The Impact of Augmented Reality Through User-Platform Interactions Towards Continuance Intention with the Effect of User Generation

Abstract

When users interact with mobile platforms in an Augmented Reality (AR) environment, cognitive and emotional engagements change through different stimuli cues that respond to users’ behavioral intentions. Although AR engages more interactions in mobile platforms, there is a significant gap in assessing UX, considering the physical distance between users and virtual products in a mobile platform. Considering the effect of user generation, the impacts of perceived engagements toward continuance intention through user-platform interactions are unexplored. This study investigated a nuanced understanding of how stimuli cues in augmented reality affect sense of immersion and sense of presence, followed by an Interaction-Engagement-Intention (I-E-I) model. A quantitative method was used to validate the proposed model. Based on an online survey with 886 responses, product fit, network quality, and Artificial Intelligence-driven Recommendation (AIR) influences were assessed for cognitive engagements. This study examined the importance of engaging satisfaction and trust as emotional engagements, influencing users’ continuance intention. The findings showed that sense of presence has a more significant influence on building trust and satisfaction. Also, trust has a more significant impact on the continuance intention to use AR mobile platforms. This study also explored the positive effects of user generation on continuance intention. This could enhance the capabilities of information system designers, researchers, marketing professionals, and solution providers to attain sustainable user retention.

1. Introduction

Augmented Reality (AR) revolutionizes the mobile platform industry through interactive engagements. AR enhances User Experience (UX) through correlated actions among cognitive, physical, and virtual encounters to collaborate, cooperate, and communicate in the physical world [1]. This requirement has increased due to rapid changes in user expectations for Heterogeneous Human-computer Interaction (HCI). The effect of those interactions involves users’ engagements with AR applications that demand to assess users’ decision-making through continuance intention [2]. Previous literature has shown that exploring the effects of interactive engagements in perceiving UX for AR in the user-platform interaction paradigm is unrealized [3].

This study addressed a fundamental research interest to explain the consequences of user-platform interaction through product fit [4], network quality [5], and artificial intelligence-driven recommendation [6]. The study investigated how AR intrinsically motivated users’ cognitive engagement to perceive user experience [7] and explained a process of assessing users’ emotions through satisfaction and trust as subsequent effects of engagement. Finally, it explained how user engagement motivates user experience through cognitive and emotional values and understanding to facilitate the decision-making process on continuance intention [8].

In sociocultural aspects, this study described age group or generation as essential factors influencing users’ choices, values, and behavioral responses to interact with AR mobile platforms. This study considered generation as joint group of age users with equal shared beliefs, expectancy, dependency, attitudes, and responses to identify user groups within Australia for assessing their continuance intention to use AR mobile platforms [9]. Furthermore, this study determined the generation of users through age groups to identify user classifications with distinct sociocultural norms and their effects on continuance intention [10]. Moreover, the acculturated culture was addressed to explain the behavioral change of user generation who migrated from native to host culture in a multicultural society. This study explored the nuances of understanding the effects of generation on users’ continuance intention considering the multicultural society in Australia [11].

This study aims to deliver a research framework to explain how stimuli cues affect user experience through the perceived sense of immersion and a sense of presence that influences continuance intention. An online survey (N = 886) was conducted with an AR mobile platform (IKEA app version 4.5.0) to validate the interaction-engagement-intention (I-E-I) model. Then, a Partial Least Square (PLS)—structural equation model (SEM) approach was applied to analyze the data guided by the following research questions: RQ1: What are the key factors determining users’ continuance intention in the user-platform interaction paradigm? RQ2: What influences cognitive and emotional engagement on continuance intention? RQ3: What are the intervening effects of user generation towards continuance intention?

This study reveals that user-platform interactions through AR influence users’ continuance intention, encourage the user-retention process, and support the mobile platform industry in understanding how to assess user experience for user-platform interactions with an AR environment. It contributes to identifying the cognitive and emotional impact in perceiving user experience that influences continuance intention to use AR mobile platforms. Additionally, insights from the potential effects of generation on user behavior suggest that online retailers can plan to expand their AR platforms globally and should consider nuances in cultural values and generation gaps in their strategies. Moreover, research outcomes and valuable insights could guide UX researchers, developers, business promoters, marketing professionals, and solution providers to understand how sociocultural factors affect users on technology adoption to continue using technological platforms.

This paper is organized with a background section reviewing literature regarding augmented reality, sources of information, sense of immersion, sense of presence, user experience, and behavioral responses. It proceeds with methodology, where a research design, measurement instrument, and data-collection procedure are developed. Then, we proceed with the results section. Finally, the study discusses implications and limitations, followed by a conclusion.

2. Literature Review

In the e-commerce revolution, user-platform interactions engage interaction-based stimuli cues. Emerging technologies like AR have revolutionized the interaction capabilities for mobile platforms [12], where users can create a virtual identity to communicate, interact, and perform activities with mobile platforms and engage in a virtual environment through user-platform interaction [2]. Therefore, influences of cognitive [13] and emotional engagements [14] through user-platform interactions may be investigated for assessing user experience (UX) in an AR environment.

Previous studies explained that users interact with virtual products through spatial registration and gesture-based interaction by changing the orientation of physical spaces in an AR environment [15]. AR provides product fit through personalization, scale measurement, positioning, and virtual try-on features in the online platform industry. Conversely, network qualities were addressed as crucial to maintaining uninterrupted immersive services using mobile platforms [5]. Further, network quality is vital to maintaining the platform’s ability to provide uninterrupted services. In another context, artificial intelligence (AI) allows more interactions between users and platforms in an AR environment that may generate recommendations [16] and persuade cognitive engagements [4] to provide user comfort. Therefore, there is a knowledge gap in investigating the effects of stimuli cues, particularly in an AR environment.

Studies have shown that AR provides real-time interactions by placing computer-generated products in physical spaces [17], while virtual products can be viewed on the mobile platform. AR allows users to engage with sensory information through a user-platform interaction paradigm, which seems to be how users are sensing close to the real world. Considering user-platform interactions in an AR mobile application, cognitive engagement may represent users’ perceptions of products or services [18]. Previous studies have examined how users’ expectations are fulfilled through AR interactions that create a sense of immersion through users’ psychological responses [17]. In another context, the situated cognition theory explains that a sense of presence occurs when a user interacts with virtual products and perceives situated feeling as a unique psychological state of “being there” in the virtual world [19]. However, understanding the impacts of stimuli cues toward sense of immersion and sense of presence through user-platform interactions [3] in an AR environment may be explored.

A user’s emotional engagement is highly impacted by cognitive states, which may subsequently influence behavioral intentions [14]. Previous literature explains satisfaction [20,21] and trust [22] as emotional feelings, where the effects of cognitive states toward users’ emotional engagements [14] in an AR environment are still unrealized. Satisfaction develops a pleasant mode and trust as a mental state where a user confirms a willingness to perform actions after fulfilling expectations. Moreover, previous studies described trust and satisfaction as emotional factors influencing users’ behavioral responses through Word-of-Mouth (WOM) and repurchase intentions [23].

The term “continuance intention” was adapted from the Expectation-Confirmation Theory (ECT) in the field of consumer behavior literature [24]. Recently, researchers studied the effects of AR on continuance intention to use mobile platforms. Further, Hung and Rust [5] describe the continuance intention through cognitive states for AR mobile platforms. However, AR features may alleviate product user uncertainty and the need to touch the product [25] through user-platform interactions in a virtual environment where the effects of these critical AR characteristics on continuance intention through perceived UX can be investigated.

Previous studies suggest that user generation may impact users’ behavioral intention [26] to use AR mobile platforms. However, Cabanillas and Santos [10] categorize users with different age groups to explain users’ behavioral responses to adopting e-commerce platforms, where the effects of generation on users’ continuance intention in an AR environment are still unrealized. Generational engagements are vital in determining users’ behaviors, especially when adopting technological systems. The age group (e.g., Generation Z, Millennials, Generation X, Boomers) shows diverse behavioral intentions, societal engagements, influence mechanisms, and adoption tendencies in the technological adoption process [27,28]. In the immersive era, generational gaps may be investigated to assess the effects of generation on users’ continuance of intention to use AR mobile platforms.

3. Hypotheses Development and Research Model

AR involves product movement through tracking and engages a sense of immersion in an immersive environment [20]. In the context of AR applications, Kowalczuk explained the effects of network qualities in engaging a sense of immersion and feeling absorbed in the AR environment [29], while AR involves users with a high level of influence when placing virtual products in physical spaces. Moreover, online mobile platforms use feeling Artificial Intelligence (AI) through adaptive and personalized recommendation systems to extend predictive data analytics [5,16]. Therefore, the following hypotheses are proposed:

H1. 

Product fit positively influences the sense of immersion in user-platform interactions.

H2. 

Network quality positively influences the sense of immersion in user-platform interactions.

H3. 

AI-driven recommendation positively influences the sense of immersion in user-platform interactions.

In AR, Hilken defines situated experience as a sense of presence in an augmented reality environment [30]. Furthermore, AR features provide users comfort in viewing virtual products in their physical spaces, which alleviates product uncertainties and reduces the need for touch [31]. On the other hand, AI transforms wireless networks through network optimization with network parameters, enhancing the UX by adopting seamless convergence of virtual and physical realms that engage the sense of presence [32]. Therefore, we propose the following hypotheses:

H4. 

Product fit positively influences the sense of presence in user-platform interactions.

H5. 

Network quality positively influences the sense of presence in user-platform interactions.

Previous studies established that a sense of immersion engages users to sense virtual products through psychological ownership that persuades satisfaction [25]. Furthermore, prior studies explained that the AR environment enhances the sense of presence through the embodiment of a product and enhances the user’s impression through satisfaction [33,34]. Therefore, we are proposing the following hypotheses:

H6. 

Sense of immersion positively influences satisfaction in user-platform interactions.

H7. 

Sense of presence positively influences satisfaction in user-platform interactions.

Immersion engages closely with subjective sensation with product virtualization, and trust acts as a mediating construct that includes the effects of perceived immersive experiences [17]. AR can alleviate those issues by incorporating immersive features and building trust to interact with mobile platforms [35]. Grubert [15] defined the virtual body as being associated with the physical space, which develops a sense of presence that creates user trust. Furthermore, sense of presence creates a realization of “being there” in the physical space, enhancing trust in products [31]. Therefore, we are proposing the following hypotheses:

H8. 

Sense of immersion positively influences trust in user-platform interactions.

H9. 

Sense of presence positively influences trust in user-platform interactions.

Bhattacharjee emphasized that satisfaction with IS use is a prime concern for users to maintain usage intention [24]. Previous studies suggested that trust influences users’ willingness to use mobile apps [36]. Moreover, Balakrishnan describes the relationship between user trust and continuance intention to use technological systems [22]. Therefore, we propose the following hypotheses:

H10. 

Satisfaction positively influences continuance intention in user-platform interactions.

H11. 

Trust positively influences continuance intention in user-platform interactions.

In the context of age variations in a multicultural society, different age groups create a generation of users that significantly moderates behavioral intention to use mobile platforms. Previous studies have categorized users from various age groups and generations to explain the relationship between behavioral responses and the continuance of intention to use e-commerce platforms [10]. This generation prefers to communicate with images for innovative content, whereas the older generation communicates with text [37]. So, the different age groups with classifications of generations affect behavioral responses to adopt immersive technologies. Therefore, we propose the following hypothesis considering the effect of generation on continuance intention.

H12. 

The generation significantly impacts users’ continuance intention in the user-platform interaction.

Previous studies have followed the Stimulus-Organism-Response (S-O-R) paradigm to investigate immersive experience, repurchase, and continuance intentions [5,14]. This model was rooted in environmental psychology, which considered stimuli cues to explain cognitive reactions [38]. However, the effects of immersive stimuli cues [12], cognitive [13], and emotional [14] engagements toward users’ continuance intention may be explained through an adaptive and integrated model, followed by the SOR framework [23].

An Interaction-Engagement-Intention (I-E-I) model (as shown in Figure 1) was developed based on the SOR framework to determine how interactions-based stimuli cues (product fit, network quality, and AI-driven recommendation) affect user experience (sense of immersion and sense of presence) in the user-platform interaction paradigm. In the context of AR, the proposed model may identify cognitive engagements that involve the processing of sensory information through user-platform interactions. Further, it includes a sense of immersion as a cognitive engagement, reinforcing the model explaining UX through deep concentration in an AR environment. Concerning the virtual try-on effects, the model includes a sense of presence to explain how spatial orientations of virtual products involve users’ engagement with AR.

This I-E-I model includes attitude as an emotional feeling to explain virtual realism through the fulfillment of cognitive actualization. Further, satisfaction was considered an emotional engagement that supported the proposed model in assessing the confirmation of users’ expectations from AR mobile platforms. Subsequently, the I-E-I model incorporates continuance intention as a behavioral response resulting from perceived cognitive and emotional states. The proposed model also considers the effects of user generation on continuance intention to use an AR mobile platform.

4. Methodology

This study considered a quantitative approach that considered respondents, instruments, data-collection method, measuring technique, items, and data analytical processes to check the reliability of constructs, validate the proposed model, and test the hypotheses. A survey guideline (as shown in Appendix A) was designed to collect the quantitative data from the survey (Annex A). An ethical application was approved by the University of Technology Sydney (UTS HREC REF NO. ETH22-7706 on 9 August 2023) as the research study includes involving human participants to collect their experiences. We have followed all the UTS guidelines and human research ethics to conduct an online questionnaire survey.

An online questionnaire survey was conducted in Australia following a convenience snowball sampling technique. The study followed this technique because of the nature of the research, which could identify potential respondents who had already experienced interacting with the AR mobile platform (IKEA) and collect data from homogenous samples, considering the age group of respondents. Also, this study designed a few pre-requisite conditions to scrutinize respondents’ eligibility to enter into the main survey questionnaire.

This study has chosen IKEA as an experimental platform to investigate UX in an AR environment for a multicultural society. In the survey, IKEA was selected as an AR platform because of its unique true-to-scale feature to check the sense of presence from respondents, its large subscriber base with a versatile group of users, and state-of-the-art AR capabilities. Although there are some global AR mobile platforms like Amazon, Target, Nike, and Adidas, their AR features are not supported in Australia. Moreover, we have received permission from IKEA to survey their online users. We considered IKEA an extreme case in this study to externally validate the proposed model and hypotheses because of some limitations with other apps in Australia. Moreover, the research phenomenon also demands data collection from different age groups and user capabilities to interact with AR. Therefore, this study has chosen IKEA as an AR mobile platform that introduced AR technology in Australia.

This study considered product fit, network quality, and AI-driven recommendation as active Independent Variables (IV) applied to respondents through interventions. Further, sense of immersion, sense of presence, satisfaction, trust, and continuance intention were designed as Dependent Variables (DV). Moreover, user generation was considered an Attribute-Independent Variable (IV) that was an intrinsic characteristic of respondents. We considered constructs and measurement items from previous literature, constructed measurement items (

Table A1. Items and Scale Measurement.

Construct	Items	Factors	Measurement Scales
Sources of information such as Stimuli cues for AR mobile platforms Cognitive engagement as an organism Emotional Engagement as an organism Behavioral Response	P NQ AIR SI SP SF TR CI	Product Fit Sun et al. [25] Network Quality Kowalczuk et al. [29] AI-driven recommendation Yin and Qiu [16] Sense of Immersion Dargan et al. [17] Sense of Presence Alimamy and Gnoth [19] Satisfaction Park and Yoo [21], David et al. [20] Trust Kumar et al. [22] Continuance Intention Nguyen and Ha [36]	PF1: I could no longer doubt that the product would fit my desired spaces. PF2: I could measure the product size to check for my desired space. NQ1: The AR mobile platform performs its functions quickly and efficiently. NQ2: The AR mobile platform is reliable (it is always active and running, performs without errors, and does what it is supposed to do). AIR1: When interacting with the platform, AI marketing technology recommends what I want based on browsing habits. AIR2: With the support of AI marketing technology, the AR mobile platform can arouse my shopping desire. SI1: During the interactions, my body was in a physical place, but my mind was in the virtual world. SI2: The AR mobile platform made me forget my immediate surroundings. SP1: I felt like the product meshed with the AR mobile platform. SP2: It seemed to me that I could do whatever I wanted with the products in the AR mobile platform. SF1: Overall, I am satisfied with the AR mobile platform. SF2: The AR mobile platform meets my expectations. TR1: I feel safer using the AR mobile platform. TR2: I am pretty sure what to expect from the platform. CI1: I intend to stay on as a member of using this AR mobile platform. CI2: I will frequently use the AR mobile platform in the future. CI3: I would prioritize the AR mobile platform over other alternative means.

, Appendix B) from validated instruments, and followed the measures of product fit from Sun et al. [25], network quality from Kowalczuk et al. [29], and AI-driven recommendations from Yin and Qiu [16]. We followed this with a sense of immersion from Dargan et al. [17] and a sense of presence from Alimamy and Gnoth [19]. This study also followed the measures of satisfaction from Park and Yoo [21] and David et al. [20], as well as trust from Kumar et al. [22]. The study relates to continuance intention as a behavioral response and measures continuance intention from Nguyen and Ha [36].

This study designed 17 questions for eight constructs to perform the measurement process. The survey data was collected from respondents of different age groups in Australia. To reduce bias, respondents were informed to fill out the questionnaire based on their perceived user experience [23]. Data were critically checked and computed using SPSS (IBM SPSS version 24.0.0.1) and cleaned through the proper validation. Finally, the sample size of 886 (N = 886) was confirmed, with a response rate of 84%. Sociodemographic characteristics were analyzed, and user profiles were prepared from responses. The respondents’ sociodemographic profile is shown in

Table 1. Respondents’ sociodemographic profile (N = 886).

Characteristics	Values	Frequency	Percentage (%)
Gender	Male	533	60.2%
	Female	345	38.9%
	I prefer not to say	8	0.9%
Generation	Generation Z (18–26 years)	202	22.8%
	Millennial (27–42 years)	471	53.2%
	Generation X (43–58 years)	167	18.8%
	Boomers II (59 years or above)	46	5.2%
Education Level	Primary	26	2.9%
	Secondary	72	8.1%
	Bachelor’s degree	524	59.1%
	Vocational or similar	129	14.6%
	Not attended in Australia	135	15.2%
Employment Status	Student	29	3.3%
	Working full-time	99	11.2%
	Working part-time	689	77.8%
	Unemployed	34	3.8%
	Self-employed	5	0.6%
	Others	30	3.4%
Annual Income	Less than 18,200 AUD	19	2.1%
	18,201–45,000 AUD	281	31.7%
	45,001–120,000 AUD	477	57.2%
	120,001–180,000 AUD	57	6.4%
	More than 180,000	22	2.5%

.

The study followed a partial least squares structural equation modeling (PLS-SEM) as a variance-based technique [39] to ensure robustness, predictive responses, and non-normality for the complex model with both formative and reflective constructs [8]. The study followed a two-stage approach: the measurement model assessment tests the reliability of all measurement items and tests the fitness of the outer model, and the structural model assessment evaluates the significance of path co-efficient for the inner model [29]. SmartPLS 4.0 (SmartPLS professional version 4.1.0.6) was used to execute the measurement assessment and develop a structure model to contrast the results. The measurement model acted as an outer model assessment with Confirmatory Factor Analysis (CFA) to determine the reliability of survey instruments and the validity of constructs [29]. Then, the structural model was used as an inner model assessment to test the hypotheses [12].

5. Results

This study used confirmatory factor analysis to check the reliability and validity of the measurement items and constructs in the measurement model [9]. The measurement model was analyzed through factor loadings, Cronbach’s alpha, composite reliability, and Average Variance Extracted (AVE) to assess the reliability of the survey instrument.

Table 2. Measurement Model Assessment.

Constructs	Items	Factor Loading	Cronbach’s Alpha	Composite Reliability (rho_a)	Average Variance Extracted (AVE)
Product fit	PF1	0.958	0.911	0.855	0.919
	PF2	0.959
Network quality	NQ1	0.946	0.879	0.879	0.892
	NQ2	0.943
AI-driven recommendation	AIR1	0.924	0.924	0.927	0.929
	AIR2	0.966
Sense of immersion	SI1	0.952	0.902	0.903	0.911
	SI2	0.956
Sense of presence	SP1	0.943	0.880	0.880	0.834
	SP2	0.946
Satisfaction	SF1	0.958	0.912	0.912	0.919
	SF2	0.960
Trust	TR1	0.943	0.881	0.882	0.894
	TR2	0.948
Continuance intention	CI1	0.819	0.750	0.752	0.724
	CI2	0.833
	CI3	0.798

describes the psychometric properties of all the constructs deployed in the study. As retrieved from the data, all the outer loadings are above 0.70, and the average variance extracted (AVE) values are above 0.50. Cronbach’s alpha value of all constructs was assessed from 0.750 to 0.924, which is higher than the recommended value of 0.7 [35]. CR values of all constructs were assessed as higher than the recommended value of 0.70. (i.e., from 0.752 to 0.927). Further, the AVE values were found to be higher than MSV (Maximum Shared Variance) and ASV (Average Shared Variance), which shows the strength of AVE and confirms the threshold for discriminant validity [14].

In the structural model assessment, we initially assessed the value of R² for all dependent variables and then observed the collinearity statistics value for all the relationships [39]. The result showed that the value of R² for all the variables was within the range of 0.11–0.25 (except sense of presence), which was recommended as a higher value. R² values for sense of immersion (0.208), sense of presence (0.194), satisfaction (0.334), trust (0.272), and continuance intention (0.389) showed reasonable variance explained. Also, the collinearity statistics (VIF) value for all the relationships was less than 4.0 (as shown in

Table 3. R-square statistics and Collinearity statistics (VIF).

R2 Statistics			Collinearity Statistics (VIF)
Construct	R-Square	R-Square Adjusted	Relationship	VIF
SI	0.208	0.204	PF -> SP	1.000
SP	0.194	0.191	NQ -> SN	1.235
SF	0.334	0.331	AIR -> SN	1.235
TR	0.272	0.271	SI -> SF	1.250
CI	0.389	0.387	SP -> SF	1.480
			SI -> TR	1.348
			SP -> TR	1.354
			SF -> CI	1.178
			TR -> CI	1.278

).

The significance of the hypotheses was tested through path co-efficient (β), t-value, and p-value in the structural model assessment. All the hypotheses were supported with the p ˂ 0.05 level. The results for the AR environment showed that all hypotheses were supported and sources of information positively influenced users’ behavioral responses. The structural model results (hypothesis testing) are provided in

Table 4. Summary of hypotheses testing results.

Hypothesized Relationship	Mean (M)	Std. Deviation	T Statistics	p Values	Hypothesis Result
H1: PF -> SI	0.150	0.035	1.439	0.001 ***	Yes
H2: NQ -> SI	0.156	0.058	2.688	0.007 **	Yes
H3: AIR -> SI	0.136	0.036	3.792	0.001 ***	Yes
H4: PF -> SP	0.197	0.035	2.787	0.003 **	Yes
H5: NQ -> SP	0.104	0.052	2.104	0.044 *	Yes
H6: SI -> SF	0.227	0.036	6.325	0.000 ***	Yes
H7: SP -> SF	0.343	0.035	9.740	0.000 ***	Yes
H8: SI -> TR	0.226	0.037	6.089	0.000 ***	Yes
H9: SP -> TR	0.321	0.036	8.964	0.000 ***	Yes
H10: SF -> CI	0.112	0.042	2.619	0.009 **	Yes
H11: TR -> CI	0.125	0.042	2.986	0.003 **	Yes

Note: p > 0.05 = ns (not significant), p < 0.05 = *, p < 0.01 = **, p < 0.001 = ***.

.

The results showed that the effects of AI-driven recommendation on a sense of immersion (t value = 3.792) and product fit toward a sense of presence (t value = 2.787) were more significant. Further, the impacts of sense of presence on the relationship between satisfaction (t value = 9.740) and trust (t value = 8.964) were more significant. Similarly, the relationship of trust toward continuance intention (t value = 2.986) was higher than that of satisfaction with continuance intention (t value = 2.619).

We have tested the effects of attribute-independent variables (age group and gender) on continuance intention. Among the variables, gender did not have an impact on continuance intention. However, the age group significantly affected users’ continuance intention. In particular, Generation Z (β = 0.595, t = 2.863) had more significant effects rather than Generation X (β = 0.505, t = 2.285) and Millennials (β = 0.521, t = 2.504) on continuance intention (as shown in

Table 5. Effects of variables test analysis result.

Effect of Attribute IV	Original Sample (O)	Sample Mean (M)	Standard Deviation	T Statistics	p Values	Significance
Generation Z -> CI	0.595	0.601	0.208	2.863	0.002 **	Significant
Millennial -> CI	0.521	0.525	0.208	2.504	0.006 **	Significant
Generation X -> CI	0.505	0.512	0.221	2.285	0.011 *	Significant

Note: p ˃ 0.05 = ns (not significant), p ˂ 0.05 = *, p ˂ 0.01 = **.

).

6. Discussion

This study contributed to the theoretical understanding of perceived user experience by considering user-platform interactions in AR mobile platforms, identifying and validating cognitive and emotional engagements, and investigating their impacts on users’ continuance intention. Although some of those variables have been determined in other contexts, we extended to empirical evidence of their applicability with user-platform interactions in the AR mobile platform context.

The I-E-I model extended the theoretical knowledge by explaining cognitive engagements through sense of immersion and sense of presence. The model also included satisfaction and trust as emotional feelings to explain the confirmation of users’ expectations from the AR mobile platform. The emotional engagements supported the model in assessing the users’ emotional feelings as an outcome of perceived cognitive engagements. The model also incorporated the generational effect on users’ continuance intention. It theorized a more vigorous and effective way to explain the perceived UX through user-platform interactions in an AR environment.

In this study, the results indicated the outcomes in continuation with the previous studies, as our findings supported the impact of perceived cognitive [12,22] and emotional factors [14] on continuance intention [39]. This study considered two cognitive variables to extant evidence as a claim that users respond to AR through user-platform interactions and co-create values through cognitive engagements. Further, the study examined two emotional variables to address users’ emotional reactions. Moreover, the findings confirmed that user generation significantly influences continuance intention. However, the results indicated that gender has no significant effect on users’ continuance intention. Therefore, users perceive UX through these perceived cognitive and emotional engagements that affect users’ continuance intention to use AR mobile platforms with the effect of user generation.

The results showed that sources of information, especially the effects of AI-driven recommendation on a sense of immersion and product fit toward a sense of presence, were more significant. Also, the influences of sense of presence toward satisfaction and trust were more significant. Similarly, the relationship of trust toward continuance intention was higher than that of satisfaction with continuance intention. The findings affirmed that generation significantly impacted users’ continuance intention in an AR environment. In particular, Generation Z influenced continuance intention more significantly than Generation X and Millennials.

This study contributed to the theoretical understanding of perceived UX considering AR by identifying and validating the cognitive and emotional factors and investigating their impacts on users’ continuance intention. The findings confirmed the significance of the relationships between cognitive and emotional engagements, extending the co-created values in perceiving UX.

Practically, this study complemented the value of optimizing app development and UX design through UX assessment. Online and interactive shopping benefits through AR were explored in this study as an integral part of information system research. In the context of processing sensory information through multi-modal AI interactions with AR, this study explained the value propositions of how AR engages more sensory information and allows users to interact with mobile platforms by alleviating product uncertainty and the need to touch the product.

Marketers can leverage the study’s findings to develop targeted marketing strategies that appeal to different demographic segments, particularly for Generation Z. Emphasizing the development of satisfaction and trust-building mechanisms can shape users’ perceptions and foster continued usage. Research findings could be applied to technologies like virtual reality (VR), the Internet of Things (IoT), and 5G in different industries.

7. Conclusions

The user experience is crucial for IS researchers in retaining users and moving toward the sustainable growth of AR mobile platforms in the e-commerce industry. AR involves more sensory information to users and supports them in changing their cognitive and emotional engagements through user-platform interactions. Consequently, these perceived values make users more satisfied and trusted in influencing continuance intention. This study empirically investigated the effects of interactions-based stimuli cues on user experience that influence continuance intention. Also, we have identified the mediation effects of trust and satisfaction on the relationship between the perceived sense of immersion, sense of presence, and continuance intention.

The UX assessment is crucial for IS researchers, and this study explained the consequences of perceived UX on continuance intention as an extension of user retention to achieve sustainable growth through AR adoption in the e-commerce industry. As a sustainable growth in the retail sector, user-platform interactions with augmented reality enhance immersive and interactive experiences, and this study confirmed the significance of examining the consequences of user-platform interactions in AR and their effects on continuance intention.

The focus on a particular AR platform, the IKEA app, may border the generalizability of findings because of its limited ability to collect data for a specific context. This study could be limited to context-dependent and multicultural environments that may not be realized in different contexts. There was a limitation in getting permission from platforms to conduct research considering their users. Further, the nature of the research could limit the study’s choice of a convenience snowball sampling technique to collect data from different age groups. Also, a particular age group, gender, or limited subscriber base would be a challenge in collecting data to validate the I-E-I model. Concerning immersive technologies, there needs to be more technical know-how and user capabilities to interact with AR mobile platforms.

Future research could address these limitations by employing more extensive and diverse samples, using objective measures, exploring different AR platforms with different contexts, and considering cross-cultural influences. The proposed model can be validated with other industries like tourism, health, education, and gaming. Moreover, a cumulative UX assessment can be extended to respondents who have been using AR for a long period.