It is found that users experience empathy and anthropomorphism as a unified"Humanlikeness"construct, and that Privacy, Personalization, and Humanlikeness drive Trust while Perceived Bias degrades it.
Abstract
Trust in AI for emotional support is not universal; it is shaped by who users are, where they come from, and what they value. Yet research in this area lacks validated psychometric instruments for assessing user perceptions in affective AI contexts and large-scale evidence on how trust formation varies across user segments. To address these gaps, we develop and validate a seven-construct psychometric scale, test a Structural Equation Model (SEM) linking system attributes to Trust and Perceived Benefits as mediators of Actual System Use, and conduct a Multi-Group Analysis (MGA) across five sociodemographic dimensions (gender, age, education, socioeconomic status, cross-national region), drawing on 1,343 active users from seven countries. We find that users experience empathy and anthropomorphism as a unified"Humanlikeness"construct, and that Privacy, Personalization, and Humanlikeness drive Trust while Perceived Bias degrades it. Notably, adoption logic diverges across groups: Privacy shapes women's trust more than men's, Anglosphere (UK, USA) users respond more positively to Humanlikeness than Europeans, and educated and higher-income users require Trust to engage, whereas older adults and lower socioeconomic groups bypass it entirely, relying on perceived practical benefits (e.g., 24/7 availability, non-judgmental support). Our findings extend technology acceptance theory and inform the equitable design of emotional support AI.
This study examines how influencer characteristics and short-form video content characteristics shape consumers’ imitation intention toward influencers in short-form live commerce. Drawing on the stimulus–organism–response (S-O-R) framework and social learning theory, this study conceptualizes cognitive trust and affective trust as two internal psychological states linking external stimuli to imitation intention. Imitation intention is defined as consumers’ willingness to regard influencers as behavioral reference models and to adopt or follow their consumption choices, suggestions, and recommendations. Survey data were collected from 306 Chinese consumers of TikTok, Kuaishou, and Xiaohongshu and analyzed using partial least squares structural equation modeling (PLS-SEM). The results show that ethics, homophily, authenticity, and entertainment significantly enhance both cognitive and affective trust, whereas originality significantly increases only cognitive trust and social attractiveness has no significant effect on either trust dimension. Cognitive trust also positively influences affective trust. Both trust dimensions significantly increase consumers’ imitation intention, with affective trust exerting the stronger effect. The mediation analysis further reveals both differentiated parallel and sequential trust pathways: cognitive trust mediates five of the six antecedent relationships, affective trust mediates four, and sequential mediation through cognitive trust and affective trust is significant for ethics, homophily, authenticity, originality, and entertainment. These findings show that trust operates through both distinct and interrelated mechanisms in translating influencer- and content-related cues into imitation intention.
Rapid social transformation in China has reshaped everyday social interaction as mobility increases and encounters with unfamiliar others become more common, raising questions about how social trust influences subjective well-being. Existing research typically treats trust either as an individual disposition or as a contextual social resource, leaving unclear how changes in both individual and community trust relate to well-being. Using panel data from the China Family Panel Studies (2012–2020) and individual fixed-effects models, this study examines how changes in generalized and domain-specific trust relate to life satisfaction across urban and rural China. Results show that increases in individual generalized trust are consistently associated with higher life satisfaction, while the association for community-level generalized trust is more sensitive to contextual setting and model specification. At the individual level, trust in embedded relations and institutional actors is positively associated with life satisfaction, whereas trust in socially distant others exhibits less consistent and more context-dependent associations. At the community level, higher neighborhood trust was positively associated with life satisfaction, particularly in rural areas, whereas stranger trust tended to exhibit negative associations with life satisfaction, especially within rural samples and models accounting for other trust domains. The findings suggest that the well-being implications of trust vary systematically across relational domains and community contexts, indicating that different forms of trust may carry distinct well-being implications rather than uniformly functioning as supportive social resources in rapidly changing societies.
Mengdie Ruan, M. N. Asadullah, Elya Nabila Abdul Bahri et al.· Social Indicators Research· 0 citations
The findings suggest that trustworthy AI integration requires addressing pre-existing structural failures before introducing algorithmic decision support, and offer design recommendations for AI-supported appraisal systems, with implications for HCI research and organizational practice.
Andreas Harnindito, Akshat Sawarni· Proceedings of the 2026 Euro...· 0 citations
Trust is an important psychological and social resource for interpersonal cooperation and societal functioning. However, measures commonly used to assess trust may capture related yet non-interchangeable constructs. Using an exploratory cross-sectional design, this study examined the response patterns, convergence and divergence, and selected demographic and regional correlates of multiple trust-related indicators in a large online sample of predominantly young Chinese adults. Participants were recruited through university networks and the questionnaire Star platform. After quality control, 5155 valid responses were included. The measures comprised a classic single-item generalized trust question, scales assessing general trust attitudes and dispositional trust, a self-reported measure of trust tendencies toward different social targets, and a hypothetical, non-incentivized trust-game scenario assessing investment choices, reciprocity expectations, and stated return decisions. In the classic trust question, 56.55% of participants indicated that most people in society could be trusted. Dispositional trust and general trust attitudes were above their respective scale midpoints with large effect sizes (Cohens d = 0.95 and 1.26), whereas self-reported interpersonal trust tendencies showed a small negative deviation from the corresponding midpoint (d = -0.17). All three hypothetical trust-game responses were statistically below the 50% response-range midpoint, although the deviations were trivial for investment choices (d = -0.09), modest for reciprocity expectations (d = -0.34), and small for stated return decisions (d = -0.15). General trust attitudes were strongly associated with dispositional trust (r = 0.853), and self-reported interpersonal trust tendencies were moderately associated with both measures. In contrast, generalized trust belief showed negligible associations with the hypothetical task responses (rs = 0.027-0.061). Regional indicators provided only small increments in explanatory power after selected demographic characteristics were controlled, and their associations varied across outcomes. The findings indicate that trust-related measures show partial convergence while capturing construct- and context-specific dimensions. Conclusions about general interpersonal trust therefore depend on the construct assessed, the target of trust, and the response context. These findings should be interpreted as measurement patterns within a predominantly young Chinese online sample rather than as a population estimate of trust in China.
Hongli Yan, Chao Fu, Xiang Li et al.· Acta Psychologica· 0 citations
Despite the rapid rise of AI‐driven virtual influencers in global marketing, there is limited understanding of how trust in these nonhuman agents impacts consumer well‐being across diverse cultural settings. This study addresses this gap by establishing trust in AI influencers as the central driver of consumer well‐being and examining how this relationship is shaped by core psychological mechanisms and moderated by cultural context. A mixed method design was adopted, combining a fuzzy Delphi expert study and in‐depth consumer interviews with a cross‐national survey of 3157 participants from the USA, UK, Canada and Australia. Measurement invariance, partial least squares structural equation modelling and multigroup analyses were applied to test the direct, indirect and moderated pathways, with individual‐level cultural orientation examined within and across these national contexts. The results confirm that trust exerts both direct and indirect effects on consumer well‐being through authenticity, parasocial bonds and persuasion knowledge model. Cultural orientation and digital literacy significantly moderate these effects, whereas persona realism shows an uncanny valley pattern. Trust emerged as the strongest predictor of well‐being, explaining nearly half of the variance across contexts. Marketers can build sustainable trust in AI influencers by prioritising transparent disclosure, coherent storytelling and ethical personalisation. Additionally, cultural sensitivity and audience literacy levels must be considered when designing effective AI influencer strategies. Regulators can use these insights to establish ethical guidelines that safeguard consumer autonomy. This research empirically assesses a sociotechnical model of trust in AI influencers across multiple countries. By integrating psychological and cultural mechanisms, it extends persuasion and authenticity theories into the nonhuman domain, offering both theoretical advances and actionable insights for global marketing practice.
Arjun J. Nair, S. Manohar· Journal of Consumer Behaviou...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
