Augmented reality (AR) is a technology that boosts or augments real-time experiences by incorporating digital information into live objects. Retailers’ application of AR has taken the entire world by storm, enriching the relationship between consumers and brands. This study explores the multifaceted usage of AR in the retail sector.
Two studies were conducted across two different time frames to understand the continued use of AR in retail dynamics. 394 respondents took part in Study One, while 368 respondents participated in Study Two. The research presented probes the moderating effect of e-word of mouth (e-WOM). The researchers also further explore the impact of AR on consumer behaviour by considering factors such as convenience, experience, curiosity, fantasy, and entertainment.
Based on Study One’s findings, the adoption of AR for online shopping is influenced more by fantasy and curiosity rather than entertainment. The findings of Study Two indicate that experience, convenience and social influence are key factors in driving the adoption of AR for online shopping. By exploring the relationship between these variables, we provide a comprehensive understanding of how AR impacts consumer perceptions.
This study extends gratification theory by analysing AR adoption in retail through a two-study approach, identifying key motivations and the moderating role of e-WOM. It differentiates fantasy from entertainment in AR adoption, emphasising curiosity and social influence. Our findings offer valuable insights for retailers and technologists to enhance AR-driven shopping experiences and consumer engagement.
1. Introduction
The use of augmented reality (AR) in global retail markets was valued at two billion dollars in 2021 and is estimated to grow to 61.3 billion dollars by 2031 with a compound annual growth rate of 41.4% from 2022 to 2031 (Davis and Aslam, 2024). Further to this, 52% of retailers assert they are prepared to integrate AR into the shopping experience (Shankar et al., 2021). AR has bridged the gap between online and offline shopping by overcoming the limitations attached to the physical and digital worlds (Butt et al., 2023; Castillo and Bigne, 2021). Studies have predicted the downfall of physical stores and the replacement of human labour by disruptive technologies like AR, virtual reality (VR), and artificial intelligence (AI) (Huang and Rust, 2018).
Retail is one sector which promises huge opportunities to implement AR (Cruz et al., 2019). AR offers a platform for consumers to experience offline and online services through smart devices (Heller et al., 2019; Kumar, 2022). Moreover, AR strongly influences the consumer shopping experience in retailing, and guides consumers in their decision-making by reducing uncertainty (Dacko, 2017). As a technology, AR is witnessing increased acceptance and adoption in the retail sector and previous researchers have attempted to study its application in retailing and other business sectors (Yim et al., 2017).
Since its application and usage are still evolving, it has become pertinent to conduct research on this topic. A key contribution of this study lies in the adoption of a two-study approach. In the first phase, the paper examines the role of entertainment, fantasy, and curiosity and their effect on attitudes towards AR, which leads to adoption intentions with of e-word of mouth (e-WOM) as a moderating variable. In the second phase, the study focuses on the role of experience, convenience, and social influence on attitudes towards AR and the adoption and continuation of AR with e-WOM as a moderating variable. Our research aims to fill this gap by conducting a survey on how consumers perceive and embrace AR in the retail industry. Specifically, this paper aims to investigate the following research questions:
Does e-WOM moderate the relationship between factors for AR and attitude towards AR?
Is continued use of AR influenced by attitude and adoption intention?
Since AR is still evolving, retailers and consumers have limited exposure towards its application in the retail sector (Pandey and Pandey, 2025). It becomes imperative to understand the factors that could facilitate the adoption of this technology, leading to improved efficiency and customer shopping experience. This study can provide valuable insights for marketers, retailers, technologists, and researchers seeking evidence-based knowledge. These findings can be especially valuable for retailers developing effective strategies for encouraging consumers to incorporate AR into their purchasing habits.
The dual-study approach employed in this paper offers new and notable insights into the evolving role of AR in retail, distinguishing between fantasy and entertainment in AR adoption, and revealing the distinct impact of curiosity and social influence on consumer attitudes. It integrates the e-WOM moderating influence to extend gratification theory and demonstrates the relationship between dynamic factors of both adoption and continued use of AR technology in retail. This provides practical recommendations for retailers, technologists, and policymakers aimed at improving shopping experiences centred on AR technology.
Moreover, the analysis spanning two periods illustrates the transition of the consumer mindset from initial curiosity and exploration to sustained commitment and trust associated with AR-based shopping. Such a perspective provides insights into the evolving consumer gratifications from the initially hedonic uses to subsequently functional and social uses. Moreover, the study outlines how e-WOM intricately shapes consumer disposition at every step, reiterating the fundamentally transformative role of e-WOM in contemporary digital decision-making processes.
2. Theoretical background
2.1 The retail sector and augmented reality (AR)
The retail sector is witnessing disruptive changes due to widespread digitalisation, transforming consumers’ shopping experience (Chakraborty et al., 2024a, b; Caboni and Hagberg, 2019), and the increase in the incorporation of interactive technologies has caught the attention of retailers and consumers (Aslam and Davis, 2024; Chakraborty and Rana, 2025). Traditional retail formats are witnessing radical changes, and immersive technologies blend with the traditional retail format offering consumers a personalised and interactive shopping experience (Sung et al., 2021). AR has been described as a combination of three elements, namely, real-time interaction, assimilation of real life, and virtual elements (Rumokoy and Frank, 2025). More recently, AR has been used extensively in the retail sector as it is designed to assist the consumer experience across the channels, which helps attract and retain customers (Watson et al., 2020).
2.2 Theory of gratification
Devised in 1940 by Katz and Blumler, Uses and Gratification (U&G) theory helps understand how a particular set of individuals use a particular type of media (Kaur et al., 2020; Şahin et al., 2025), what is the fundamental reason behind their usage, and most importantly, what kind of gratification they derive out of using it (Wei et al., 2024; Korhan and Ersoy, 2016). Some researchers (Chiu and Huang, 2015) have deliberated on the usage and level of gratification derived from Internet usage while making a purchase decision (Yadav et al., 2024). U&G theory is highly applicable in the current context for several reasons, including that: this theory is well-adopted and verified in an academic context for contemplating the motives for using a given means or a forum (Ray et al., 2019); that it can help understand different user types (Xu et al., 2025); and that it can provide a deep understanding of the connotation and the key factors affecting the online purchase behaviours of consumers (Pushparaj and Kushwaha, 2024).
3. Research model and hypothesis development
The current study has developed a research model by incorporating the Theory of Gratification for adopting AR in retail. The basic assumption associated with U&G theory is that users are actively connected with innovative technology (Xu et al., 2025). The study has been divided into two phases in which six independent variables are connected to attitudes towards the adoption of AR in retail. The six independent variables consist of entertainment, fantasy, and curiosity in Study One, while in Study Two, experience, convenience and social influence act as independent variables and attitude towards AR, adoption of AR, and continued use of AR act as dependent variables. e-WOM towards AR is the moderating variable examined in this paper.
3.1 Entertainment
Entertainment is a passive experience consumers develop while watching a theatrical performance, affecting their emotions and satisfaction levels (Gupta et al., 2024). Entertainment is a kind of value addition in AR that enriches the overall experience of shopping (Brengman et al., 2019). Advanced technologies like AR will significantly influence the future framework of entertainment (Alsoud et al., 2024; Talaviya et al., 2020). The implementation of AR entertainment can make the shopping process more enjoyable and interactive for shoppers by gamifying routine tasks (De Canio et al., 2021). Hence, we propose:
Entertainment positively affects attitude towards AR.
3.2 Fantasy
The extent to which the situation or environment in which an individual is present is different from the actual real scenario is called fantasy or virtuality (Park, 2017; Wrzus et al., 2024). Even with similar immersive technology being used, the user’s fantasy may vary depending on how store design matches the real world (Kim and Choo, 2023), and the consumer experience is enhanced in a fantasy-based store compared to a reality-based store. Perceived fantasy attained by shopping contributes to consumers’ enjoyment by allowing them to escape the real world (Hussain et al., 2025; Song et al., 2007). The AR shopping interface allows brands and consumers to establish deep emotional connections that transcend simple consumer purchase engagement because of the fantasy level of immersion and escapism provided by AR (Preece and Skandalis, 2024). Hence, we propose:
Fantasy positively affects attitude towards AR.
3.3 Curiosity
Previous studies on consumer retailing have defined curiosity as a temporary state of motivation triggered by a new stimulus (Xue et al., 2023). When consumers are inquisitive, the probability of them using VR shopping features increases as it transforms into an enjoyable activity (Erensoy et al., 2024), and this type of shopping exploration creates new experiences, which leads to the possibility of having positive attitudes towards this technology (Kumar et al., 2024). A significant relationship exists between the use of highly immersive technology and perceptual curiosity (Kim and Choo, 2023). Hence, we propose:
Curiosity positively affects attitude towards AR.
3.4 Experience
Experience is an important medium for differentiating between users and non-users (Kaushik et al., 2020). The rapid expansion of innovative digital technology offers immense potential to retailers to provide personalised visual experiences to customers and building interactive relationships (Wang, 2021). Some past studies have examined how consumers form positive emotional connections with products when interactive technologies offer detailed visualisations that help them gather information (Shamim et al., 2025; Yim et al., 2017). In addition to making consumers more informed, such tailored experiences increase consumers’ trust and confidence in their decisions (Zimmermann et al., 2023). Hence, we propose:
Experience positively affects attitude towards AR.
3.5 Convenience
AR offers consumers an intelligent, convenient and efficient shopping process (Chakraborty et al., 2024a, b). It enriches the overall shopping experience by providing useful product information and by offering facilities like self-checkout, navigation and self-shopping (Xue et al., 2023). AR apps help consumers search for specific products, save time and effort, and improve their touchpoints by offering more convenience (Xue et al., 2023). By effortlessly embedding digital assistance throughout the purchasing process, customer loyalty is reinforced while repeated use is encouraged due to heightened convenience (Kang et al., 2023). Thus, we propose:
Convenience positively affects attitude towards AR.
3.6 Social influence
Social influence refers to how peer groups influence and change consumer behaviour. (Xue et al., 2023). The key motivators for using AR-based devices have been linked to self-expression, socialising and entertainment (Rauschnabel et al., 2019). The novelty offered by AR triggers discussion regarding its novel content, which may pressure individuals regarding its use of AR features, motivating like-minded people to use this technology (McLean and Wilson, 2019; Hong et al., 2025). The collective social enthusiasm for AR can boost its adoption as consumers look for affirmation and social interaction through shared technological experiences (Wang et al., 2023a, b). Hence, we propose:
Social influence positively affects attitude towards AR.
3.7 Adoption of AR
The behavioural intention of consumers to use AR can be gauged by their attitude, which aids in understanding key consumer differences (Jiang et al., 2021). Media enjoyment and usefulness significantly influence the user’s attitude towards AR (Yim et al., 2017). Studies have shown that in the retail sector, consumers respond to new technology regarding shopping behaviour (Jiang et al., 2021). Particularly in retail settings, a favourable disposition towards AR not only propels first-time adoption but also influences sustained interaction and continued use over the long term (Wang et al., 2023a, b). Thus, we propose:
Attitude towards AR positively influences the adoption of AR.
3.8 Continued use of AR
Previous studies have revealed that lacking primary expertise negatively affects the intention to adopt technology (Lin and Lin, 2008). Knowledge enhancement helps overcome the barriers and challenges involved in adopting new technology (Berg and Lingen, 2019). The continued use of any learning system is also determined by user satisfaction level. Satisfaction is a strong predictor of continued used by a user. (Rajeh et al., 2021), and perceived enjoyment, ease of use and usefulness were significant predictors of continued use intentions for AI-based service providers (Ashfaq et al., 2020). Hence, we propose:
Adoption of AR positively influences the continued use of AR.
3.9 Moderating role of e-word of mouth (e-WOM)
Positive e-WOM generated through favourable or unfavourable experiences plays a crucial role in shaping the consumer’s purchase intention (Patel et al., 2022). Customer satisfaction and intention to repurchase directly impact e-WOM (Verma et al., 2023). Previous studies have shown that the application of AR generates positive e-WOM and favourable behavioural intentions (Khasawneh and Rabata, 2023). Customer engagement, e-WOM and loyalty have also been shown to be influenced by the quality of service received by consumers (Sampat and Sabat, 2021). Hence, we propose:
e-WOM has a moderating effect on the association between: (a) entertainment; (b) fantasy; (c) curiosity; (d) experience; (e) convenience; and (f) social influence and attitude towards AR.
Drawing these hypotheses together, Figure 1 depicts the proposed model for the study.
The model shows two grouped sections on the left labeled “Study 1” and “Study 2”. Under “Study 1”, three rectangular text boxes are arranged vertically and labeled “Entertainment”, “Fantasy”, and “Curiosity”. Right-pointing arrows labeled “H 1”, “H 2”, and “H 3” extend respectively from these three boxes toward a central text box labeled “Attitude towards A R”. Under “Study 2”, three rectangular text boxes are arranged vertically and labeled “Experience”, “Convenience”, and “Social Influence”. Right-pointing arrows labeled “H 4”, “H 5”, and “H 6” extend respectively from these three boxes toward the same central text box labeled “Attitude towards A R”. From “Attitude towards A R”, a right-pointing arrow labeled “H 7” leads to a text box labeled “Adoption of A R”. A further right-pointing arrow labeled “H 8” leads from “Adoption of A R” to a text box labeled “Continued use of A R in Retail”. Additionally, a text box below the central area is labeled “E-W O M towards A R”. Multiple upward dashed arrows labeled “H 9 a, H 9 b, H 9 c, H 9 d, H 9 e, H 9 f” from “E-W O M towards A R” points to the arrow “H 1”, “H 2”, “H 3”, “H 4”, “H 5”, and “H 6”.Proposed model
The model shows two grouped sections on the left labeled “Study 1” and “Study 2”. Under “Study 1”, three rectangular text boxes are arranged vertically and labeled “Entertainment”, “Fantasy”, and “Curiosity”. Right-pointing arrows labeled “H 1”, “H 2”, and “H 3” extend respectively from these three boxes toward a central text box labeled “Attitude towards A R”. Under “Study 2”, three rectangular text boxes are arranged vertically and labeled “Experience”, “Convenience”, and “Social Influence”. Right-pointing arrows labeled “H 4”, “H 5”, and “H 6” extend respectively from these three boxes toward the same central text box labeled “Attitude towards A R”. From “Attitude towards A R”, a right-pointing arrow labeled “H 7” leads to a text box labeled “Adoption of A R”. A further right-pointing arrow labeled “H 8” leads from “Adoption of A R” to a text box labeled “Continued use of A R in Retail”. Additionally, a text box below the central area is labeled “E-W O M towards A R”. Multiple upward dashed arrows labeled “H 9 a, H 9 b, H 9 c, H 9 d, H 9 e, H 9 f” from “E-W O M towards A R” points to the arrow “H 1”, “H 2”, “H 3”, “H 4”, “H 5”, and “H 6”.Proposed model
4. Research methodology
Since AR is an emerging technology, people’s attitudes and perceptions evolve over time (Santiago et al., 2024); hence two studies were conducted using a convenience sampling method. The study’s time frame covered two different periods (Study One: February – July 2023 and Study Two: April – September 2024) to understand consumer attitudes towards the adoption and continued use intention of AR in retail. The links between factors shaping consumers attitude towards AR in retail and e-WOM evolve depending on the situation and time period. The chosen respondents had made online purchases from AR-based retail stores in the previous year. This study’s participants were selected from Tier One and Tier Two cities of India (Tier One and Tier Two cities mean Metropolitan and Developed cities).
Customer understanding of AR in retail necessitates the dual-study design employed in this research. Given rapid developments in technology, user attitudes will continue to evolve based on exposure and familiarity. For the first study, we focused on initial adoption drivers, such as entertainment, fantasy, and curiosity, which capture the exploration and novelty-seeking motivations of consumers in the early stages of technological adoption. These variables are also aligned to the hedonic and imaginative features of AR in users’ early experiences. This phase of the research helps understand the psychological motivations that first-time consumers are seeking in retail experiences that are AR-enabled.
The second study, however, emphasises the continuation and sustained engagement phase with a focus on experience, convenience, and social influence. As consumer familiarity with AR increases, the enthusiasm tied to novelty wears off, and expectations shift toward usefulness and social validation. This phase appreciates that, after curiosity wanes, continued engagement is sustained exclusively through functional utility, convenience, and social proof. Through a comparison of both phases, the study demonstrates a shift in motivational factors from hedonic to utilitarian and social aspects. This provides a longitudinal perspective on AR adoption that would be unattainable through a solitary cross-sectional study.
In contrast to the limitations of a traditional single-study design, which provides only a static picture, this two-study approach has, albeit to a slight extent, worked towards enhancing both practical and theoretical value by applying U&G theory at different points in the adoption cycle. It demonstrates how gratifications associated with AR evolve over time – from fantasy-driven adoption to experience- and convenience-driven continuity – and how e-WOM differentially manages these connections across various stages.
The two-study design was necessary for this research on consumers’ attitudes and behaviours concerning AR in retail to capture the shifts over time, as perceptions of new technologies such as AR change with increased familiarity and exposure. This research was able to capture both the initial adoption factors driven by entertainment, fantasy, and curiosity, as well as sustained engagement stemming from experience, convenience, and social influence, alongside assessing the shifting moderating role of e-WOM. This approach enhances the understanding of AR adoption and continued use, thus increasing the relevance and validity of the study.
A detailed questionnaire was developed based on feedback received from a team of experts comprising 34 people from industry and academia. An online survey was conducted among respondents who were selected based on their shopping frequency through AR-based retailers. Only those respondents who had purchased more than six times in the last year were chosen for the study. The first study included factors like entertainment, fantasy, and curiosity and their effect on attitude towards AR in retail, and in the second study, experience, convenience, and social influence and their effect on attitude were examined along with the moderating role of e-WOM. A five-point Likert scale was used to measure the responses.
For Study One, questionnaires were sent to 1843 respondents through email and social media, out of which 394 took part in this study (after filtering out incomplete responses). The respondents comprised 54.57% males, with a majority of 39.09%, belonging to the age group of 25–40 years. Additionally, 44.67% were graduates, and 38.32% had a monthly income between 501-750USD. For Study Two, we asked the 394 respondents from Study One to participate again, with 368 respondents taking part in this study. The demographic profile of respondents comprised 38.59% belonging to the age group of 25–40 years, 56.52% were males, 45.38% were graduates, and 39.95% had a monthly income between 501-750USD.
The constructs were extracted from the existing literature such as entertainment (Zeng et al., 2023), fantasy and curiosity (Kim and Choo, 2023), experience (Kim et al., 2022), convenience (Jiang et al., 2021), social influence (Kang et al., 2023), attitude towards AR (Romano et al., 2022), adoption of AR (Dogra et al., 2023), e-WOM (Legman et al., 2023) and continued use of AR in retail (Kim et al., 2022).
5. Findings
5.1 Analysis of the collected data
The data analysis was conducted using a two-step methodology. Both exploratory factor analysis and confirmatory factor analysis (CFA) were conducted to assess the reliability and validity of the scales used in this study. Structural Equation Modelling (SEM) was used to evaluate the theoretical model through AMOS 26. Process macro 3.5.3 was used to investigate the moderating variable.
5.2 Common method bias
Using a multidimensional approach, we addressed the issue of common method bias that may arise from gathering data from a single source. Harman’s single-factor test was conducted during the initial stage to assess the mono-method variance. SPSS was used to assess the reliability of the constructs, revealing a total variance score of 38.65. If the score falls below 50% (the threshold), it can be inferred that the data has no common method bias.
5.3 Results of CFA (Study 1 and Study 2)
CFA was employed to assess the validity and reliability of the indices (see Table 1). All the factor loadings were above 0.7, except for CNT. This indicates that the items in the factor could meet the reliability threshold limit (Fornell and Larcker, 1981). The values for CNT were 0.696 and 0.683, respectively, which met the criteria for the study. As part of the analysis, a composite reliability test was conducted which indicated that the values exceeded the threshold of 0.7 and were considered acceptable. The composite reliability value exceeded the threshold of 0.70, while the value of Average Variance Extracted (AVE) surpassed 0.50, meeting the acceptable parameters established by Fornell and Larcker (1981). Discriminant validity was successfully established by demonstrating that the square root of the AVE exceeded the correlation coefficients between the constructs (Fornell and Larcker, 1981). Based on the findings, it was evident that all constructs demonstrate strong reliability (see Table 2). Applying the heterotrait-monotrait (HTMT) technique, we examined the discriminant validity between the constructs (see Table 3) and found all constructs exhibited discriminant validity.
Constructs, items and sources
| Constructs | Item nos | Items | Sources | Study 1 | Study 2 |
|---|---|---|---|---|---|
| FL | FL | ||||
| Entertainment (ENT) | ENT1 | I think that AR tools in retail are very entertaining | Zeng et al. (2023) | 0.800 | |
| ENT2 | The enthusiasm for AR tools is catching, it would pick me up | 0.696 | |||
| ENT3 | The retail AR tools do not sell any product but entertain me | 0.717 | |||
| Fantasy (FAN) | FAN1 | The AR-based websites were fantastic | Kim and Choo (2023) | 0.796 | |
| FAN2 | The AR-based websites seem unrealistic | 0.803 | |||
| FAN3 | My experience with AR-based websites seemed imaginary | 0.882 | |||
| Curiosity (CUR) | CUR 1 | The AR-based website grabs my attention | Kim and Choo (2023) | 0.747 | |
| CUR 2 | The AR-based websites make me curious | 0.785 | |||
| CUR 3 | I wonder about the AR-based websites/store | 0.722 | |||
| CUR 4 | I want to know and experience more of such stores/websites | 0.812 | |||
| Experience (EXP) | EXP 1 | I think that using AR-based websites was very exciting | Kim et al. (2022) | 0.944 | |
| EXP 2 | I felt that I could touch the products | 0.914 | |||
| EXP 3 | I felt that I was part of the scene which was displayed | 0.885 | |||
| EXP 4 | My experience was not very pleasant | 0.801 | |||
| Convenience (CON) | CON 1 | I found AR-based websites are user friendly | Jiang et al. (2021) | 0.942 | |
| CON 2 | Shopping was very smooth and hassle-free due to AR-based websites | 0.920 | |||
| CON 3 | The services offered by AR based websites are prompt and efficient | 0.828 | |||
| Social Influence (SCP) | SCP1 | I think using AR in shopping make me look trendy | Kang et al. (2023) | 0.920 | |
| SCP2 | My friends and peer group encourage me to use AR-based shopping websites | 0.853 | |||
| SCP3 | I feel outdated if I do not use AR-based shopping website | 0.882 | |||
| Attitude towards AR (ATT) | ATT1 | The use of AR in retail is a good idea | Romano et al. (2022) | 0.777 | 0.872 |
| ATT2 | I view the AR tools with a positive approach | 0.829 | 0.707 | ||
| ATT3 | I think other customers should also use AR-based websites | 0.809 | 0.820 | ||
| ATT4 | It is sensible to use AR tool in the retail sector | 0.874 | 0.732 | ||
| Adoption of AR (ADP) | ADP1 | I will recommend AR-based shopping websites to my friends | Dogra et al. (2023) | 0.774 | 0.907 |
| ADP2 | AR-based shopping is very appealing | 0.860 | 0.888 | ||
| ADP3 | I will give priority to AR-based shopping over traditional shopping | 0.786 | 0.763 | ||
| E-Word of Mouth (e-WOM) | E-WOM 1 | I highly recommend AR in retail to others | Legman et al. (2023) | 0.734 | 0.918 |
| E-WOM 2 | I have a highly favourable view of AR-based websites | 0.786 | 0.942 | ||
| E-WOM 3 | I will encourage my friends and family members to explore AR-based websites | 0.704 | 0.905 | ||
| Continued use of AR in Retail (CNT) | CNT 1 | I will be willing to shop from AR-based websites | Kim et al. (2022) | 0.629 | 0.827 |
| CNT 2 | The likelihood of me using the AR-based websites for shopping is very high | 0.717 | 0.901 | ||
| CNT 3 | I am willing to recommend AR-based shopping websites to my friends | 0.811 | 0.905 |
Reliability and validity tests
| CR | AVE | MSV | MaxR(H) | ENT | FAN | CUR | ATT | ADP | CNT | EWOM | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Study 1 | |||||||||||
| ENT | 0.783 | 0.546 | 0.144 | 0.791 | 0.739 | ||||||
| FAN | 0.867 | 0.686 | 0.223 | 0.876 | 0.286*** | 0.828 | |||||
| CUR | 0.851 | 0.589 | 0.212 | 0.855 | −0.273*** | −0.350*** | 0.767 | ||||
| ATT | 0.893 | 0.677 | 0.24 | 0.898 | 0.222*** | 0.472*** | −0.461*** | 0.823 | |||
| ADP | 0.849 | 0.652 | 0.24 | 0.856 | 0.238*** | 0.313*** | −0.384*** | 0.490*** | 0.808 | ||
| CNT | 0.765 | 0.523 | 0.178 | 0.784 | 0.380*** | 0.287*** | −0.422*** | 0.392*** | 0.406*** | 0.723 | |
| EWOM | 0.786 | 0.551 | 0.185 | 0.79 | 0.213** | 0.431*** | −0.239*** | 0.282*** | 0.130* | 0.111† | 0.742 |
| Study 2 | |||||||||||
| EXP | 0.937 | 0.788 | 0.552 | 0.949 | 0.888 | ||||||
| CON | 0.926 | 0.807 | 0.552 | 0.94 | 0.743*** | 0.898 | |||||
| SCP | 0.916 | 0.784 | 0.474 | 0.921 | 0.647*** | 0.689*** | 0.886 | ||||
| ATT | 0.865 | 0.617 | 0.397 | 0.881 | 0.598*** | 0.594*** | 0.630*** | 0.786 | |||
| ADP | 0.89 | 0.731 | 0.249 | 0.907 | 0.410*** | 0.445*** | 0.499*** | 0.414*** | 0.855 | ||
| CNT | 0.91 | 0.771 | 0.344 | 0.916 | 0.548*** | 0.587*** | 0.497*** | 0.498*** | 0.195*** | 0.878 | |
| EWOM | 0.944 | 0.849 | 0.269 | 0.947 | 0.510*** | 0.476*** | 0.425*** | 0.396*** | 0.223*** | 0.519*** | 0.922 |
Note(s): Significance of correlations †p < 0.100, *p < 0.050, **p < 0.010, ***p < 0.001
HTMT analysis
| ENT | FAN | CUR | ATT | ADP | CNT | EWOM | |
|---|---|---|---|---|---|---|---|
| Study 1 | |||||||
| ENT | |||||||
| FAN | 0.285 | ||||||
| CUR | 0.27 | 0.337 | |||||
| ATT | 0.232 | 0.471 | 0.458 | ||||
| ADP | 0.237 | 0.309 | 0.39 | 0.492 | |||
| CNT | 0.377 | 0.267 | 0.428 | 0.399 | 0.419 | ||
| EWOM | 0.212 | 0.442 | 0.228 | 0.286 | 0.132 | 0.114 | |
| Study 2 | |||||||
| EXP | |||||||
| CON | 0.742 | ||||||
| SCP | 0.638 | 0.706 | |||||
| ATT | 0.59 | 0.596 | 0.628 | ||||
| ADP | 0.425 | 0.459 | 0.49 | 0.414 | |||
| CNT | 0.535 | 0.596 | 0.502 | 0.489 | 0.197 | ||
| EWOM | 0.502 | 0.475 | 0.424 | 0.394 | 0.223 | 0.517 | |
5.4 Hypothesis testing: Study 1 and Study 2
An analysis was conducted using SEM with AMOS 26 to test the hypotheses for Study One. Based on the analysis of model fit, the results indicate a Chi-Square Minimum Discrepancy (χ²) (CMIN)/df ratio of 1.851, a Goodness-of-Fit Index (GFI) value of 0.933, an Adjusted Goodness-of-Fit Index (AGFI) value of 0.913, a Normed Fit Index (NFI) value of 0.923, a Tucker–Lewis Index (TLI) value of 0.956, a Comparative Fit Index (CFI) value of 0.963, and a Root Mean Square Error of Approximation (RMSEA) value of 0.047. Fantasy (H2: β = 0.422, p < 0.05) had a significant effect on attitude. There was a significant relationship between curiosity and attitude (H3: β = −0.440, p < 0.05) and attitude had a positive relationship with adoption (H7: β = −0.403, p < 0.05). Furthermore, adoption was found to be significantly related to continued use of AR (H8: β = −0.573, p < 0.05). Therefore, H2, H3, H7, and H8 were supported, whereas hypothesis H1 was not supported (see Table 4).
Hypothesised results
| Relationships | Estimate | S.E. | C.R. | P | Hypothesis |
|---|---|---|---|---|---|
| Study 1 | |||||
| ATT ← ENT | 0.049 | 0.068 | 0.719 | 0.472 | H1 |
| ATT ← FAN | 0.422 | 0.068 | 6.181 | *** | H2 |
| ATT ← CUR | −0.440 | 0.074 | −5.965 | *** | H3 |
| ADP ← ATT | 0.403 | 0.045 | 9.054 | *** | H7 |
| CNT ← ADP | 0.573 | 0.083 | 6.87 | *** | H8 |
| Study 2 | |||||
| ATT ← EXP | 0.190 | 0.058 | 3.279 | 0.001 | H4 |
| ATT ← CON | 0.146 | 0.066 | 2.210 | 0.027 | H5 |
| ATT ← SCP | 0.275 | 0.052 | 5.305 | *** | H6 |
| ADP ← ATT | 0.497 | 0.066 | 7.527 | *** | H7 |
| CNT ← ADP | 0.283 | 0.074 | 3.823 | *** | H8 |
SEM was conducted using AMOS 26 to perform hypothesis testing for Study Two. The model fit analysis revealed the following results: CMIN/df = 2.756, GFI = 0.903, AGFI = 0.874, NFI = 0.929, TLI = 0.945, CFI = 0.953, RMSEA = 0.069. Experience (H2: β = 0.190., p < 0.05) and Social Influence (H2: β = 0.275, p < 0.05) had a significant relationship with attitude. It was also observed that attitude had a positive relationship with adoption intention, while adoption intention was significantly related to continued use of AR. Support was found for H4, H5, H6, H7, and H8 (see Table 4).
5.5 Moderation analysis
Process Macro 3.5.3 was used to measure the moderating effect of e-WOM on the relationship between entertainment, fantasy, curiosity, and attitude toward AR. The moderating role of e-WOM was found in the relationship between curiosity and attitude (H9c: β = −0.154, p < 0.05), while no moderating effect was found on H9b and H9a. Process Macro 3.5.3 was used to measure the moderating effect of e-WOM on the relationship between experience, convenience, and social influence on attitude towards AR. The moderating role of e-WOM was found in the relationship between experience and attitude (H9d: β = −0.070, p < 0.05), while no moderating effect was found on H9e and H9f.
6. Discussion
In this study we explored the various factors that impact people’s attitudes toward AR. Based on Study One’s findings, the adoption of AR for online shopping is influenced more by fantasy and curiosity rather than entertainment. Indeed, research has found that imagination is sparked by the wonders of fantasy, enabling users to envision the potential of AR in enhancing the shopping experience (Park, 2017). Conversely, curiosity drives a strong desire to comprehend and delve into seeking information and bridging the gap between AI research and its practical implementation (Sun, 2023).
The findings of Study Two indicate that experience, convenience, and social influence are key factors in driving the adoption of AR for online shopping, as they align with the motivations of users. The study reveals that having experience with AR can create a sense of familiarity and comfort with the technology, reducing apprehension and building trust. This contrasts with previous research, which has shown feelings of insecurity, incompetency, and apprehension among consumers concerning the usage of AI (Moore et al., 2022). Convenience matches the need for a smooth and efficient shopping experience, supporting previous findings (Klaus and Zaichkowsky, 2022). Social influence was found to have a positive relationship with adoption intention. One reason may be that when friends or influencers share their positive experiences with AR for online shopping it enhances social approval among other consumers, which would support previous research (Venkatesh et al., 2022).
This research revealed a strong correlation between a favourable attitude towards AR, its adoption, and the sustained use of AR for online shopping, with this relationship found to be sequential. Positive experiences further strengthen the initial attitude and motivate ongoing usage. Study One discovered that e-WOM plays a role in influencing the connection between curiosity and AR adoption. However, it found no significant impact of e-WOM on fantasy or entertainment. This could be because curiosity naturally drives individuals to seek information and validation.
Based on Study Two, it is suggested that e-WOM plays a significant role in influencing the adoption of AR technology, which is consistent with previous findings (Alam et al., 2024). However, it was found that this influence is primarily moderated by the level of experience rather than convenience or social influence. Positive e-WOM can enhance a positive experience with AR, resulting in a more favourable attitude toward its adoption (Khasawneh and Rabata, 2023).
6.1 Theoretical implications
In this research, U&G theory has been expanded to include the traditional user motivation focus and the sequential stage of AR adoption in online shopping. Traditionally, U&G has focused on users’ media selection to fulfil their particular needs (Gui et al., 2021). This research extends the theory by showing how different gratifications include curiosity, fantasy, and social influence, and the way they shape user attitudes towards AR adoption (Ibáñez-Sánchez et al., 2022). We have also showed how trust and e-WOM moderate some of these associations, thus highlighting the phenomenon of social verification in technology adoption (Hameed et al., 2024). Further, distinguishing the impacts of fantasy and entertainment brings further refinement of U&G theory, as it shows that not all hedonic gratifications are relevant to adoption; rather fantasy engagement is stronger than entertainment engagement.
This study also explored how e-WOM influences the connection between user motivations and AR adoption, and we show how e-WOM has a crucial impact on moderating the relationship between curiosity and adoption. Experience also plays a significant role in determining how e-WOM affects adoption, and within this context, e-WOM serves as a means of validation, where positive e-WOM strengthens positive experiences. This emphasises the need to considering e-WOM throughout various stages of the AR adoption process.
6.2 Practical implications
This study provides valuable insights for retail service providers and technology-based companies. With the rise of AI-based technologies and the growing acceptance of consumers towards them, service providers could analyse the factors that impact consumer adoption of AR. Our findings suggest that it is important to prioritise enhancing the entertainment value and appeal of AR for consumers, as this will encourage them to embrace AR within the retail industry, potentially boosting sales and improving consumer satisfaction and retention.
Furthermore, given the findings that e-WOM plays a moderating role, it is crucial to prioritise the provision of innovative services in a user-friendly manner to cater to consumer needs. Efforts should be focused on providing a delightful shopping experience to consumers through personalised services and improving the sensory experience using AR-based technology to understand customer needs better. It is also important to promote a positive attitude towards AR by raising awareness about its usefulness and value-added features.
7. Limitations and future directions
This research has sought to introduce new elements to the research field, although it has certain limitations. The study examines the use of AR in the retail industry by analysing consumer behaviour only in the Indian market. Therefore, future research should expand the geographical scope explored. Further research should prioritise examining individuals who do not utilise VR for shopping. This would enhance the capacity of service providers to attract individuals who are not yet users of AR in the retail sector, thus expanding the market for AR-based retail stores. Our research has primarily focused on using gratification theory, but it would be beneficial for future scholars to consider incorporating other theories into their studies.
In particular, this research was designed to examine the primary reasons influencing attitudes of consumers regarding the use of AR in retail, zeroing in on the intention to adopt and continue using the technology. As much attention as there is on AR in retail, not much has been given to factors influencing adoption intentions. This research contributes by looking at the impact of e-WOM on the relationship between the aforementioned factors and attitudes towards AR in retail for the first time. We expect the results will be valuable and actionable for scholars as well as practitioners in retail and service businesses, thereby advancing knowledge and practice in the field.
