How do consumers evaluate products in virtual reality? A literature review for a research agenda

Purpose – This paper aims to review the existing literature about consumers ’ evaluation of products in virtual reality (VR), provide an accurate overview of this ﬁ eld, systematise the knowledge developed so far, identifytheresearchgaps andpropose a future research agenda. Design/methodology/approach – A systematic literature review was performed on Scopus and Web of Science, resulting in a ﬁ nalpoolof31 articles. Findings – Four main themes were identi ﬁ ed, and a detailed research agenda is proposed based on the ﬁ ndings and following thetheory, context, characteristics,methodologyframework. Research limitations/implications – The provision of formal inclusion and exclusion criteria may have resulted in additional potentially relevant articles not indexed in the data set under consideration. Originality/value – The paper highlights how products are perceived in VR, the consumers ’ responses, the peculiarities of VR compared to other conditions and VR as a product test environment. To the best of the authors ’ knowledge, this paper seems to represent the ﬁ rst systemic review that focusses solely on how consumers assess products in VR. The results lead to a broad proposal of directions for future research that can expand knowledge on VR in marketing. Practical implications concern the use of VR to design product strategiesandasa testingandprototypingenvironment.


Introduction
The growing adoption of virtual reality (VR) applications among consumers corresponds to increasing attention from academia and practitioners.VR is now a mature and accessible technology with advanced features (Corallo et al., 2020).Immersive VR devices such as headmounted displays (HMDs) are widely available and affordable (Flavi an et al., 2019a), and VR is increasingly positioned as a relevant marketing platform (van Berlo et al., 2021).VR applications are now common in several sectors (Rodrigues and Loureiro, 2022), such as tourism, health care, entertainment, education and real estate.Among the most promising from a marketing perspective are retail and shopping.Several major business players, such as Amazon, Alibaba and eBay, are investing in implementing VR solutions in their commerce system (Zarantonello and Schmitt, 2022;Xi and Hamari, 2021;Martínez-Navarro et al., 2019).VR allows consumers to be immersed in a fully virtual environment characterised by immersion, presence and interactivity (Zarantonello and Schmitt, 2022).
VR shopping environments and marketplaces can realistically simulate physical ones and address the limitations of e-commerce (Luna-Nevarez and McGovern, 2021).VR can enrich the shopping experience with new, exciting and immersive solutions.It is accessible at any time and without the spatial and temporal limitations of the physical world.Firms can address shoppers' multisensory needs, simplifying the decision-making process and engaging them (Mishra et al., 2021).Thus, VR represents one of the technological megatrends that will have a relevant impact on consumers' lives and activities (Xi and Hamari, 2021).The metaverse, of which VR is one of the enabling technologies, has ignited the interest of scholars and practitioners (Dwivedi et al., 2022).
Marketing research is recently experiencing considerable interest in VR, despite its first steps dating back to the 1990s, realigning with other disciplines such as engineering, psychology and computer science, which present a more established contribution (Wedel et al., 2020;Loureiro et al., 2019).Yet, despite the growing attention around VR, it is still unclear how much is known about consumer behaviour in an immersive VR.Marketing literature seems to be still fragmented and scattered, although rapidly evolving, and knowledge should be deepened (Cowan et al., 2021;Hollebeek et al., 2020;Wedel et al., 2020).Particular attention should be placed on consumer assessment of products in VR, and research is needed to better understand the potential of VR in a retail context (Kim et al., 2021).Unlike other marketing applications, such as tourism and hospitality, in VR, consumers are asked not only to experience the environment but also to analyse and evaluate a virtual product.Multisensory experiences, such as those in VR, are contextspecific (Mishra et al., 2021).The evaluation of a tourism destination may be different from a virtual gaming experience.As an example, Flavi an et al. (2021a) focussed on the effectiveness of VR in the hotel industry, which is characterised by the intangibility of the product offering.Orús et al. (2021) claim that the differences in their results compared to previous literature can be attributed to the differences between tangible and intangible industries, such as retailing and hospitality.
VR can also be relevant in prototyping and product evaluation (Hube et al., 2020), as well as in the pre-purchase phase and in stimulating consumer attention towards new products (Hoyer et al., 2020).In research, VR overcomes the limitations of traditional studies.New product ideas can be tested with the possibility of immediately varying each feature and reducing issues about cost, complexity, time, availability and perishability of products.VR decreases the effect of external factors, increases ecological validity and provides more and richer data about users' behaviour (Dwivedi et al., 2022;Lombart et al., 2019;Siegrist et al., 2019).
Within this context, some questions arise: Q1. How do consumers evaluate products in VR?
Q2.What responses, attitudes and behaviours are elicited?
Q3.What are the peculiarities of VR compared with other environments?
And finally: Q4.What do we know and what do we not know about consumers' product evaluations in VR?
Our aim is to understand how products are perceived and evaluated in VR, what consumer attitudes, perceptions and responses emerge during the evaluation of products in VR and what advances the literature has produced so far.This seems to be an emerging theme in Products in virtual reality which knowledge is not yet consolidated and needs systematisation.In line with Wedel et al. (2020), who claim that researchers could revisit consumer concepts and theories in VR, it is necessary to outline the state of the art of literature and identify the opportunities for future research.Despite the greater attention in other marketing areas, such as tourism and hospitality, to the best of our knowledge, there seems to be a lack of reviews addressing consumer evaluation of products in VR.
A preliminary search was conducted to frame the present work.Loureiro et al. (2019) provided a text mining-based review to understand the use of VR in marketing, identifying the use of VR in manufacturing and new product development, among the emerging topics.Xi and Hamari (2021) reviewed 72 articles, conference proceedings and book chapters to understand how VR influences the shopping experience and which technologies have been implemented in the shopping context.Cowan and Ketron (2019) proposed a framework based on user immersion and user experience, whereas Hollebeek et al. (2020) provided a categorisation of VR archetypes.
Based on the above, a literature review can provide a deep understanding of consumer assessment of products in VR and investigate the use of VR in product research in light of consumer perceptions.Accordingly, this paper aims to present the theory, context, characteristics and methodology identified in the reviewed articles, analyse the most relevant topics that emerged and provide a future research agenda.For the present study, a systematic literature review was conducted.Moreover, we adopted the theory, context, characteristics, methodology framework (TCCM; Paul and Rosado-Serrano, 2019) to present the future research agenda.

Virtual reality
In VR, users are immersed in 3D virtual environments where they can navigate and interact, which triggers sensory stimulation (Flavi an et al., 2021a, p. 1).The synthetic world of VR may or may not mimic the properties of a real-world environment (Loureiro et al., 2019, p. 514).Hence, VR is a computer-generated simulation of a situation that incorporates the user, who perceives it via one or more of the senses (currently mostly audio-visual) and interacts with it in a manner that appears to be real (Wedel et al., 2020, p. 443).
VR has specific characteristics that distinguish it from other Extended Reality technologies (Flavi an et al., 2019b).The "Reality-Virtuality Continuum" (Milgram and Kishino, 1994) has been the reference for classification of different realities for many years.Recently, Flavi an et al. (2019b) extended this framework, identifying an independent dimension named pure mixed reality (PMR) and reaching a more comprehensive classification.According to the authors, the continuum ranges between reality and virtuality.At one end is the real environment, where users interact with elements in the real world, while at the other end is the Virtual Environment, which includes Virtual Worlds and VR.Virtual Worlds, such as Second Life, are continuous virtual environments in which, at any time, avatars representing users can interact with other avatars.The term VR has also been unrealistically used (Xi and Hamari, 2021) to refer to different classes of technology, such as Virtual Worlds, the internet or other digital assets and environments.Between these extremes, the physical and virtual worlds coexist and are integrated and mediated by technology.In augmented reality (AR), virtuality overlaps reality, i.e. digital contents are superimposed on the real environment; thus, AR enhances the physical environment (Heller et al., 2021).In augmented virtuality, reality overlaps virtuality, i.e. real contents are superimposed on the virtual environment.In the middle of this continuum is PMR, where virtuality and reality are merged.Users can interact both with digital and real content, which are integrated.All realities are independent of each other and are characterised by specific features.
Based on this classification, Flavi an et al. (2019b) propose the "EPI Cube", a framework that considers the interplay between technological embodiment, perceptual presence and behavioural interactivity, and they suggest a comprehensive taxonomy of existing and new reality-virtuality technologies.The vertices of the EPI Cube represent radical examples of technologies.Applications of interest for this study are those that fall under Vertices 7 and 8 and that refer, respectively, to 360°HMDs videos and VR HMDs.
The framework proposed by Cowan and Ketron (2019) is based on user immersion and user experience.The applications that fall under automated virtual environment, relevant for this study, ensure the highest level of user experience and realism, using HMD and the Cave Automatic Virtual Environment (CAVE).
Based on the above, we refer to VR as a fully immersive environment that offers a sense of embodiment, as users see themselves as components of the virtual environment (Flavi an et al., 2019b, p. 550) in which they can interact with virtual content and objects in real-time.

Systematic literature review
A systematic literature review was conducted (Tranfield et al., 2003;Fink, 2009) in line with the Preferred Reporting Items for Systematic Literature Reviews and Meta-Analyses (PRISMA) guidelines (Page et al., 2021).
Data were obtained from Scopus and Web of Science online databases.Based on previous studies and in line with PRISMA, keywords were selected to build a search item appropriate to identify potentially relevant articles, screening titles, abstracts and keywords.They are presented below.
(produc* OR goods OR commodit* OR merchandis* OR item* OR shop* OR store* OR retail* OR mall OR supermarket) AND ("virtual real*" OR "VR") The first group of keywords covers the word product (and synonyms), as well as shopping and retail, extending Xi and Hamari (2021); "virtual real*" and "VR" include all forms of the words Virtual Reality and their abbreviations.The asterisk allows searches for different suffixes.
The search was then refined to include only articles published in peer-reviewed academic journals.Only scientific articles written in English were included, and other types of publications, such as conference proceedings and book chapters, were excluded.The subject area filter "business, management and accounting" was applied in Scopus, and the category filters "business" and "management" were set in Web of Science to concentrate the analysis only on marketing-related articles.Articles published in any country up to 31 May 2022 and accessible online were considered.Based on these criteria, 521 records were then obtained from Scopus and 198 records from Web of Science.After merging the two raw data sets, all duplicates (113) were removed.Finally, a total of 606 records were obtained.
The article selection process is described below and summarised in Figure 1, adapted from the PRISMA flow diagram (Page et al., 2021).
Starting with the complete data set, an initial screening was carried out.Firstly, those subjects and categories that are not related to marketing, such as logistics, organisation, decision sciences (221), were excluded.Some examples may be found in Supply Chain Management, Information and Organization, Decision Support Systems, International Journal of Project Management.Four records were removed, as they corresponded to other types of publications (book chapters).Successively, in line with Chevtchouk et al. (2021), only journals ranked at least Q1 or Q2 in the SCImago Journal Ranking (2021) were included.

Products in virtual reality
A total of 108 records did not meet this criterion and were excluded.Finally, 273 records were identified to be assessed for eligibility.
The analysis continued by carefully reading the title, abstract, introduction and conclusions of the resulting records.A paper was considered not relevant if it does not deal with VR but with other types of product presentations, such as simple 3 D or it refers to Virtual Worlds (38); or it deals with other reality technologies and formats other than VR, such as AR (27); or it is a conceptual article or a literature review (23); or it deals with nonmarketing topics (such as economics, production management, or VR from a technical perspective), or it focusses on topics other than product-related marketing research, e.g.tourism, hospitality, real estate (154).Every article that fits in at least one of the exclusion criteria was not considered.Each doubtful case was resolved by reading the whole article.An article was considered potentially relevant if it deals with products evaluation or presentation in VR, or analyses consumer behaviour or attitudes related to products in VR, or considers VR as a testing environment involving products to some extent, or compares VR with other experimental conditions in product research.A common point for all selected articles is that one or more products are used in the study.A forward and backward check was performed on all excluded items.Finally, to ensure the validity of the final selection of articles, all articles were read in full.In conclusion, a final pool of 31 articles was obtained.

Descriptive analysis
By looking at the distribution by year, a clear trend in the growth of papers emerges, reaching its peak in 2021 when about half of the contributions were published.Journal of Business Research features the highest number of articles.
Table 1 presents a summary of the theories, contexts, characteristics and methodologies used in the reviewed articles, following the TCCM framework.This framework allows the development of insight clearly and comprehensively, bringing out both theoretical and empirical aspects of a research domain (Castagnoli et al., 2022).It is an effective tool to ensure a comprehensive understating of a given area of research (Roy Bhattacharjee et al., 2022), and it takes into account the breadth of the state of the art (Akhmedova et al., 2021).It is particularly suitable to highlight the knowledge gaps and suggest new directions for future research (Tsiotsou et al., 2022).
First, theories have been reported whereas expressly stated in the articles.As context, we mapped the settings within which the studies were conducted.To this end, we included details about the countries where the research was carried out (if explicit), products used in the studies (in VR or not), the sample involved and the type of VR or other technologies, the format used and possible comparisons with other conditions.With characteristics, we referred to the main constructs, variables and relations that emerged from the article.Finally, the methodology reports the type of studies (not considering pre-tests).

Discussion of results
This section outlines key research streams, relevant topics and contributions of the reviewed articles.Four macro themes were identified.The content of the articles revolved around the fundamental elements of the following topics: (1) products in VR; (2) consumers in VR; (3) VR compared to other conditions; and (4) product tests in VR.
A summary is proposed in Table 2.

Products in virtual reality
VR does not necessarily have to mimic real-world shopping situations to allow for virtual product appraisal (van Berlo et al., 2021).The high interactivity of VR leads to a product evaluation close to physical reality, according to Peukert et al. (2019).The authors consider the resolution and the quality of stimuli to be relevant in product diagnosticity.On the other hand, highly immersive shopping environments seem to positively influence a hedonic path through telepresence but negatively influence a utilitarian path through product diagnosticity.This effect is probably due to a technical limitation of the technology used, which did not allow the information on the packaging to be clearly read.Kang et al. (2020) found that consumers do not seem to place particular importance on the graphic quality when a product is presented in a 3D environment, compared to a 2D display or a 3D website.Probably, consumers in VR are exposed to multiple cues in addition to graphic quality, and they might consider it less relevant when evaluating informativeness.This result could also

Products in virtual reality
be due to the novelty of VR.Interactivity and visual-spatial cues affect the informativeness and playfulness of VR (Kang et al., 2020).Interactivity allows shoppers to freely move and manipulate the product, while a visual-spatial cue refers to the possibility of visualising a full-size product in 3D.
Product type has a significant impact on purchase intentions in a VR environment, and utilitarian products seem to lead to more positive results.Probably the immersive experience of VR makes the purchase less monotonous, leading to greater customer satisfaction and purchase intention (Mishra et al., 2021).Products perceived as an extension of the body (tools such as hammers and pencils) seem to be more suitable for a VR-retail experience (Alzayat and Lee, 2021).These product categories could benefit more from VR than products perceived as a presentation of the body (clothes and accessories).This contrasts with the fact that clothing and accessories have been considered leaders in virtual-based experiences (Alzayat and Lee, 2021).VR can deliver realistic cues for food and drink and elicit natural product-searching processes (Huang et al., 2021;Bign e et al., 2016).
The vivid presentation of a product in VR and the interaction with virtual content can make consumers feel as if they are in a store (Kim et al., 2021).Seeing and interacting with the product in VR did not seem to influence choice satisfaction, according to Meißner et al. (2020), probably because it was not possible to judge the product more accurately.A different result may be obtained when evaluating a product, whereby the reference environment is important or when the use of a product is more complex and requires detailed evaluations.Interacting with the product in VR reduced the colour-flavour incongruency effect on product evaluation, but the same effect was not found for brand perception (Huang and Wan, 2019).Finally, in contexts other than traditional shopping, virtual product appeal strengthens brand attitude but not purchase intention (van Berlo et al., 2021).
The possibility to touch a product is one of the key components of shopping and retail experiences.VR can simulate real products and be a medium to effectively

SJME
haptically explore products (Alzayat and Lee, 2021).The importance of sensory input also depends on product type, and haptic cues have a relevant impact on consumer brand evaluations, even if they are not physical products (Cowan et al., 2021).The need for touch negatively moderates the relationship between VR and hedonic shopping value (Alzayat and Lee, 2021), so individuals with a high need for touch derive less hedonic value in VR-based retail.The vicarious haptic effect is strongest for individuals highly stimulated by an immersive VR experience.Luangrath et al. (2022) found that this effect increases body ownership and, in turn, the psychological ownership of the product and the product evaluation.Ringler et al. (2021) showed how consequential product sound affects consumers' perception of a product and willingness to pay.According to Schnack et al. (2021a), VR could generate sensory cues not comparable to those in a physical store (such as the absence of smell or physical touch), which potentially impacts the way human personality plays in shaping behaviour.The authors did not identify a link between shopper personality and shopping behaviour in an immersive VR store.
Companies in different industries already provide solutions in which consumers can evaluate products in VR.Prada offers the possibility of directly experiencing its product offerings in VR [1], IKEA has long been providing a VR environment where consumers can try out and evaluate [2] different furniture configurations.

Consumers in virtual reality
Consumer attention and purchase conversion rates may be increased in VR shops, triggering impulse buying and packaging inspection (Schnack et al., 2021a).VR also leads to more favourable word-of-mouth recommendations, which may help reduce barriers to VR adoption in businesses (Mishra et al., 2021).Meißner et al. (2020) and Peukert et al. (2019) found that shoppers are less price-sensitive in VR. Brand evaluations and purchase intention are also influenced by product category knowledge, presence induced and haptic input (Cowan et al., 2021). Notably, van Berlo et al. (2021) found that the sense of presence does not influence brand recall.Shoppers also place significant importance on the visual appeal of a product when they experience a highly playful VR environment (Kang et al., 2020).Body ownership and control show a significant relationship with perceived usefulness and, with telepresence, with perceived playfulness of the VR shopping experience (Han et al., 2020).Moreover, perceived usefulness and playfulness lead to the behavioural intention to visit the VR store and purchase products.According to Loureiro et al. (2021b), the vividness and presence of the VR store are increased by consumers' pleasure, triggered by the escapism (as experience); consumer intentions are strengthened as a result.Vividness and presence drive the desire to continue using the VR store and recommend it.Because of pleasure, consumers have a more vivid sensation of presence and a richer representation of the VR store.
Emotions experienced in VR have an impact on the sense of presence and lead to an increase in purchase intention.The affective evaluation of VR acts on brand recall, which also influences purchase intention (Martínez-Navarro et al., 2019).Consumer-brand interactions in VR can evoke emotional responses in terms of both arousal and valence (van Berlo et al., 2021).This drives the perceived hedonic value of the brand, which in turn drives consumers' commercial behaviour.
Consumers in a VR store show similar shopping patterns to real-world behaviour during a multi-category shopping trip (Schnack et al., 2020).Huang et al. (2021) also argue that consumers may exhibit similar shopping behaviours in virtual and brick-and-mortar shops.As Peukert et al. (2019) point out, the effect of immersion on user behaviour and VR shopping acceptance might be more pronounced when users can experience an entire shop.3D VR stores increase consumers' cognitive elaboration and purchase intentions, especially for shoppers that are browsing more than searching (Park and Kim, 2021).However, Kim et al. (2021) claim that VR interactivity is limited when compared to the experience with products in physical stores, while greater than an online platform.Moreover, VR provides more realistic images of products in a 3D naturalistic environment than an online platform.The high degree of vividness can enhance the shopping experience, impacting perceived usefulness and enjoyment.
For example, in the business context, there are solutions for analysing consumer behaviour when evaluating products in VR, as in the case of Kellogg's, for the study of product placement and assortment on the shelves (Accenture, 2019).Pfeiffer et al. (2020) found similarities between physical and virtual supermarkets, calling for research that contrasts VR with physical reality.VR stores can lead to advantages over physical stores regarding positive consumer responses.Martínez-Navarro et al. (2019) compared different VR formats and devices.HMDs are the most effective among VR devices, as they lead to higher levels of presence (compared to powerwalls), but they also generate more discomfort (compared to PCs).However, discomfort perceived in a VR store does not influence the sense of presence or brand recall.This is relevant, as the discomfort was the only advantage attributable to the physical store over the virtual one.3D and 360°S VR do not elicit significant differences in consumers' cognitive and affective responses.Only purchase intentions are affected by the VR format.However, VR stores are more effective in generating cognitive and affective responses.In all VR conditions, brand recall was found to be significantly higher than in physical stores.Finally, VR increases purchase intention compared to physical stores.

Virtual reality compared to other conditions
Consumers seem more open to trying different products in an immersive VR (Meißner et al., 2020), showing stronger variety-seeking, even though they do not seem to be more satisfied with their choices compared to non-immersive VR.Harz et al. (2021) compared two types of VR with a test using real products.Laboratory VR produced more consistent consumer behaviour than online VR and created superior behavioural consistency compared with the real product test because of the sense of presence and vividness.Cowan et al. (2021) showed that 360°VR retail environments are best used online, if retailers aim to influence brand responses, compared to in-store.However, when consumers show high product category knowledge, an online environment in 360°VR gets a worse brand attitude and purchase intention.When haptic information is presented in 360°VR, these negative effect results were attenuated.
According to Kim et al. (2022), consumers experience higher immersion during a VR experience than during a website experience.The flow state results are enhanced, leading to a higher intention to visit the physical store.VR retail positively affects hedonic purchase value compared to an online retail website, but this does not seem to be the case for the utilitarian purchase value.This process is mediated by telepresence, which lets consumers feel as if they are in a simulated retail space (Alzayat and Lee, 2021).However, according to Kang et al. (2020), purchase intention is lower in a 3D VR than in a 3D Web condition.One explanation is that as informativeness is relevant in shaping purchase behaviour if shoppers find the 3D VR shop less informative, they tend to have a lower willingness to purchase.
Comparing VR, AR and mobile apps, Mishra et al. (2021) argue that consumers show higher purchase intentions for a utilitarian product in VR than in apps and a positive experience for a hedonic product in AR compared to apps.Park and Kim (2021) examined the effect of an AR virtual try-on app and a 3D VR shop on clothing purchase intentions.If consumers are in a browsing mode, VR is more effective in increasing purchase intention than AR and static images, whereas VR and AR are both more effective than static images when consumers are in a searching mode.According to Hilken et al. (2021), AR is better suited for improving purchase intentions than VR, whereas VR is more effective in enhancing brand attitude than AR.Moreover, the combination of AR and VR can improve both purchase intentions and brand attitudes, but only if VR is used after AR.This effect seems to be due to an alignment with the online-to-offline journey in experiential retail.Customers simulate procedural action sequences with VR once they have decided on which products to consume with AR.

Product tests in virtual reality
Companies such as BMW [3] and Volvo [4] use VR for product testing and prototyping.Indeed, academic research is also considering VR in this sense.A VR supermarket allows researchers to obtain a highly controlled experimental environment (Pfeiffer et al., 2020) whose immersive and interactive experience allows them to measure both actual brand choice and process variables and to offset the limitations of a field study (Bernritter et al., 2021).VR offers a naturalistic experimental environment suitable for the study of shopper behaviour.Intending to generalise findings to real store settings, VR can provide high external validity (Meißner et al., 2019).

Products in virtual reality
VR laboratory studies produce consistent and realistic consumer behaviour (Harz et al., 2021).A VR experience allows participants to achieve a better and more complete understanding of the product (Naderi et al., 2020).Immersive VR also reduces external variables, such as distractions or lack of realism, allows for controlling confounding effects and generates an immersive, interactive and more realistic experience.VR increases realism and choice immersion, permitting to capture the consumer's product evaluation and the actual brand choice (Esmark Jones et al., 2018;Ketelaar et al., 2017) and allowing the measurement of process variables that are usually hard to assess in field experiments (Ketelaar et al., 2018).VR leads to high experiential control and ecological validity.Also, it opens opportunities that cannot be pursued by traditional techniques (Meißner et al., 2020).When it comes to locomotion techniques in VR shopping environments, instant teleportation and motion-tracked walking are the most used (Schnack et al., 2021b).Although the latter more closely simulates what happens in a real store (but at increased complexity and costs), the absence of physical walking in the instant teleportation does not seem to impact shoppers' emotional states or purchase behaviours.
VR allows easier manipulation of variables and offers an efficient and effective experimental condition.It can be applied earlier in product and packaging design (Huang et al., 2021).It also offers a variety of purchase metrics and benefits that overcome the limitations of traditional methods (Schnack et al., 2020).As reported by Harz et al. (2021), VR laboratory studies provide forecasting advantages when compared to online VR or traditional tests and can also be applied in the early stage of new product development, as companies can use a virtual blueprint for their tests.
VR represents a valuable research tool for conducting consumer behaviour experiments in the food and beverage domain.Lombart et al. (2019) analysed easily perishable products: it is possible to reduce waste and other issues, replicate studies more easily, ensure high control over the product stimuli and guarantee higher validity in the results.In addition, in VR, consumers can interact and manipulate products, as they could not do in other conditions.
As highlighted by Schnack et al. (2021a), any differences identified between physical and VR could be due to differences between subjective and objective research instruments, i.e. the nature of the instruments used to collect data.Many research studies present limitations due, for instance, to the use of self-reported measurement.Combining VR with other technologies overcomes these criticalities and increases the experiments' efficiency.Bernritter et al. (2021)

Emerging future research topics and research agenda
This section presents the key areas for future research based on the TCCM framework.A summary is proposed in Table 3. Uncovered areas of investigation and mixed findings emerge from the analysis, which needs to be addressed to build a solid corpus in this domain.
literature shows that certain dissimilarities between the two conditions subsist (Naderi et al., 2020;Siegrist et al., 2019).Future research could also investigate the effect of navigating and interacting with products in VR without the use of controllers but through motion detection technologies.The consequences of using VR devices for an extended time could also influence consumers' reactions (Mishra et al., 2021).More results on the comparison of VR with other reality technologies and formats, such as AR and PMR, are needed.VR and other reality technologies and formats offer new forms of data (Rauschnabel et al., 2022) useful in understanding consumers from several perspectives (behavioural, physiological, emotional and attitudinal).Finally, VR could also be combined with other technologies to gain a deeper understanding of consumers' product evaluations and complement behavioural data with biometric measures, such as electroencephalogram, eye-tracking, heart rate, skin conductance and haptic feedback (Schnack et al., 2021a;Schnack et al., 2021b;van Berlo et al., 2021;Schnack et al., 2020).

Characteristics: new research directions
Four research directions emerge on this theme.First, future studies should address whether the high level of cognitive engagement during a VR experience can discourage consumers from paying attention to specific stimuli (Martínez-Navarro et al., 2019).
Second, a promising area is the impact of sensory characteristics, such as touch, sound, fragrance and taste, on product evaluation and subsequent purchase decisions (Alzayat and Lee, 2021;Cowan et al., 2021;Harz et al., 2021;Loureiro et al., 2021b;Ringler et al., 2021).Flavi an et al. (2021b) analyse the effect of ambient scent on the VR tourism experience, highlighting the paucity of multisensory studies in the food and beverage domain and the lack of consistency about the effect of the senses in VR experiences.
Third, mixed findings seem to be emerging.Among others, for instance, there seems to be less consensus on consumers' purchase intention (Mishra et al., 2021;Park and Kim, 2021;Kang et al., 2020;Martínez-Navarro et al., 2019) and choice satisfaction (Mishra et al., 2021;Meißner et al., 2020) in VR with respect to other conditions.The importance of the visual and graphical quality of stimuli in VR (Kang et al., 2020;Peukert et al., 2019), as well as differences between the behaviour in a virtual store versus a real one (Kim et al., 2021;Pfeiffer et al., 2020;Schnack et al., 2020), should be also clarified.
Finally, none of the reviewed research considers behavioural analysis in B2B markets.Future research could address this gap, revealing how product evaluation in VR occurs in the business market.

Methodology: new research directions
First, most of the reviewed articles apply the methodology of experiments.The frequency of samples consisting of students and the recourse to laboratory studies could mean that consumer behaviour in VR and the possible differences with the real world are not fully representative.Many studies call for field experiments to increase the external validity and generalisability of results.Given the diffusion of VR among consumers, a viable alternative for analysing actual behaviour could be the use of conjoint analysis (Branca et al., 2023).
Second, as the rate of adherence to the technology improves over time (Mishra et al., 2021), it would be appropriate to conduct longitudinal research to verify any changes in consumer response over time (Loureiro et al., 2021a).
Finally, the use of qualitative studies, such as focus groups and in-depth interviews regarding product assessment in VR, could complement the knowledge obtained from quantitative studies.

Conclusion and limitations
VR is gaining increasing interest in the academic and business worlds and is going to revolutionise the consumer shopping experience (Meißner et al., 2019).To understand consumers' attitudes and behaviours in product evaluation in VR, a systematic literature review was conducted, answering previous calls for insight into marketing-based VR applications and peculiarities in consumers' attitudes and behaviour in VR (Cowan et al., 2021;Hollebeek et al., 2020;Wedel et al., 2020).The article reconstructs the state of the art and identifies several opportunities for future research on an evolving topic, which has produced only the first results of a strongly growing trend.
In terms of theoretical implications, the analysis revealed the main key research streams and relevant topics, resulting in four themes: (1) the study of products in VR; (2) the response of consumers in VR; (3) the comparison of research on products in VR and other experimental conditions; and (4) the peculiarities of using VR as a testing environment for products.
Based on these results and through the application of the TCCM framework, directions for future research agendas have been developed.This study contributes to systematising knowledge on product research in VR, moving from recent conceptual articles (Hollebeek et al., 2020;Cowan and Ketron, 2019;Flavi an et al., 2019b) and addressing recent reviews that analyse VR in marketing from other perspectives.The article extends Loureiro et al. (2019), in which VR in manufacturing and new product development were identified among the applications of VR in marketing, and Xi and Hamari (2021), who focus on the shopping context and experience.Furthermore, it highlights promising areas of investigation that are still not covered, calling for future research and it brings out mixed findings that need to be addressed.
This article also has practical implications, informing managers and practitioners about consumers' product evaluation in VR.The reviewed studies present numerous insights for the construction of effective product offerings and shopping experiences in VR that can overcome the limitations of e-commerce (Luna-Nevarez and McGovern, 2021) and physical reality and engage consumers (Mishra et al., 2021).The use of VR can have several positive impacts.It allows for offering digital and personalised experiences in stores (Cowan et al., 2021), presenting products more playfully than in traditional modes (Kang et al., 2020) and providing multisensory experiences (Mishra et al., 2021).In VR, it is possible to generate better consumer responses than in physical environments (Martínez-Navarro et al., 2019) and stimulate the willingness to try more productsand thus to support the launch of new offers or brandsby reaching consumers who seem less price sensitive in VR (Meißner et al., 2020) and more inclined to impulse purchase (Schnack et al., 2021a).Furthermore, the design, prototyping, development and consumer evaluation of products can be disruptively influenced by this technology (Bu et al., 2021;Huang et al., 2021;Hube et al., 2020).Thus, VR can have an impact even before products reach the actual purchasing process.Conducting studies in VR allows for leaner and cheaper but equally effective processes (Branca et al., 2023).As VR does not have the limitations of physical reality, further advantages include the possibility of expanding the boundaries of traditional experiments, obtaining greater ecological validity, controlling experimental conditions, devising rigorous processes and obtaining a wider and more complex stream of data (Harz et al., 2021;Meißner et al., 2020;Pfeiffer et al., 2020;Schnack et al., 2020;Lombart et al., 2019;Meißner et al., 2019).This article is not without limitations.Given the inclusion and exclusion criteria, there can be relevant contributions not included in this paper.In addition, only high-quality journals were considered.Also, the number of articles considered in the review is not particularly high, given the emergent nature of the topic analysed.Future studies may also include conference proceedings and other types of publications to broaden the knowledge base.Finally, bibliometric analysis or other quantitative methodologies could be used to further explore this research domain.
Figure 1.Procedure for selecting articles to be reviewed

Table 1 .
Products in virtual reality Loureiro et al. (2021b) j Extended S(Stimuli) -O (Organism) -R (Response) framework Portugal j Shoes j 200 students j Oculus Rift HMD Escapism affects consumers' cognitive and affective states that increase pleasure.