crawford et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Accordingly, clarifying the mechanisms underlying the formation of AI dependence is crucial for harnessing technology potential while safeguarding students\u0026rsquo; learning and development.\u003c/p\u003e \u003cp\u003eIndividuals\u0026rsquo; attitudes toward external objects shape their behavioral intentions, with positive attitudes strengthening intentions and negative attitudes weakening them (Ajzen, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2001\u003c/span\u003e). Research indicates that perceived usefulness and ease of use of AI enhance their acceptance, thereby constituting a psychological foundation for technological dependence. Moreover, reliance on highly anthropomorphic and emotionally expressive AI interactions (Moussawi \u0026amp; Koufaris, \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) may erode socio-emotional competencies. This erosion may be reflected in diminished conflict management skills and a narrowed empathic scope (Yang \u0026amp; Xie, \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSelf-Determination Theory provides a framework for understanding ho\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003ew\u003c/span\u003e AI attitudes translate into AI dependence. It posits that satisfaction of basic psychological needs facilitates intrinsic motivation and the internalization of extrinsic motivation (Ryan \u0026amp; Deci, 2017, 2020). Previous studies further indicates that attitudesinfluence behavioral patterns and quality only after being internalized into different types of motivation (Deci \u0026amp; Ryan, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e1985\u003c/span\u003e; Ryan \u0026amp; Deci, \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Vansteenkiste et al., \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2010\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFor example, Li et al. (2025) found that intrinsic motivation, identified regulation, and introjected regulation were significantly associated with behavioral engagement in AI-supported learning, with intrinsic motivation exerting stronger effects on affective and cognitive engagement. Other SDT-based research further indicate that autonomy is the most robust predictor of sustained technology use motivation (Annamalai et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eExisting studies have provided valuable insights into the relationships among AI attitudes, AI use motivation, and AI dependence (Li et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2025a\u003c/span\u003e; Annamalai et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Mohamed et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Chiu et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Despite these contributions, several important gaps remain unaddressed. First, Research has largely centered on technology acceptance theories and related models (e.g. Strzelecki, \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Menon \u0026amp; Shilpa, 2023; Nawaz et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Liu \u0026amp; Ma, 2023; Saif et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). SDT-based studies, by contrast, have largely examined how AI supports students\u0026rsquo; learning, with limited attention to the mechanisms underlying the formation or prevention of dependence (Annamalai et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Li et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2025b\u003c/span\u003e; Chiu et al.,2024; Xia et al., \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Second, research on technological dependence has mainly focused to traditional instrumental forms, such as smartphone use (Zhang et al., \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). When AI dependence is considered, it is often treated as a background variable in the studies of learning outcomes (Jia et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), leaving its psychological mechanisms unexplored. Finally, most studies adopt a variable-centered approach to examine the mechanisms of AI dependence (Zhang et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Zhong et al., \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), which may obscure internal heterogeneity. In contrast, person-centered approaches, such as latent profile analysis, allow for more nuanced identification of dependence patterns and their predictors across different user groups (Bray \u0026amp; Dziak, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Lubke \u0026amp; Muth\u0026eacute;n, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2005\u003c/span\u003e).\u003c/p\u003e"},{"header":"Literature Review","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eAI attitudes and AI dependence\u003c/h2\u003e \u003cp\u003eBehavioral attitude refers to the extent to which an individual evaluates a behavior positively or negatively. In the Technology Acceptance Model, usage attitude is defined as an affective evaluation of using a specific technological system. These frameworks, However, were developed before the AI era. No consensus yet exists on how to conceptualize attitudes toward AI. Some measurement instruments distinguish between positive and negative AI attitudes (Arce et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Kaya et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Schepman \u0026amp; Rodway, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Bergdahl et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), Others adopt a tripartite framework encompassing cognitive, affective, and behavioral components (Suh \u0026amp; Ahn, \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Existing research thus provides both theoretical and practical insights. Building on this literature, the present study conceptualizes AI attitude as an integrated configuration of cognition, emotion, and behavior formed through individuals\u0026rsquo; engagement with AI. Such attitudes may shape students\u0026rsquo; modes, frequency, and intensity of technology use, as well as their willingness to engage in learning (Fidan, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Arce et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Suh \u0026amp; Ahn, \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Ajlouni et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). These processes may, in turn, contribute to dependence-related experiences. Previous studies indicate that attitudes technology-related attitudes, including positive and negative attitudes, fear of missing out, and task-switching tendencies, predict dependence outcomes (Cocoradă et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Fidan (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) further highlight the dual nature of AI attitudes: positive attitudes enhance engagement but may elevate the risk of technological dependence. Similarly, Arce et al. (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) report that both positive and negative attitudes predict AI dependence. The relationship between AI usage attitudes and AI dependence therefore warrants further clarification.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eIndividual motivation and AI dependence\u003c/h3\u003e\n\u003cp\u003eSelf-Determination theory argues that behavioral outcomes are shaped not o\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003en\u003c/span\u003ely by explicit attitudes or technological characteristics, but by the degreeo\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003ef\u003c/span\u003e motivational internalization (Deci \u0026amp; Ryan, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e1985\u003c/span\u003e; Ryan \u0026amp; Deci, \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). In traditional contexts, motivation is strongly associated with dependence-related behaviors, These include exercise dependence (Gonz\u0026aacute;lez-Cutre \u0026amp; Sicilia, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), nicotine dependence, substance addiction (Kalivas \u0026amp; Volkow, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2005\u003c/span\u003e), internet dependence (Sun et al.,2008), and smartphone addiction (Zhang et al., \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Motivation may facilitate openness to novel technologies. However, it may also produce maladaptive outcomes, including dependence. Consistent with this view, Frenkenberg and Hochman (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) found that the intensity of AI use varies across motivational orientations.\u003c/p\u003e \u003cp\u003eWithin SDT, motivation is classified as intrinsic motivation, extrinsic motivation, and intermediate forms. Identified regulation represents a relatively autonomous form of extrinsic motivation grounded in the personal endorsement of a behavior\u0026rsquo;s value. Because of its proximity to intrinsic motivation, it is often viewed as a relatively neutral motivational orientation (Hagger et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). It emphasizes the perceived importance of engaging in a behavior. Empirical studies show that intrinsic motivation and identified regulation promote active learning (Harding et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). Individuals with stronger intrinsic motivation are more likely to translate personal interests into action (De Bilde et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2011\u003c/span\u003e), which facilitating technology acceptance (Zhang et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). However, their relationship with dependence remains ambiguous. Some evidence suggests that students may develop reliance on AI tools to satisfy intrinsic needs for competence and relatedness (Salah et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn contrast, other studies indicate that intrinsic motivation may alleviates technological addiction.It may also reduce ChatGPT dependence indirectly by lowering procrastination. External motivation refers to behavior driven by external incentives (Barkoukis et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Ryan \u0026amp; Deci, \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). Its sources lies in outcomes separate from the activity itself (De Bilde et al.,2011). Prior studies shows that external motivation supports successful remote learning and enhaces engagement (Namestovski et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). It also occupies a central position in AI motivation systems (Li et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2025b\u003c/span\u003e) and may directly facilitate AI dependence. Identified regulation, by contrast, involves internalizing the value of a goal into their self-concept, enabling autonomous action (Guay et al.,2017). Students with higher identified regulation demonstrate greater autonomy, flexibility, and creativity. Although higher identified regulation has been associated with lower social media dependence, AI-related research often subsumes it under general extrinsic motivation, overlooking its relative autonomy.\u003c/p\u003e \u003cp\u003e \u003cb\u003eAI attitudes an\u003c/b\u003e \u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003ed\u003c/span\u003e \u003cb\u003elearning motivation\u003c/b\u003e\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eAttitudes are object-specific, whereas motivation is goal-oriented. The two constructs are nevertheless closely related. Motivation is partly shaped by attitudes (Pham, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In educational settings, attitudes influence learners\u0026rsquo; general orientations toward learning and constitute a foundation for motivation (Masgoret \u0026amp; Gardner, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). Positive attitudes enhance motivation, whereas negative attitudes weaken it. Importantly, interventions can transform negative attitudes into positive ones,thereby restoring motivation (Oroujlou \u0026amp; Vahedi, \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Positive attitudes also promote the sustained use of deep learning strategies, which reinforce intrinsic motivation. In this sense, attitudes function as a driving force throughout the learning process (Oroujlou \u0026amp; Vahedi, \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). In the AI context, positive usage attitudes are associated with stronger learning motivation (Tiwari et al., \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) and greater creativity (He, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). By contrast, negative or ambivalent attitudes may frustrate basic psychological needs and foster passive AI use. This pattern may increase the risk of dependence. For example, high AI anxiety can undermine self-efficacy and competence satisfaction (Bewersdorff et al., 2024). Students may then adopt externally regulated behaviors to avoid failure (Zhang et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Even when students acknowledge the instrumental value of AI tools, their motivation may remain passive. Under academic pressure, this orientation can increase vulnerability to problematic use (Zhang et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Zhong et al., \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Passive AI use is often linked to lower self-reported cognitive effort (Lee et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) and may further intensify dependence through need frustration (Zhong et al., \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e\n\u003ch3\u003eThe present study\u003c/h3\u003e\n\u003cp\u003eTo address these research gaps, this study aims to examine (i) the effects of AI attitudes on AI dependence and (ii) the mediating role of individual motivation, and (iii) the latent profiles of AI dependence and their predictors. Accordingly, the following research questions are proposed:\u003c/p\u003e \u003cp\u003eRQ1: How do university students\u0026rsquo; AI use attitudes influence their AI dependence through AI use motivation?\u003c/p\u003e\n\u003cp\u003eRQ2: What latent profiles of AI dependence can be identified?\u003c/p\u003e\n\u003cp\u003eRQ3: How do AI use attitudes and use motivations predict profile membership?\u003c/p\u003e\n\u003cp\u003eThe conceptual model is shown in Fig. 1 and Fig. 2.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eThis study adopted both variable-centered and person-centered perspectives to explore the mediating mechanism of university students\u0026apos; AI attitudes and use motivations on AI dependence, as well as the latent profiles of AI dependence and the predictive factors of profile membership. The overall research design and procedures are as follows.\u003c/p\u003e\n\u003cdiv id=\"Sec11\"\u003e\n \u003ch2\u003eParticipants\u003c/h2\u003e\n \u003cp\u003eA total of 405 undergraduate students from a typical comprehensive public university in China participated in this study. Among them, 115 were freshmen (28.4%), 81 were sophomores (20.0%), 69 were juniors (17.0%), and 140 were seniors and above (34.6%). In terms of gender, 352 were female (86.9%) and 53 were male (13.1%).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec12\"\u003e\n \u003ch2\u003eInstruments\u003c/h2\u003e\n \u003cdiv id=\"Sec13\"\u003e\n \u003ch2\u003eAI attitude scale\u003c/h2\u003e\n \u003cp\u003eThe AI attitude of students was assessed using a revised version of the University Students\u0026rsquo; AI Attitude Scale developed by Katsantonis (2024) and Bewersdorff et al. (2024), which evaluated university students\u0026rsquo; AI attitudes from three dimensions: cognitive, affective, and behavioral. The cognitive dimension included 1 item (Do you believe that AI tools play an important role in higher education?). The affective dimension contained 1 item (Do you support using AI tools to assist in learning?). The behavioral dimension was measured with 4 items (e.g., When using AI tools, do you attempt to explore different functions to complete tasks?). Confirmatory Factor Analysis (CFA) confirmed the strong structural validity of the scale, with the model fit indices showing a good fit: \u003cem\u003e\u0026chi;\u003c/em\u003e\u0026sup2;(0)\u0026thinsp;=\u0026thinsp;0.00, CFI\u0026thinsp;=\u0026thinsp;1.00, TLI\u0026thinsp;=\u0026thinsp;1.00, RMSEA\u0026thinsp;=\u0026thinsp;0.00, SRMR\u0026thinsp;=\u0026thinsp;0.00.\u003c/p\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec14\"\u003e\n \u003ch2\u003eAI use motivations scale\u003c/h2\u003e\n \u003cp\u003eStudents\u0026rsquo; AI use motivations were assessed using a revised version of the Academic Motivation Scale by Barkoukis et al. (2008). The scale consisted of 9 items divided into three dimensions: intrinsic motivation, extrinsic motivation, and identified regulation. The intrinsic motivation dimension included three items (e.g., \u0026ldquo;I use AI learning tools because I find the learning process itself to be enjoyable\u0026rdquo;; \u003cem\u003e\u0026alpha;\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.81). The extrinsic motivation dimension was measured with three items (e.g., \u0026ldquo;I use AI learning tools because I want to achieve good grades in the course\u0026rdquo;; \u003cem\u003e\u0026alpha;\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.81).The identified regulation dimension comprised three items (e.g., \u0026ldquo;I believe that using AI learning tools is important for my academic growth and future career development\u0026rdquo;; \u003cem\u003e\u0026alpha;\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.71).. CFA verified the strong structural validity of the scale, with the model fit indices indicating a good alignment: \u003cem\u003e\u0026chi;\u003c/em\u003e\u0026sup2;(24)\u0026thinsp;=\u0026thinsp;114.73, CFI\u0026thinsp;=\u0026thinsp;0.94, TLI\u0026thinsp;=\u0026thinsp;0.90, RMSEA\u0026thinsp;=\u0026thinsp;0.09, SRMR\u0026thinsp;=\u0026thinsp;0.04.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec15\"\u003e\n \u003ch2\u003eAI dependency scale\u003c/h2\u003e\n \u003cp\u003eStudents\u0026rsquo; AI dependence was assessed using a revised version of the Facebook Addiction Scale by Andreassen et al. (2012). The revision referred to the AI Dependency Scale adapted by Zhang et al. (2024), as well as the University Student AI Dependency Scale developed and validated by Morales-Garc\u0026iacute;a et al. (2024). The scale included 5 items (e.g., I feel less confident when completing academic tasks without the help of AI; \u003cem\u003e\u0026alpha;\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.89). CFA confirmed the strong structural validity of the scale, with the model fit indices showing a good fit: \u003cem\u003e\u0026chi;\u003c/em\u003e\u0026sup2;(5)\u0026thinsp;=\u0026thinsp;29.54, CFI\u0026thinsp;=\u0026thinsp;0.98, TLI\u0026thinsp;=\u0026thinsp;0.96, RMSEA\u0026thinsp;=\u0026thinsp;0.11, SRMR\u0026thinsp;=\u0026thinsp;0.02.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec16\"\u003e\n \u003ch2\u003eResearch procedure\u003c/h2\u003e\n \u003cp\u003eAll scales were translated and adapted to the Chinese context to ensure their applicability in Chinese higher education. Three members of the research team independently translated the scales from English to Chinese and made adaptations based on the learning context of Chinese university students. A third team member mediated to resolve discrepancies in the translations. The final version of the questionnaire was digitized and uploaded to the online survey platform Wenjuanxing.\u003c/p\u003e\n \u003cp\u003eThe sample size was determined a priori using the power analysis tool G*Power 3.1. The results showed that a sample of 405 participants was sufficient to detect a small effect size (0.1) with a 95% power and a 5% level of statistical significance (Faul et al., 2007). The online questionnaire was distributed to university students through WeChat groups, a widely used communication platform in China, to ensure the demographic diversity of the sample and maximize the heterogeneity of the data.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec17\"\u003e\n \u003ch2\u003eAnalytic strategy\u003c/h2\u003e\n \u003cp\u003eThe survey data were analyzed using SPSS 27.0 and Mplus 8.3. Descriptive statistics and correlation analysis were performed with SPSS. Subsequently, a structural equation model (SEM) was employed to examine the influence of AI attitude on AI dependency, and to further test the multiple chain mediating effect of AI uses intrinsic motivation, AI uses identity regulation, AI uses external motivation.\u003c/p\u003e\n \u003cp\u003eWe also employed the person-centered latent profile analysis (LPA) in Mplus 8.3. This approach is designed to identify subgroups of students who demonstrate distinct patterns of variable responses (Lubke \u0026amp; Muth\u0026eacute;n,2005). Compared with traditional clustering methods (e.g. K-means clustering), LPA is more accurate because it models the measurement errors and provides a combination of goodness-of-fit indices for model comparison and selection (Bray \u0026amp; Dziak,2018). LPA enables researchers to (1) identify subgroups of students characterized by distinct combinations of key variables, and (2) investigate factors that predict one\u0026rsquo;s group or profile membership.\u003c/p\u003e\n \u003cp\u003eTo answer RQ2, LPA was used to identify the AI dependency profiles. A combination of goodness-of-fit indices was used to determine the optimal number of profiles (Nylund, Asparouhov \u0026amp; Muth\u0026eacute;n,2007). Specifically, a lower value of Akaike Information Criteria (AIC), Bayesian Information Criteria (BIC), and sample size adjusted BIC (aBIC) indicate better model fit. A value of entropy higher than 0.8 demonstrates an accurate class separation. A significant p-value of Lo-Mendel-Rubin\u0026rsquo;s Likelihood ratio test (LMR) and Bootstrap Likelihood ratio test (BLRT) suggested that the K class model fits better than the K-1 class model.\u003c/p\u003e\n \u003cp\u003eTo answer RQ3, AI attitude, AI uses identity regulation, AI uses intrinsic motivation, and AI uses external motivation variables were used as independent variables to explore their relationship with AI dependency profile membership. The Mplus automated three-step procedures were used (Asparouhov \u0026amp; Muth\u0026eacute;n,2014). The odds ratio (OR) value could show to what degree the teaching and learning environment could predict the students\u0026rsquo; socio-emotional profile. OR values greater than 1 indicate an increased likelihood of membership in a specific profile compared with the reference profile.4.5 Common Method Bias Testing\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec18\"\u003e\n \u003ch2\u003eCommon method bias\u003c/h2\u003e\n \u003cp\u003eThe Harman single-factor test (Podsakoff et al., 2003) was used to conduct an unrotated exploratory factor analysis on all measurement items of AI attitude, AI use motivations (intrinsic, identified regulation, extrinsic), and AI dependence. The results showed that four common factors with eigenvalues greater than 1 were extracted, and the first factor accounted for 33.42% of the total variance, which was below the critical threshold of 40%. This indicated that there was no serious common method bias in this study.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eDescriptive statistics and bivariate correlations\u003c/h2\u003e \u003cp\u003eDescriptive statistics and bivariate correlation coefficients are shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Student\u0026rsquo; AI attitude showed a positive correlation with AI uses identity regulation (\u003cem\u003er\u003c/em\u003e = .59, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), AI uses intrinsic motivation (\u003cem\u003er\u003c/em\u003e = .45, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), AI uses external motivation (\u003cem\u003er\u003c/em\u003e = .40, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and AI dependency (\u003cem\u003er\u003c/em\u003e = .22, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). AI uses identity regulation showed a positive correlation with AI uses intrinsic motivation (\u003cem\u003er\u003c/em\u003e = .59, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), AI uses external motivation (\u003cem\u003er\u003c/em\u003e = .42, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and AI dependency (\u003cem\u003er\u003c/em\u003e = .19, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). AI uses intrinsic motivation showed a positive correlation with AI uses external motivation (\u003cem\u003er\u003c/em\u003e = .40, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and AI dependency (\u003cem\u003er\u003c/em\u003e = .12, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.014). AI uses external motivation showed a positive correlation with AI dependency (\u003cem\u003er\u003c/em\u003e = .44, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cem\u003eDescriptive statistics: bivariate correlations, means, deviations (N\u0026thinsp;=\u0026thinsp;405)\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e1.Al attitude\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2.Al uses identity regulation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.59\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3.Al uses intrinsic motivation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.45\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.59\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4.Al uses external motivation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.40\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.42\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e.40\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5.AI dependency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.22\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.19\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e.12\u003csup\u003e*\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e.44\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eM\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e3.07\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eSD\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.70\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.81\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.85\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"6\"\u003e\u003cem\u003eNote\u003c/em\u003e: \u003csup\u003e*\u003c/sup\u003e\u003cem\u003ep\u0026thinsp;\u0026lt;\u003c/em\u003e\u0026thinsp;0.05; \u003csup\u003e**\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01; \u003csup\u003e***\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eMultiple chain mediation effect of AI use motivations on the relationship between AI attitudes and AI dependence\u003c/b\u003e \u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThe research model demonstrated a good fit to the data: \u003cem\u003eχ\u003c/em\u003e\u0026sup2;/\u003cem\u003edf\u003c/em\u003e\u0026thinsp;=\u0026thinsp;3.25, RMSEA = .07 (\u0026lt;\u0026thinsp;.08), SRMR = .05 (\u0026lt;\u0026thinsp;.08), TLI = .90 (\u0026gt;\u0026thinsp;.90), CFI = .92 (\u0026gt;\u0026thinsp;.90). The results of the measure model are shown in Fig.\u0026nbsp;3 and Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. AI attitude positively predicts AI uses identity regulation (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.47, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), AI uses intrinsic motivation (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.51, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), AI uses external motivation (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.67, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), but does not significantly predict AI dependency (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.38, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.164). AI uses identity regulation positively predicts AI uses intrinsic motivation (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.29, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.035), AI uses external motivation (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.23, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.048), and AI dependency (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.65, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). AI uses intrinsic motivation (\u003cem\u003eβ\u003c/em\u003e = \u0026minus;\u0026thinsp;.33, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.014) and AI uses external motivation (\u003cem\u003eβ\u003c/em\u003e = \u0026minus;\u0026thinsp;.20, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.375) negatively predicts AI dependency.\u003c/p\u003e \u003cp\u003e \u003cb\u003eFigure 3\u003c/b\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003ePath diagram of the multiple chain mediation model\u003c/h2\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eNote\u003c/strong\u003e \u003cp\u003eMeasurement model and covariates were omitted from the path diagram due to brevity. \u003cem\u003ep\u0026thinsp;\u0026lt;\u003c/em\u003e\u0026thinsp;0.05; \u003csup\u003e**\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01; \u003csup\u003e***\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePath Analysis Results\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIndependent variables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEst.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSE.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003eP\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003eAI Dependency\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI attitude\u0026rarr;AI Dependency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.38\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.164\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI uses identity regulation\u0026rarr;AI Dependency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI uses intrinsic motivation\u0026rarr;AI Dependency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;\u0026thinsp;.33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.014\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI uses external motivation\u0026rarr;AI Dependency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;\u0026thinsp;.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.375\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003eAI uses intrinsic motivation\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI attitude\u0026rarr;AI uses intrinsic motivation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.51\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI uses identity regulation\u0026rarr;AI uses intrinsic motivation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e0.035\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003eAI uses external motivation\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI attitude\u0026rarr;AI uses external motivation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI uses identity regulation\u0026rarr;AI uses external motivation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.048\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI uses identity regulation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI attitude\u0026rarr;AI uses identity regulation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cem\u003eR\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI Dependency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.368\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI uses identity regulation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.218\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI uses intrinsic motivation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.488\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI uses external motivation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.634\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003e\u003cem\u003eNote\u003c/em\u003e: \u003csup\u003e*\u003c/sup\u003e\u003cem\u003ep\u0026thinsp;\u0026lt;\u003c/em\u003e\u0026thinsp;0.05; \u003csup\u003e**\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01; \u003csup\u003e***\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eMediation analysis indicated that 5 indirect paths contributed to the overall effect (see Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). AI uses intrinsic motivation negatively mediated the associations between AI attitude and AI Dependency (\u003cem\u003eβ\u003c/em\u003e = \u0026minus;\u0026thinsp;.17, 95%\u003cem\u003eCI\u003c/em\u003e[-0.424, -0.056]). AI uses identity regulation positively mediated the associations between AI attitude and AI Dependency (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.30, 95%\u003cem\u003eCI\u003c/em\u003e[0.178, 0.482]). Likewise, the mediation chain of AI uses intrinsic motivation and AI uses identity regulation negatively mediated the associations between AI attitude and AI Dependency (\u003cem\u003eβ\u003c/em\u003e = \u0026minus;\u0026thinsp;.05, 95%\u003cem\u003eCI\u003c/em\u003e[-0.148, -0.006]). However the mediating effect of AI uses external motivation (\u003cem\u003eβ\u003c/em\u003e = \u0026minus;\u0026thinsp;.14, 95%\u003cem\u003eCI\u003c/em\u003e[-0.493, 0.014]) and the mediation chain of AI uses external motivation and AI uses identity regulation (\u003cem\u003eβ\u003c/em\u003e = \u0026minus;\u0026thinsp;.02, 95%\u003cem\u003eCI\u003c/em\u003e[-0.095, 0.002]) between AI attitude and AI Dependency were nonsignificant, as their confidence intervals included zero.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cem\u003eMediation effect analysis\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003ePath\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eEst.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eSE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003e95%CI\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eBootLLCI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eBootULCI\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI attitude\u0026rarr;AI uses intrinsic motivation\u0026rarr;AI Dependency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;\u0026thinsp;.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.424\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.056\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI attitude\u0026rarr;AI uses external motivation\u0026rarr;AI Dependency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;\u0026thinsp;.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.493\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.014\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI attitude\u0026rarr;AI uses identity regulation\u0026rarr;AI Dependency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.178\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.482\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI attitude\u0026rarr;AI uses intrinsic motivation\u0026rarr;AI uses identity regulation\u0026rarr;AI Dependency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;\u0026thinsp;.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.148\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.006\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI attitude\u0026rarr;AI uses external motivation\u0026rarr;AI uses identity regulation\u0026rarr;AI Dependency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;\u0026thinsp;.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.095\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003e\u003cem\u003eNote\u003c/em\u003e: \u003csup\u003e*\u003c/sup\u003e\u003cem\u003ep\u0026thinsp;\u0026lt;\u003c/em\u003e\u0026thinsp;0.05; \u003csup\u003e**\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01; \u003csup\u003e***\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eLatent profile analysis of university students\u0026rsquo; AI dependence\u003c/h2\u003e \u003cp\u003eLatent Profile Analysis was performed in Mplus 8.3 with the 5 items of the AI Dependency Scale as indicators. The optimal number of profiles was determined based on multiple goodness-of-fit indices including AIC, BIC, aBIC, entropy, LMR, and BLRT (Nylund, Asparouhov \u0026amp; Muth\u0026eacute;n, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2007\u003c/span\u003e), and the fit indices of models with 1 to 7 profiles are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cem\u003eLPA model fit statistics for AI dependency indicators.\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of\u003c/p\u003e \u003cp\u003eProfiles\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eK\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAIC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eBIC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eaBIC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMinimum\u003c/p\u003e \u003cp\u003eClass Size\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eEntropy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eaLRT\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5852.59\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5892.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e5860.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5157.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5221.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e5170.48\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e39.75%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.788\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e688.30\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e3\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e22\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e4897.99\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e4986.08\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e4916.27\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e26.91%\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u003cb\u003e0.850\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e\u003cb\u003e263.87\u003c/b\u003e\u003csup\u003e\u003cb\u003e***\u003c/b\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4817.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4930.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e4841.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e5.19%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.868\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e89.56\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e34\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4718.33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4854.46\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e4746.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e4.44%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.972\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e64.88\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4695.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4855.27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e4728.35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.98%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.973\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e34.26\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e46\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4744.85\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4929.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e4783.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.49%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.905\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e8.93\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe results showed that: the AIC, BIC, and aBIC values decreased with the increase in the number of profiles; the three-profile model had an entropy value of 0.850 (\u0026gt;\u0026thinsp;0.8), and the LMR and BLRT tests were both significant, indicating good class separation of the model; although the fit indices of the four-profile and above models were slightly improved, the minimum class size was less than 10%, which indicated overclassification. Therefore, the three-profile model was identified as the optimal model.\u003c/p\u003e \u003cp\u003eCombined with the AI dependence level of each profile, the three profiles were named: Conservative Users, Moderate Users, and Dependent Users. The basic characteristics and descriptive statistics of each profile are presented in Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e4\u003c/span\u003e:The first profile was labeled dependent user profile (29.4% of participants, \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;119) due to having the highest levels of AI dependency. The second profile was labeled moderate user profile (43.7% of participants, \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;177) with moderate values in the five socio-emotional skills. The third profile was labeled conservative user profile (26.9% of participants, \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;109) due to relatively low levels of AI dependency.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cem\u003eComparisons of mean differences in indicator variables across the three profiles.\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eConservative user\u003c/p\u003e \u003cp\u003e(\u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;109)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eModerate user\u003c/p\u003e \u003cp\u003e(\u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;177)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eDependent user\u003c/p\u003e \u003cp\u003e(\u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;119)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003eANOVA\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003eF\u003c/em\u003e(2,402)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cem\u003eη\u003c/em\u003e\u0026sup2;\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI attitude\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3.59(0.70)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.60(0.55)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.83(0.54)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e6.31\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI uses identity regulation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3.93(0.73)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.80(0.68)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.11(0.67)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e7.25\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI uses intrinsic motivation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3.62(0.90)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.47(0.72)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.68(0.84)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI uses external motivation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3.02(0.71)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.19(0.70)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.73(0.66)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e33.99\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.15\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eThe role of AI attitude, AI uses identity regulation, AI uses intrinsic motivation, and AI uses external motivation in predicting profile membership\u003c/b\u003e \u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e summarizes the relationship between AI attitude, AI uses identity regulation, AI uses intrinsic motivation, AI uses external motivation and profile membership. Compared with the moderate user profiles, students were characterized by lower levels of AI attitude (OR\u0026thinsp;=\u0026thinsp;2.08, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.001), AI uses intrinsic motivation (OR\u0026thinsp;=\u0026thinsp;1.46, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.031), AI uses identity regulation (OR\u0026thinsp;=\u0026thinsp;2.19, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.001), and higher AI uses external motivation (OR\u0026thinsp;=\u0026thinsp;3.75, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) were more likely to be members of the Dependent user profile than the Moderate user profiles. Likewise, Compared with the conservative user profiles, students were characterized by lower levels of AI attitude (OR\u0026thinsp;=\u0026thinsp;2.09, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.005), and higher AI uses external motivation (OR\u0026thinsp;=\u0026thinsp;5.23, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) were more likely to be members of the Dependent user profile than the conservative user profiles.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cem\u003eLogistic Regression Results of Gender and SDT with AI on AI dependency profile\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"10\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003eModerate user vs Conservative user\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e \u003cp\u003eDependent user vs Conservative user\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e \u003cp\u003eDependent user vs Moderate user\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCoef.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003eP\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eCoef.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003eP\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eCoef.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003e\u003cem\u003eP\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI attitude\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.975\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.738\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e2.092\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.730\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.076\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI uses intrinsic motivation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.277\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.758\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.150\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.104\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.110\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.600\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.381\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e1.464\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.031\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI uses identity regulation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.345\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.709\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.124\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.440\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.553\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.063\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.785\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.191\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI uses external motivation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.334\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.396\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.111\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.655\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e5.233\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.321\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.748\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003eEmpirical findings\u003c/h2\u003e \u003cp\u003eFirst, this study finds that attitudes toward AI do not directly predict dependence on generative artificial intelligence; instead, their influence is primarily transmitted indirectly through motivational mechanisms (RQ1). This finding is consistent with prior research suggesting that positive AI attitudes may be associated with a higher risk of dependence (Zhai et al., \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), while further revealing the complexity of the underlying mechanisms. Specifically, attitudes per se do not exert a direct effect on AI dependence; rather, they shape individuals\u0026rsquo; usage motivations, which in turn regulate or amplify dependence-related outcomes. This result echoes the conclusions of Tiwari et al. (\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), who demonstrated that positive attitudes toward technology mainly influence learning behaviors by enhancing learning motivation. Taken together, these findings indicate that the relationship between AI attitudes and AI dependence is mediated rather than direct, thereby refining existing understandings of how these two constructs are linked.\u003c/p\u003e \u003cp\u003eThe second empirical finding highlights heterogeneity across motivational types in their influence on generative AI dependence. The critical distinction lies not in the strength of motivation, but in whether the motivation sustains learners\u0026rsquo; autonomy (RQ1). Specifically, intrinsic motivation exhibits a significant inhibitory effect on AI dependence. This result aligns with prior studies showing that intrinsic motivation promotes higher-order cognitive processing and autonomous learning (Mart\u0026iacute;n et al., 2023; Mohamed et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), suggesting that AI use driven by interest, exploration, and understanding supports critical and reflective engagement rather than fostering dependence. In contrast, identified regulation significantly promotes AI dependence. This finding implies that when students internalize AI use as a valuable and necessary learning strategy, their reliance on the tool is more easily legitimized and normalized as a dependency-oriented practice. Meanwhile, the effect of external motivation is not statistically significant. This does not negate the role of external motivation; rather, it suggests that AI use primarily driven by grades or task completion often lacks sustained psychological support and is therefore insufficient, on its own, to produce stable patterns of dependence (Li et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2025a\u003c/span\u003e). Its influence may instead depend on specific experiential pathways or contextual conditions (Ye et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe third empirical finding identifies three distinct profiles of generative AI use among university students, namely, dependent, moderate, and conservative users. Individuals holding more positive attitudes toward AI are more likely to be classified into the dependent user group, and those with stronger overall motivation are particularly prone to membership in this group, with external motivation playing an especially salient role. On the one hand, these results indicate that AI dependence is not merely a matter of continuous variation in degree, but rather reflects meaningful latent group differentiation, thereby complementing prior research that has predominantly relied on variable-centered approaches (Zhang et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Zhong et al., \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). On the other hand, the profile analysis further demonstrates that external motivation serves as a key discriminating factor between dependent and non-dependent users, with dependent students exhibiting significantly higher levels of external motivation. This finding resonates with previous studies linking avoidance-oriented or instrumental motives to elevated risks of dependence (Huang et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Ye et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec25\" class=\"Section2\"\u003e \u003ch2\u003eTheoretical contributions\u003c/h2\u003e \u003cp\u003eThe first and second empirical findings elucidate how AI attitudes influence AI dependence through distinct motivational pathways, thereby providing empirical evidence for the relationship between AI attitudes and dependence and underscoring the critical role of student motivation in appropriate AI use. These findings offer empirical support for self-determination theory (SDT) in the context of generative AI. Most SDT-based studies have focused on how need satisfaction or motivational regulation enhances students\u0026rsquo; technology acceptance or learning engagement (Chiu, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Li \u0026amp; Chiu, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Shen et al., \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). In contrast, the present study adopts a different perspective by examining the association between specific SDT motivational types and generative AI dependence, offering both theoretical and empirical justification for more nuanced distinctions among motivational forms and their potential consequences in AI-supported learning contexts.\u003c/p\u003e \u003cp\u003eThe third empirical finding further reveals that students with more positive AI attitudes and stronger usage motivation are not necessarily protected from dependence; rather, they may be more likely to fall into the AI-dependent category. This conclusion diverges from traditional variable-centered findings and aligns with the inverted U-shaped hypothesis of motivation, suggesting that AI use is influenced not only by motivational type but also by motivational intensity. Moderate levels of motivation may enable students to effectively leverage AI to support learning while avoiding the motivational trap of overreliance. This insight provides a theoretical reference for guiding students toward balanced and autonomous AI use in the era of generative AI.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section2\"\u003e \u003ch2\u003ePractical implications\u003c/h2\u003e \u003cp\u003eThe findings of this study yield three practical implications for educational AI developers, higher education administrators, and university instructors. First, the quality of customized generative conversational AI currently relies heavily on users\u0026rsquo; prompts, highlighting the urgent need for task- and learner-adaptive design. As demonstrated in this study, generative AI dependence is not directly driven by usage attitudes, but primarily shaped through motivational mechanisms, particularly the buffering role of intrinsic motivation and the risk-enhancing effect of identified regulation. For educational AI developers, tool design should therefore extend beyond improving efficiency and convenience to actively support motivational regulation. For example, visualizing AI reasoning processes and reducing the opacity of algorithmic decision-making may help learners engage cognitively during AI interaction. Such designs can stimulate intrinsic motivation while weakening implicit messages that frame AI as an indispensable learning tool, thereby encouraging learners to use AI while maintaining autonomous cognitive processing and avoiding the internalization of AI as an irreplaceable learning strategy.\u003c/p\u003e \u003cp\u003eSecond, higher education administrators should remain vigilant regarding the reinforcing effect of performance-oriented evaluation systems on students\u0026rsquo; external motivation when formulating policies and governance frameworks related to generative AI. The results indicate that external motivation plays a decisive role in distinguishing dependent from non-dependent user groups, suggesting that institutional designs emphasizing grades, efficiency, or task completion may structurally amplify dependence risks. As Xu and Gao (\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) argue, AI governance should prioritize human-centered leadership across all levels, aiming to empower humans and safeguard human agency while preventing technological alienation. Accordingly, universities may proactively regulate technology dependence by revising assessment systems, for instance, by incorporating process-oriented evaluation, reflective assignments, or staged comprehension checks, to shift students\u0026rsquo; focus toward learning processes and depth of understanding rather than treating generative AI as the optimal tool for assessment performance. Such institutional arrangements can attenuate the role of external motivation in driving dependence at the organizational level. While existing studies have largely focused on universal AI ethics guidelines or usage norms (Hagendorff, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Morley et al., \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Contractor et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), future research should further explore differentiated AI usage strategies across user groups to prevent dependence.\u003c/p\u003e \u003cp\u003eThird, university instructors play a critical role in reducing dependence risk by intentionally guiding students to reflect on their generative AI use during instruction. Teachers\u0026rsquo; behaviors are a key determinant of students\u0026rsquo; motivation (Ahmadi et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), making it essential to understand how instructional guidance and support shape motivational regulation in AI-supported learning (Chiu et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). The findings demonstrate that intrinsic motivation effectively inhibits generative AI dependence and that motivational autonomy is more influential than attitudes alone. Accordingly, instructors may integrate AI-use reflection into routine teaching practices, for example, by asking students to articulate their purposes, processes, and perceived learning impacts of AI use in assignments or learning journals. Such practices encourage students to become aware of their underlying motivations and avoid unreflective substitution of AI for thinking. Through this approach, generative AI can be positioned as a tool that supports understanding and exploration rather than a shortcut that replaces cognition, thereby reducing the likelihood of dependence at the pedagogical level. This implication is consistent with calls for higher education instructors to develop essential competencies for the AI era by optimizing instructional strategies and helping students appropriately position the role of AI in learning (Chiu, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusion, limitations, and future research","content":"\u003cp\u003eUsing a mixed-methods approach, this study examined the effects of AI attitudes and usage motivation on generative AI dependence. The results indicate that university students\u0026rsquo; attitudes toward AI do not exert a direct influence on AI dependence; rather, their effects are fully mediated by usage motivation. Specifically, intrinsic motivation significantly inhibits dependence, identified regulation significantly promotes dependence, and the direct effect of external motivation is unstable. Moreover, students\u0026rsquo; AI dependence can be categorized into three distinct profiles: dependent, moderate, and conservative users. Both positive AI attitudes and high levels of usage motivation increase the likelihood of belonging to the dependent group, with external motivation emerging as a key predictor of profile membership. By integrating variable-centered and person-centered perspectives, this study advances understanding of the mechanisms underlying AI dependence and provides theoretical support and practical insights for guiding students toward appropriate AI use in higher education.\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eDespite its theoretical and practical contributions, this study has several limitations. First, the reliance on self-report measures makes it difficult to capture the dynamic interactions between learners and AI in authentic learning contexts, which may constrain deeper understanding of the mechanisms underlying AI dependence. Future research may employ qualitative approaches, such as interviews or case studies, to explore the complex interplay among these factors in specific contexts. Second, the use of cross-sectional data limits causal inference among variables. Longitudinal designs are therefore recommended to more clearly identify causal relationships. Finally, factors such as task type, disciplinary background, and task difficulty may influence motivational structures and dependence patterns. Future studies could incorporate diverse learning contexts into analytical models to enhance the external validity of the findings.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e "},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData Availability Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthical Statements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eEthical Approval\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEthical approval for this study was obtained from the Ethics Committee of our University on 08 August 2025. All procedures involving participants were conducted in accordance with the relevant guidelines and regulations, and strictly adhered to the ethical principles of the Declaration of Helsinki. The ethical approval covered the entire research project, including the research protocol, participant recruitment, questionnaire-based data collection, data analysis, and the publication of anonymised research results. The Ethics Committee confirmed that the legitimate rights and interests of the research participants were adequately protected and formally approved the overall research plan.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eInformed Consent\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was a non-interventional questionnaire survey. Participants received verbal notification between October 9 and December 8, 2025, before taking part. The questionnaire cover page explained the study objectives, procedures, voluntary participation, and data use. Participants were informed that their responses would remain confidential and anonymous, data would be used for academic purposes only, and participation posed no foreseeable risks. They also agreed to publication of the results. Consent covered participation, use of collected data for analysis, and publication of anonymized findings. All participants completed the survey voluntarily, with no obligation.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAhmadi A, Noetel M, Parker P, Ryan RM, Ntoumanis N, Reeve J, Lonsdale C (2023) A classification system for teachers\u0026rsquo; motivational behaviors recommended in self-determination theory interventions. J Educ Psychol 115(8):1158. ttps://doi.org/10.1037/edu0000783\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAjlouni AO, Wahba FA-A, Almahaireh AS (2023) Students\u0026rsquo; attitudes towards using ChatGPT as a learning tool: The case of the University of Jordan. Int J Interact Mob Technol 17(18):99\u0026ndash;117. ttps://doi.org/10.3991/ijim.v17i18.41753\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAjzen I (2001) Nature and operation of attitudes. Ann Rev Psychol 52:27\u0026ndash;58. ttps://doi.org/10.1146/annurev.psych.52.1.27\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndreassen CS, Torsheim T, Brunborg GS, Pallesen S (2012) Development of a Facebook Addiction Scale. Psychol Rep 110(2):501\u0026ndash;517. ttps://doi.org/10.2466/02.09.18.PR0.110.2.501-517\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAnnamalai N, Bervell B, Mireku D, Andoh R, Debo D (2025) Artificial Intelligence in Higher Education: Modelling Students\u0026rsquo; Motivation for Continuous Use of ChatGPT Based on a Modified Self-Determination Theory. Computers Educ Artif Intell 8(1):100346. ttps://doi.org/10.1016/j.caeai.2024.100346\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArce CM, Gavilanes JC, Arce EM, Haro EM, Bonilla-Jurado D (2025) Artificial Intelligence in Higher Education: Predictive Analysis of Attitudes and Dependency Among Ecuadorian University Students. Sustainability 17(17):7741. ttps://doi.org/10.3390/su17177741\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAsparouhov T, Muth\u0026eacute;n B (2014) Auxiliary variables in mixture modeling: Three-step approaches using Mplus. Struct Equ Model 21(3):329\u0026ndash;341. ttps://doi.org/10.1080/10705511.2014.915181\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarkoukis V, Tsorbatzoudis H, Grouios G, Sideridis G (2008) The assessment of intrinsic and extrinsic motivation and amotivation: Validity and reliability of the Greek version of the Academic Motivation Scale. Assess Education: Principles Policy Pract 15(1):39\u0026ndash;55. ttps://doi.org/10.1080/09695940701876128\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarrett A, Pack A (2023) Not quite eye to A.I.: student and teacher perspectives on the use of generative artificial intelligence in the writing process. Int J Educational Technol High Educ 20:59. ttps://doi.org/10.1186/s41239-023-00427-0\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBergdahl J, Latikka R, Celuch M, Savolainen I, Mantere ES, Savela N, Oksanen A (2023) Self-determination and attitudes toward artificial intelligence: Cross-national and longitudinal perspectives. Telematics Inform 82:102013. ttps://doi.org/10.1016/j.tele.2023.102013\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBray BC, Dziak JJ (2018) Commentary on latent class, latent profile, and latent transition analysis for characterizing individual differences in learning. Learn Individual Differences 66:105\u0026ndash;110. ttps://doi.org/10.1016/j.lindif.2018.06.001\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChiu TKF (2021) Applying the self-determination theory (SDT) to explain student engagement in online learning during the COVID-19 pandemic. J Res Technol Educ 54(sup1):S14\u0026ndash;S30. ttps://doi.org/10.1080/15391523.2021.1891998\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChiu TKF (2024) The impact of Generative AI (AI) on practices, policies and research direction in education: a case of ChatGPT and Midjourney. Interact Learn Environ 32(10):6187\u0026ndash;6203. ttps://doi.org/10.1080/10494820.2023.2253861\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChiu TKF, Moorhouse BL, Chai CS, Ismailov M (2024) Teacher support and student motivation to learn with Artificial Intelligence (AI) based chatbot. Interact Learn Environ 32(7):3240\u0026ndash;3256. ttps://doi.org/10.1080/10494820.2023.2172044\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChiu TKF, Rospigliosi PA (2025) Encouraging human-AI collaboration in interactive learning environments. Interact Learn Environ 33(2):921\u0026ndash;924. ttps://doi.org/10.1080/10494820.2025.2471199\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCocoradă E, Maican CI, Cazan A-M, Maican MA (2018) Assessing the smartphone addiction risk and its associations with personality traits among adolescents. Child Youth Serv Rev 93:345\u0026ndash;354. ttps://doi.org/10.1016/j.childyouth.2018.08.006\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eContractor D, McDuff D, Haines JK, Lee J, Hines C, Hecht B, Li H (2022) Behavioral use licensing for responsible AI. In \u003cem\u003eProceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency\u003c/em\u003e (pp. 778\u0026ndash;788). ttps://doi.org/10.1145/3531146.3533143\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCrawford J, Allen KA, Pani B, Cowling M (2024) When artificial intelligence substitutes humans in higher education: the cost of loneliness, student success, and retention. Stud High Educ 49(5):883\u0026ndash;897. ttps://doi.org/10.1080/03075079.2024.2326956\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Bilde J, Vansteenkiste M, Lens W (2011) Understanding the association between future time perspective and self-regulated learning through the lens of self-determination theory. Learn Instruction 21(3):332\u0026ndash;344. ttps://doi.org/10.1016/j.learninstruc.2010.03.002\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeci EL, Ryan RM (1985) Intrinsic motivation and self-determination in human behavior. Plenum. ttps://doi.org/10.1007/978-1-4899-2271-7\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFaul F, Erdfelder E, Lang AG, Buchner A (2007) G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 39(2):175\u0026ndash;191. ttps://doi.org/10.3758/bf03193146\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFidan \u0026Uuml; (2025) Individual Privacy Perception in the Digital Age: The Interaction of Artificial Intelligence Attitude and Dependency. OPUS J Soc Res 22(5):869\u0026ndash;881. ttps://doi.org/10.26466/opusjsr.1725180\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFrenkenberg A, Hochman G (2025) It\u0026rsquo;s Scary to Use It, It\u0026rsquo;s Scary to Refuse It: The Psychological Dimensions of AI Adoption-Anxiety, Motives, and Dependency. Systems 13(2):82. ttps://doi.org/10.3390/systems13020082\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGerlich M (2025) AI tools in society: Impacts on cognitive offloading and the future of critical thinking. Societies 15(1):6. ttps://doi.org/10.3390/soc15010006\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGonz\u0026aacute;lez-Cutre D, Sicilia A (2012) Motivation and exercise dependence: a study based on self-determination theory. Res Q Exerc Sport 83(2):318\u0026ndash;329. ttps://doi.org/10.1080/02701367.2012.10599863\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuay F, Denault A-S, Renauld S (2017) School attachment and relatedness with parents, friends and teachers as predictors of students\u0026rsquo; intrinsic and identified regulation. Contemp Educ Psychol 51:416\u0026ndash;428. ttps://doi.org/10.1016/j.cedpsych.2017.10.001\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHagendorff T (2020) The Ethics of AI Ethics: An Evaluation of Guidelines. Minds Machines 30:99\u0026ndash;120. ttps://doi.org/10.1007/s11023-020-09517-8\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHagger MS, Hardcastle SJ, Chater A, Mallett C, Pal S, Chatzisarantis NL (2014) Autonomous and controlled motivational regulations for multiple health-related behaviors: between- and within-participants analyses. Health Psychol Behav Med 2(1):565\u0026ndash;601. ttps://doi.org/10.1080/21642850.2014.912945\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarding TS, Vanasupa L, Savage RN, Stolk JD (2007) Work-in-progress - Self-directed learning and motivation in a project-based learning environment. In \u003cem\u003e2007 37th Annual Frontiers In Education Conference - Global Engineering: Knowledge Without Borders, Opportunities Without Passports\u003c/em\u003e (pp. F2G-3\u0026ndash;F2G-4). IEEE. ttps://doi.org/10.1109/FIE.2007.4418013\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHe WJ (2023) Positive and negative affect facilitate creativity motivation: Findings on the effects of habitual mood and experimentally induced emotion. Front Psychol 14:1014612. ttps://doi.org/10.3389/fpsyg.2023.1014612\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHsu TC, Huang HL, Hwang GJ, Chen MS (2023) Effects of incorporating an expert decision-making mechanism into chatbots on students\u0026rsquo; achievement, enjoyment, and anxiety. Educational Technol Soc 26(1):218\u0026ndash;231. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.jstor.org/stable/48707978\u003c/span\u003e\u003cspan address=\"https://www.jstor.org/stable/48707978\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang S, Lai X, Ke L, Li Y, Wang H, Zhao X, Dai X, Wang Y (2024) AI Technology panic-is AI Dependence Bad for Mental Health? A Cross-Lagged Panel Model and the Mediating Roles of Motivations for AI Use Among Adolescents. Psychol Res Behav Manage 17:1087\u0026ndash;1102. ttps://doi.org/10.2147/PRBM.S440889\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJia W, Pan L, Neary S (2025) Effect of AI Dependency on University Students' Academic Achievement: The Mediating Role of Self-Efficacy and Moderating Role of Perceived Teacher Caring. Behav Sci 15(10):1348. ttps://doi.org/10.3390/bs15101348\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKalivas PW, Volkow ND (2005) The neural basis of addiction: a pathology of motivation and choice. Am J Psychiatry 162(8):1403\u0026ndash;1413. ttps://doi.org/10.1176/appi.ajp.162.8.1403\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKasneci E, Se\u0026szlig;ler K, K\u0026uuml;chemann S, Bannert M, Dementieva D, Fischer F, Kasneci G (2023) ChatGPT for good? On opportunities and challenges of large language models for education. Learn individual differences 103:102274. ttps://doi.org/10.1016/j.lindif.2023.102274\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKatsantonis A, Katsantonis I (2024) University students\u0026rsquo; attitudes toward artificial intelligence: An exploratory study of the cognitive, emotional, and behavioural dimensions of AI attitudes. Educ Sci 14(9):988. ttps://doi.org/10.3390/educsci14090988\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKaya F, Aydin F, Schepman A, Rodway P, Yetişensoy O, Kaya D, M (2024) The Roles of Personality Traits, AI Anxiety, and Demographic Factors in Attitudes toward Artificial Intelligence. Int J Hum Comput Interact 40(2):497\u0026ndash;514. ttps://doi.org/10.1080/10447318.2022.2151730\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLee H-P, Sarkar A, Tankelevitch L, Drosos I, Rintel S, Banks R, Wilson N (2025) The impact of generative AI on critical thinking: Self-reported reductions in cognitive effort and confidence effects from a survey of knowledge workers. In \u003cem\u003eCHI\u0026rsquo;25: Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems\u003c/em\u003e (p. 1121, pp. 1\u0026ndash;22). ttps://doi.org/10.1145/3706598.3713778\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi J, King RB, Chai CS, Zhai X, Lee VWY (2025a) The AI Motivation Scale (AIMS): a self-determination theory perspective. J Res Technol Educ 1\u0026ndash;22. ttps://doi.org/10.1080/15391523.2025.2478424\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi J, Zhang J, Chai CS, Lee VWY, Zhai X, Wang X, King RB (2025b) Analyzing the network structure of students\u0026rsquo; motivation to learn AI: A self-determination theory perspective. npj Sci Learn 10(48). ttps://doi.org/10.1038/s41539-025-00339-w\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Y, Chiu TKF (2025) The mediating effects of needs satisfaction on the relationship between teacher support and student engagement with generative artificial intelligence (AI) chatbots from a self-determination theory (SDT) perspective. Educational Technol Soc 30:20051\u0026ndash;20070. ttps://doi.org/10.1007/s10639-025-13574-w\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu G, Ma C (2024) Measuring EFL learners\u0026rsquo; use of ChatGPT in informal digital learning of English based on the technology acceptance model. Innov Lang Learn Teach 18(2):125\u0026ndash;138. ttps://doi.org/10.1080/17501229.2023.2240316\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLubke GH, Muth\u0026eacute;n B (2005) Investigating population heterogeneity with factor mixture models. Psychol Methods 10(1):21\u0026ndash;39. ttps://doi.org/10.1037/1082-989X.10.1.21\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMart\u0026iacute;n N\u0026uacute;\u0026ntilde;ez JL, Ar AY, P\u0026eacute;rez Fern\u0026aacute;ndez R, Abbas A, Radovanovic D (2023) Does intrinsic motivation mediate? Perceived artificial intelligence (AI) learning and computational thinking of students during the COVID-19 pandemic. Computers Educ Artif Intell 4:100128. ttps://doi.org/10.1016/j.caeai.2023.100128\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMasgoret A-M, Gardner RC (2003) Attitudes, motivation, and second language learning: A meta-analysis of studies conducted by Gardner and Associates. Lang Learn 53(1):123\u0026ndash;163. ttps://doi.org/10.1111/1467-9922.00212\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMohamed AM, Shaaban TS, Bakry SH, Guill\u0026eacute;n-G\u0026aacute;mez FD, Strzelecki A (2025) Empowering the faculty of education students: Applying AI\u0026rsquo;s potential for motivating and enhancing learning. Innov High Educ 50(2):587\u0026ndash;609. ttps://doi.org/10.1007/s10755-024-09747-z\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorales-Garc\u0026iacute;a WC, Sairitupa-Sanchez LZ, Morales-Garc\u0026iacute;a SB, Morales-Garc\u0026iacute;a M (2024) Development and validation of a scale for dependence on artificial intelligence in university students. Front Educ 9:1323898. ttps://doi.org/10.3389/feduc.2024.1323898\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorley J, Floridi L, Kinsey L, Elhalal A (2020) From What to How: An Initial Review of Publicly Available AI Ethics Tools, Methods and Research to Translate Principles into Practices. Sci Eng Ethics 26(4):2141\u0026ndash;2168. ttps://doi.org/10.1007/s11948-019-00165-5\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoussawi S, Koufaris M (2019) Perceived intelligence and perceived anthropomorphism of personal intelligent agents: Scale development and validation. In \u003cem\u003eHawaii International Conference on System Sciences.\u003c/em\u003e ttps://doi.org/10.24251/HICSS.2019.015\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNamestovski Ž, Major L, Moln\u0026aacute;r G, Szűts Z, Esztelecki P, Kőr\u0026ouml;si G (2018) External motivation, the key to success in the MOOCs framework. Acta Polytech Hungarica 15(6):125\u0026ndash;142. ttps://doi.org/10.12700/APH.15.6.2018.6.7\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNawaz N, Arunachalam H, Pathi BK, Gajenderan V (2024) The adoption of artificial intelligence in human resources management practices. Int J Inform Manage Data Insights 4(1):100208. ttps://doi.org/10.1016/j.jjimei.2023.100208\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNylund KL, Asparouhov T, Muth\u0026eacute;n BO (2007) Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Struct Equ Model 14(4):535\u0026ndash;569. ttps://doi.org/10.1080/10705510701575396\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOroujlou N, Vahedi M (2011) Motivation, attitude, and language learning. Procedia-Social Behav Sci 29:994\u0026ndash;1000. ttps://doi.org/10.1016/j.sbspro.2011.11.333\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePham T (2021) Attitude and motivation in language learning: A review. J Engl Lang Teach Appl Linguistics 3(5):64\u0026ndash;72. ttps://doi.org/10.32996/jeltal.2021.3.5.7\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRyan RM, Deci EL (2000) Self-Determination Theory and the Facilitation of Intrinsic Motivation, Social Development, and Well-Being. Am Psychol 55:68\u0026ndash;78. ttps://doi.org/10.1037/0003-066x.55.1.68\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSalah M, Abdelfattah F, Alhalbusi H et al (2024) \u003cem\u003eMe and my AI bot: Exploring the 'AIholic' phenomenon and university students' dependency on generative AI chatbots - Is this the new academic addiction?\u003c/em\u003e [Preprint]. Research Square. ttps://doi.org/10.21203/rs.3.rs-3508563/v2\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaif N, Khan SU, Shaheen I, Alotaibi FA, Alnfiai MM, Arif M (2024) Chat-GPT; validating Technology Acceptance Model (TAM) in education sector via ubiquitous learning mechanism. Comput Hum Behav 154:108097. ttps://doi.org/10.1016/j.chb.2023.108097\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchepman A, Rodway P (2020) Initial validation of the general attitudes towards Artificial Intelligence Scale. Computers Hum Behav Rep 1:100014. ttps://doi.org/10.1016/j.chbr.2020.100014\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShen X, Mo X, Xia T (2025) Exploring the attitude and use of AI-image among art and design college students based on TAM and SDT. Interact Learn Environ 33(2):1198\u0026ndash;1215. ttps://doi.org/10.1080/10494820.2024.2365959\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStrzelecki A (2024) Students\u0026rsquo; Acceptance of ChatGPT in Higher Education: An Extended Unified Theory of Acceptance and Use of Technology. Innov High Educ 49:223\u0026ndash;245. ttps://doi.org/10.1007/s10755-023-09686-1\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuh W, Ahn S (2022) Development and validation of a scale measuring student attitudes toward artificial intelligence. SAGE Open 12(2):21582440221100463. ttps://doi.org/10.1177/21582440221100463\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun S, Rubin AM, Haridakis PM (2008) The Role of Motivation and Media Involvement in Explaining Internet Dependency. J Broadcast Electron Media 52(3):408\u0026ndash;431. ttps://doi.org/10.1080/08838150802205595\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTiwari CK, Bhat MA, Khan ST, Subramaniam R, Khan MAI (2024) What drives students toward ChatGPT? An investigation of the factors influencing adoption and usage of ChatGPT. Interact Technol Smart Educ 21(3):333\u0026ndash;355. ttps://doi.org/10.1108/ITSE-04-2023-0061\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVansteenkiste M, Smeets S, Soenens B, Lens W, Matos L, Deci EL (2010) Autonomous and controlled regulation of performance-approach goals: Their relations to perfectionism and educational outcomes. Motivation Emot 34(4):333\u0026ndash;353. ttps://doi.org/10.1007/s11031-010-9188-3\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang CY, Lin JJ (2023) Utilizing artificial intelligence to support analyzing self-regulated learning: A preliminary mixed-methods evaluation from a human-centered perspective. Comput Hum Behav 144:107721. ttps://doi.org/10.1016/j.chb.2023.107721\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang H, Fu T, Du Y et al (2023) Scientific discovery in the age of artificial intelligence. Nature 620:47\u0026ndash;60. ttps://doi.org/10.1038/s41586-023-06221-2\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXia Q, Chiu TKF, Lee M, Sanusi IT, Dai Y, Chai CS (2022) A self-determination theory (SDT) design approach for inclusive and diverse artificial intelligence (AI) education. Comput Educ 189:104582. ttps://doi.org/10.1016/j.compedu.2022.104582\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu W, Gao Z (2025) An Intelligent Sociotechnical Systems (iSTS) Framework: Enabling a Hierarchical Human-Centered AI (hHCAI) Approach. IEEE Trans Technol Soc 6(1):31\u0026ndash;46. ttps://doi.org/10.1109/TTS.2024.3486254\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang W, Xie Y (2024) Can robots elicit empathy? The effects of social robots\u0026rsquo; appearance on emotional contagion. Computers Hum Behavior: Artif Hum 2(1):100049. ttps://doi.org/10.1016/j.chbah.2024.100049\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYe JH, Zhang M, Nong W et al (2025) The relationship between inert thinking and ChatGPT dependence: An I-PACE model perspective. Educational Technol Soc 30:3885\u0026ndash;3909. ttps://doi.org/10.1007/s10639-024-12966-8\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhai C, Wibowo S, Li LD (2024) The effects of over-reliance on AI dialogue systems on students\u0026rsquo; cognitive abilities: a systematic review. Smart Learn Environ 11:28. ttps://doi.org/10.1186/s40561-024-00316-7\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang KZK, Chen C, Lee MKO (2014) Understanding the role of motives in smartphone addiction. In \u003cem\u003ePACIS 2014 Proceedings\u003c/em\u003e (p. 131). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://aisel.aisnet.org/pacis2014/131\u003c/span\u003e\u003cspan address=\"https://aisel.aisnet.org/pacis2014/131\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang S, Zhao J, Tan W (2008) Extending TAM for online learning systems: An intrinsic motivation perspective. Tsinghua Sci Technol 13(3):312\u0026ndash;317. ttps://doi.org/10.1016/S1007-0214(08)70050-6\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang S, Zhao X, Zhou T, Kim JH (2024) Do you have AI dependency? The roles of academic self-efficacy, academic stress, and performance expectations on problematic AI usage behavior. Int J Educational Technol High Educ 21:34. ttps://doi.org/10.1186/s41239-024-00467-0\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Y, Ding Y, Huang H, Peng Q, Wan X, Lu G, Chen C (2022) Relationship between insecure attachment and mobile phone addiction: A meta-analysis. Addict Behav 131:107317. ttps://doi.org/10.1016/j.addbeh.2022.107317\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhong W, Luo J, Lyu Y (2024) How Do Personal Attributes Shape AI Dependency in Chinese Higher Education Context? Insights from Needs Frustration Perspective. PLoS ONE 19(11):e0313314. ttps://doi.org/10.1371/journal.pone.0313314\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"humanities-and-social-sciences-communications","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"palcomms","sideBox":"Learn more about [Humanities \u0026 Social Sciences Communications](http://www.nature.com/palcomms/)","snPcode":"41599","submissionUrl":"https://submission.springernature.com/new-submission/41599/3","title":"Humanities and Social Sciences Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"AI attitude, AI dependence, AI use motivation, Self-determination theory(SDT),, generative artificial intelligence (AI)","lastPublishedDoi":"10.21203/rs.3.rs-8954144/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8954144/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eWhether students develop dependence on generative artificial intelligence (AI) technologies largely depends on their attitudes toward using such tools. Existing studies have primarily explained the formation of AI dependence from the perspectives of usage attitudes and technological characteristics, while paying limited attention to the psychological mechanisms through which attitudes exert their influence. Grounded in Self-Determination Theory, the present study examines how different types of use motivation mediate the relationship between students\u0026rsquo; AI attitudes and AI dependence. A total of 405 university students completed a comprehensive questionnaire measuring AI attitudes, use motivations, and AI dependence. The results indicate that: (i) AI attitudes do not directly predict AI dependence; instead, their effects are transmitted through multiple motivational pathways, with intrinsic motivation inhibiting AI dependence, whereas identified regulation facilitates its development; (ii) three distinct AI dependence profiles were identified: dependent, moderate, and conservative users. These profiles exhibited clear stratification, with dependent users showing the highest scores across all dimensions and conservative users the lowest; and (iii) individuals with stronger motivations are more likely to be classified into the dependent user profile, with external motivation playing a particularly salient role. These findings enrich the literature on AI technology dependence and offer practical implications for higher education administrators, tool developers, and educational researchers.\u003c/p\u003e","manuscriptTitle":"From Appropriate Use to Dependence: How AI Attitudes and Use Motivations Shape University Students’ AI Dependence","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-23 19:00:18","doi":"10.21203/rs.3.rs-8954144/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-17T23:39:23+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-14T18:48:01+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-07T13:48:04+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"270173461708271407004915924002999630063","date":"2026-03-23T05:19:46+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-20T09:53:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"130628070252852279239059333166439123662","date":"2026-03-20T08:14:13+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"128963274715666742176814306971228479477","date":"2026-03-19T22:58:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"17963843219731832083221559668317802399","date":"2026-03-19T10:26:30+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-18T15:31:16+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-18T14:15:18+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-03-18T14:03:09+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-12T07:38:09+00:00","index":"","fulltext":""},{"type":"submitted","content":"Humanities and Social Sciences Communications","date":"2026-03-12T03:17:59+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"humanities-and-social-sciences-communications","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"palcomms","sideBox":"Learn more about [Humanities \u0026 Social Sciences Communications](http://www.nature.com/palcomms/)","snPcode":"41599","submissionUrl":"https://submission.springernature.com/new-submission/41599/3","title":"Humanities and Social Sciences Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"667d50f4-73f6-47d6-97b6-e15b23e05c88","owner":[],"postedDate":"March 23rd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":64844874,"name":"Business and commerce/Information systems and information technology"},{"id":64844875,"name":"Biological sciences/Psychology"},{"id":64844876,"name":"Social science/Psychology"},{"id":64844877,"name":"Social science/Science technology and society"}],"tags":[],"updatedAt":"2026-05-02T20:08:11+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-23 19:00:18","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8954144","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8954144","identity":"rs-8954144","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}