Abstract: The popularity of smartphones is not only changing people's lives, but also their psychological perceptions. Recent studies have adapted the self-expansion theory to understand the impact of the smartphone on individuals' self, namely self-expansion via smartphone (SES), which is interpreted as expanding personal abilities and self-concepts through smartphones (Hoffner et al., 2016; Wu et al., 2023). Most of the existing literature discussed the impacts of people's use of services or applications on self-expansion from the view of smartphones as a tool. However, the intelligent and interactive design of smartphones remind us that smartphones are not only tools to be manipulated by people. They can also actively initiate interaction with people and meet their emotional needs. This might lead people to view their smartphones not as tools but as self-aware beings, and the resultant new human-computer relationship could also expand individuals' self. Hence, the current study examines the causes and mechanisms of SES from a relational perspective. Specifically, we investigate whether there is a significant difference in the level of “self-expansion via smartphone” and “the inclusion of smartphones in the self (ISS)” between people with different levels of smartphone anthropomorphism, and a moderated mediation model is constructed to test the mediating role of ISS between smartphone anthropomorphism and SES, as well as the moderating role of perceived threat in such a mediation model. Moreover, individual differences in anthropomorphism, smartphone addiction tendency, gender, and age were included in the analysis as covariates.
In the first study, we developed an 8-item smartphone perceived threat questionnaire. A total of 347 college students were surveyed through the questionnaire method to explore the relations among smartphone anthropomorphism, SES, ISS, and perceived threat. SPSS macro PROCESS was used to test our hypothesis. In the second study, a situational experiment with 157 college students (9 participants who did not answer seriously or guessed the experiment purpose were excluded) was carried out in the 2 (smartphone anthropomorphism: anthropomorphic vs. nonanthropomorphic) × 2 (threat: threat-priming vs. control) group design. Participants were informed that they would be interviewed about smartphones. Firstly, participants were randomly assigned to either a threat-priming or control condition. In the threat-priming condition, participants read a popular science article about the potentially harmful effects of smartphones. Participants in the control condition read about the history of smartphones. Next, participants were randomly assigned to one of two smartphone anthropomorphism conditions, operationalized by displaying an advertising video and asking several questions about smartphones. Specifically, in the anthropomorphic condition, the smartphone incarnates a human in the advertising video, and questions were written with lifelike, agentic phrasing. In the nonanthropomorphic condition, the advertising video showed smartphones' capabilities, and questions were written with relatively neutral phrasing. Finally, participants responded to a series of questionnaires on the dependent, mediated, and control variables. After analysis of variance, we used a moderated mediation model to test whether the relationship between smartphone anthropomorphism (1 = anthropomorphic, 0 = non-anthropomorphic) and SES was mediated by ISS, and whether the indirect effect differed by threat condition (1 = threat priming, 0 = control; model 8 in Hayes, 2013).
The results of Study 2 found that, compared to participants in the non-anthropomorphic condition, participants who were induced smartphone anthropomorphism mindset had a higher level of ISS (F(1, 140) = 4. 47, p < .05, η² = .03) and SES (F(1, 140) =6. 81, p < .05, η² = .035), Perceived threat moderated the relation between smartphone anthropomorphism and ISS. Smartphone anthropomorphism can increase participants' ISS when the perceived threat was not primed (F(1, 140)=9. 70, p < .01, η²= .07), while this effect did not exist when the perceived threat was primed. The moderated mediation analysis of two studies consistently showed that the relation between smartphone anthropomorphism and SES was fully mediated by ISS. Moreover, a boundary condition existed. That is, perceived threat limited the ability of anthropomorphism to facilitate SES from the pe. Specifically, the correlation between smartphone anthropomorphism and ISS, as well as the indirect effect of ISS, both became weaker as the perceived threat increased.
In conclusion, the present study reveals how smartphone anthropomorphism affects self-expansion via smartphone and its underlying mechanism from the perspective of relation. The findings contribute to extending research on smartphone self-expansion and provide new insights into how smartphones serve as an affective technology that helps people develop their identities and selves.

Key words: smartphone, anthropomorphism, self-expansion, perceived threat.

摘要： 基于自我扩展理论,从关系视角探索手机自我扩展的成因及其机制。研究1通过问卷法构建有调节的中介模型,初步探索拟人化、手机纳入自我、手机自我扩展间的关系及威胁感在其中的调节作用,研究2采用2×2的组间实验对假设模型进行进一步验证。结果一致发现：（1）手机纳入自我在拟人化与手机自我扩展之间起完全中介作用;（2）威胁感调节了这一中介效应,当个体感知的手机威胁感高时,拟人化对手机纳入自我的作用、手机纳入自我的中介效应均减小。上述结果表明手机拟人化可以实现关系视角下的手机自我扩展,威胁感是这一过程发生的边界条件。

关键词: 智能手机, 拟人化, 自我扩展理论, 威胁感