College of Forestry, Fujian Agriculture and Forestry University, Fuzhou 350002, China. Key Laboratory of National Forestry and Grassland Administration for Orchid Conservation and Utilization at College of Landscape Architecture, Fujian Agriculture and Forestry University, Fuzhou 350002, China
Auto-generated profile, based on publicly available information
Zhongjian Liu
College of Forestry, Fujian Agriculture and Forestry University, Fuzhou 350002, China. Key Laboratory of National Forestry and Grassland Administration for Orchid Conservation and Utilization at College of Landscape Architecture, Fujian Agriculture and Forestry University, Fuzhou 350002, China
The purpose of this study was to examine the antecedents of academic emotions among university students through real-time and multilevel analyses. We explored the link between state self-control and academic emotions, the influence of self-determination on the relationship between them, and the moderating role of trait self-control. The data was collected over seven consecutive days during which university students (N = 155) completed smartphone questionnaires. Data was organized into hierarchical two-level structures, where situations (Level 1) nested within individuals (Level 2). The results showed that positive emotions were negatively predicted by state self-control while negative emotions were positively predicted by state self-control. Moreover, state self-control under low self-determination was a significant negative predictor of positive emotions, whereas high self-determination had no predictive effect on positive emotions. The relationship between self-determination and negative emotions was further moderated by trait self-control. The limitations of this study and future research directions are also discussed.
6 February 2025 in Journal of Computer Assisted Learning
As generative artificial intelligence becomes increasingly embedded in everyday, educational, and work-related contexts, increasing attention has been directed to dependence-related forms of AI chatbot use. Problematic smartphone use (PSU) may reflect broader dysregulated digital engagement and may therefore be relevant to understanding dependence-related AI chatbot use. However, little is known about whether PSU is associated with AI chatbot dependence or whether AI-related evaluative factors show different association patterns across PSU severity profiles. Guided by the I-PACE model, data from 538 AI chatbot users were analyzed using latent profile analysis and a supplementary continuous-risk model. Latent profile analysis yielded three severity-based PSU groups: low-symptom PSU (16%), moderate-symptom PSU (46%), and high-symptom PSU (38%). These groups formed an ordered severity gradient rather than distinct symptom configurations, and AI chatbot dependence differed across them. Emotional experience was implicated in several significant statistical indirect associations between AI-use evaluations and AI chatbot dependence. Profile-specific analyses suggested possible differences across PSU groups; however, formal between-group evidence was limited to the performance expectancy pathway. The supplementary continuous-risk model supported the positive association between PSU and AI chatbot dependence but did not reproduce the profile-specific performance expectancy finding. These findings suggest that dependence-related AI chatbot use may be situated within broader patterns of dysregulated digital engagement, although the observed profile differences primarily reflected severity rather than distinct typological patterns. Given the cross-sectional self-report design, the findings should be interpreted cautiously and require longitudinal, multi-method, and more diverse-sample replication.
As generative artificial intelligence becomes increasingly embedded in everyday and educational contexts, scholarly concern has shifted from initial acceptance of AI chatbots to the risk that sustained use may evolve into dependence. However, little is known about whether individuals with different levels of problematic smartphone use (PSU) exhibit distinct pathways toward AI chatbot dependence. Based on the AI device use acceptance (AIDUA) framework, this study surveyed 538 participants recruited through an online platform to examine whether latent PSU profiles differ in their levels of AI chatbot dependence and in the cognitive-affective pathways associated with such dependence. The results indicate that: (1) Latent profile analysis identified three PSU profiles that primarily reflected symptom severity: low-symptom PSU (16%), moderate-symptom PSU (46%), and high-symptom PSU (38%); significant differences in AI chatbot dependence levels were observed across these PSU subcategories; (2) Emotional experience mediated the associations of hedonic motivation, performance expectancy, and effort expectancy with AI chatbot dependence; (3) Within-group mediation patterns varied across PSU profiles. In the moderate-PSU group, emotional experience emerged as a relatively salient within-group mechanism, whereas in the high-PSU group, performance expectancy played a more central role. Formal between-group comparisons further indicated that cross-group heterogeneity was primarily concentrated in the performance expectancy pathway. These findings extend research on AI chatbot use beyond acceptance processes to dependence outcomes, clarify heterogeneity across individuals in the link between PSU and AI chatbot dependence, and provide a basis for more differentiated risk assessment and intervention strategies.
Although behavioural research has extensively documented anthropomorphism in human-robot interaction, the neural basis of perceiving anthropomorphic robots remains inconsistent across individual fMRI studies. To address this gap, we conducted a coordinate-based meta-analysis of 19 fMRI studies (430 participants) using activation likelihood estimation (ALE), followed by meta-analytic connectivity modelling (MACM) and functional decoding via Neurosynth. The ALE meta-analysis identified two clusters that were consistently activated across studies: the posterior cingulate cortex/precuneus (PCC/precuneus) and the right temporoparietal junction/inferior parietal lobule (rTPJ/IPL). MACM further revealed that these two nodes, together with the medial prefrontal cortex, anterior cingulate cortex, and bilateral TPJ, form a stable mentalizing network that supports both self-referential processing and mental state attribution, as confirmed by Neurosynth decoding. Exploratory subgroup analysis showed that active interaction tasks predominantly engaged the rTPJ, whereas explicit evaluation tasks mainly recruited the right posterior temporal lobe, indicating a task-dependent dissociation. These findings provide quantitative neural evidence for the Computers as Social Actors theory and reconcile previous inconsistencies by revealing a core mentalizing network for anthropomorphic robot perception, along with a dual-pathway mechanism modulated by task type. The results have practical implications for designing socially interactive robots that effectively engage human social cognitive networks.