The Research Trend of Security and Privacy in Digital Payment

: The aim of this study is to synthesize the rapidly increasing literature on privacy and security risk of digital payment. By reviewing 591 studies, the literature on this topic was evaluated using a bibliographical approach to highlight the intellectual development of the ﬁeld and recommend potential research directions in this still-emerging ﬁeld. According to our assessment, academics have continued to focus on perceived privacy and security, while more multigroup analyses based on subdimensions of risk are needed. In addition, the vast majority of studies have not considered the inter-relationship between risk attributes. We analyse the potential causes of the lack of research diversity and provide additional suggestions to improve digital payment research in the future. This study will be valuable for academics, analysts, regulators, practitioners, and investors.


Introduction
The term "digital payment" refers to a paying technique in which payment data and instructions are transmitted via digital devices, such as a personal digital assistant or cellular phone. Global acceptance of digital payments has been increasing in the last year. The internet's development in recent years has aided in the spread of this payment instrument, as new financial needs have been created by electronic commerce which were unable to be fulfilled effectively by the existing traditional payment systems. The growth can also be attributed to the economic and technological advancements of the internet, the expansion of social networking, the growth of mobile phone users, the expanded usage of electronic money, and the growing service of credit cards [1][2][3]. Due to the pervasiveness of smartphones today, consumers benefit from the simplicity and ease of paying for products and services via this kind of payment channel, which lowers transaction costs. Customers can also access and control their transactions remotely via the web-based user interface.
However, mobile payments are underutilised in several countries, such as China, India, Singapore, and Malaysia, where consumers choose to pay for goods and services with traditional methods [2]. According to Schierz et al. [4], only one percent of mobile phone users have used a mobile payment system. This lack of adoption could be attributed to security and privacy concerns among users. Risks of digital payment could be linked to service risk, device risk, network risk, and platform risk. For example, it has been suggested that the risk perceived among consumers of online shopping is one of the main factors that hinder its development [5]. Although digital payment services have become established and widely used in many nations, including Japan, Singapore, and Korea, digital payments are still in their infancy on a global scale, especially in Europe [6].
Even in smart cities, the adoption of such innovative technology is still low. For example, Kumar et al. [7] reported that in the smart city Coimbatore, only 74.3% of the population uses mobile payments. The slow or low adoption of digital payment technologies could be linked to a variety of factors, including security considerations, privacy issues, usage barriers, value barriers [8], a lack of security, system complexity, and privacy concerns [9,10]. In addition, Kartika et al. [11] argued that in smart cities, it is important to improve the security and confidentiality of information in noncash financial transactions in order to promote such services. Yang et al. [12] observed that due to regulatory shortcomings in data protection, customers find it increasingly difficult to enforce their rights in e-commerce. Customers' privacy and financial worries are heightened as a result of regulatory ambiguity around mobile payments. El Haddad et al. [13] suggested that despite technology improvements, e-commerce still faces a considerable concern in terms of trust and perceived risk and their relationships with user satisfaction.
Much research has examined digital payment from a technical and user acceptance standpoint. Researchers have a diverse range of interests, including the antecedents and determinants of users' satisfaction [14,15], network operators [16], consumer acceptance [17,18], continued use behaviour [17,18], and stakeholders' expectations [19] of a variety of services, such as mobile payment, quick response code payment [20], mobile and electronic wallets [21,22], internet and mobile banking [23,24], and digital payment [25]. The study indicates that the number of authors and publications on digital payments has expanded over the past two decades. Despite the vast number of studies on digital payment variability, research on the antecedents of digital payment has produced conflicting findings [2]. It has been suggested that additional, in-depth studies on the adoption process of these tools are essential, as is active monitoring of the effects of various financial solutions on customers' perceptions and daily lives [3]. As such, it is critical to conduct a review of the existing body of knowledge.
Despite the growing number of studies on digital payments, there seems to be a scarcity of peer-reviewed research on risk perception of digital payment. A review of the existing studies on digital payment proves the presence of a variety of previously explored research themes, and the prior review studies either focus on specific aspects of digital payment, such as customer adoption behaviour [2], digital payment utilisation persistence [26], and mobile payment adoption [27]; or a specific context, such as golf counties [28], Sub-Saharan Africa [29], and Thailand [30]. Thus, this study provides an overview of the available literature and contributes to the field's enrichment and the development of future research areas. To accomplish this, a bibliometric approach is used to analyse the present state of the literature on digital payment security and privacy. This offers a thorough synthesis of the literature in this field and discusses future research directions and consequences for digital payment providers and policymakers.
Based on all of the above, the rest of the article is organised as follows. First, the research method and sample literature collection are presented. This is followed by the analysis, findings, and future research discussions. The research concludes with the limitations of our study in Section 5.

Materials and Methods
We thoroughly analysed digital payment literature, encompassing articles published over 22 years. The field was mapped using a systematic review that combined qualitative and quantitative methodologies [31,32]. Systematic reviews are a frequently used technique for organizing and synthesising research results. They are especially beneficial when dealing with vast and complicated research bodies, such as those in digital payment.
While narrative approaches may be beneficial, they have been criticised for their high degree of subjectivity and lack of generalizability [33,34], while systematic reviews have defined methodologies for conducting a complete literature review. The fundamental ideas of systematic reviews are as follows: specific objectives, reproducibility, a wide and comprehensive search based on merit, hence minimising reviewer bias, and the incorporation of a synthesised technique to organise the literature [35][36][37][38][39].
The systematic review approach utilised in this research consisted of five steps: (1) identification of keywords and terms, (2) identification of articles, (3) quality evaluation, (4) data extraction, and (5) data synthesis. The data mining keywords were chosen based on comparable evaluations on behavioural finance [40]. Given that the purpose of this study is to perform a systematic evaluation of security and privacy issues in digital payment, phrases such as "digital payment", "Electronic payment", "Mobile payment", and "Security" OR "Privacy" were used (see Figure 1). To discover all published research in the field, data mining was performed using the Scopus database. This database is often regarded as the most comprehensive, including research from a wide variety of subjects compared to other databases (i.e., Web of Science). Massaro et al. [41] suggested that Scopus is one of the largest abstract and citation databases of peer-reviewed literature. Rasel and Win [42] used Scopus database because of its broader coverage of relevant and quality publications. This database allows a search for publications with prespecified keywords, for example, in the title, abstracts, or keywords. At this point, the sample size was 2691 papers. ideas of systematic reviews are as follows: specific objectives, reproducibility, a wide and comprehensive search based on merit, hence minimising reviewer bias, and the incorporation of a synthesised technique to organise the literature [35][36][37][38][39].
The systematic review approach utilised in this research consisted of five steps: (1) identification of keywords and terms, (2) identification of articles, (3) quality evaluation, (4) data extraction, and (5) data synthesis. The data mining keywords were chosen based on comparable evaluations on behavioural finance [40]. Given that the purpose of this study is to perform a systematic evaluation of security and privacy issues in digital payment, phrases such as "digital payment", "Electronic payment", "Mobile payment", and "Security" OR "Privacy" were used (see Figure 1). To discover all published research in the field, data mining was performed using the Scopus database. This database is often regarded as the most comprehensive, including research from a wide variety of subjects compared to other databases (i.e., Web of Science). Massaro et al. [41] suggested that Scopus is one of the largest abstract and citation databases of peer-reviewed literature. Rasel and Win [42] used Scopus database because of its broader coverage of relevant and quality publications. This database allows a search for publications with prespecified keywords, for example, in the title, abstracts, or keywords. At this point, the sample size was 2691 papers. At this stage, any studies that were deemed irrelevant were removed from the original sample. By focusing on certain fields, namely, business management, finance, economics, environment, and social science, the number of sample studies decreased to 940 studies that fell within these fields. Titles and abstracts of these studies were screened for irrelevant research. Only studies that clearly addressed digital payment and security or privacy risk were included. Final sample included in this study is 591 documents. The sample literature was evaluated following an approach that is widely applied in similar review research [31,43,44]. The methods used in the analytical structure of this study are presented in Figure 2. At this stage, any studies that were deemed irrelevant were removed from the original sample. By focusing on certain fields, namely, business management, finance, economics, environment, and social science, the number of sample studies decreased to 940 studies that fell within these fields. Titles and abstracts of these studies were screened for irrelevant research. Only studies that clearly addressed digital payment and security or privacy risk were included. Final sample included in this study is 591 documents. The sample literature was evaluated following an approach that is widely applied in similar review research [31,43,44]. The methods used in the analytical structure of this study are presented in Figure 2.

Results and Discussion
To determine the basic continuous pattern of publication on this topic, a descriptive analysis of 591 publications was conducted. The publication trend was examined in terms of total publications by journal, country, area, and year. The yearly pattern of articles shows that the interest in studying this subject has increased over the last decade as a result of technological advancements that enable more dependable, user-friendly, adaptable, and functionally rich mobile payment systems. In our sampled literature, the earliest document goes all the way back to 1989, when Bürk and Pfitzmann [45] compared the security and degrees of unobservability of different digital payment systems. Indeed, before 2016, there was limited published research, with an average of less than 10 studies each year ( Figure 3). Following this, the amount of research investigating the security and privacy of digital payment steadily increased each year. The year 2021 had the most publications, with 128 articles (about 22% of publications on this topic).

Results and Discussion
To determine the basic continuous pattern of publication on this topic, a descriptive analysis of 591 publications was conducted. The publication trend was examined in terms of total publications by journal, country, area, and year. The yearly pattern of articles shows that the interest in studying this subject has increased over the last decade as a result of technological advancements that enable more dependable, user-friendly, adaptable, and functionally rich mobile payment systems. In our sampled literature, the earliest document goes all the way back to 1989, when Bürk and Pfitzmann [45] compared the security and degrees of unobservability of different digital payment systems. Indeed, before 2016, there was limited published research, with an average of less than 10 studies each year ( Figure 3). Following this, the amount of research investigating the security and privacy of digital payment steadily increased each year. The year 2021 had the most publications, with 128 articles (about 22% of publications on this topic).  Table 1 summarises the contribution of existing digital payment literature by subject area (coauthored papers have been considered at multiple geographic locations). The sample studies were conducted in 08 different countries, with 80 countries contributing less than five studies and 91 countries contributing just one study. The table lists countries with more than five documents, the average year of publication, and the total citations. The results reveal that developed countries, such as China, India, and the United States, contributed the most. The average year of publication for these countries was before 2015-2018, indicating that they were the drivers of interest in investigating this topic. The growing interest in these markets was due to the continuous development of digital payment applications. For example, the development of facial recognition payment in China has led the world. However, despite the economic size of these countries and the existence of various types of digital payments, they remain mostly cash-based markets. This triggers scholars' interest in better understanding the security and privacy concerns related to digital payment technology. It should be noted that China has contributed the most to the prevalent literature on digital payment, while African countries have contributed the least. Additionally, further work may concentrate on comparing samples from various nations based on customers' perceptions of the adoption of various payment approaches [21].   Table 1 summarises the contribution of existing digital payment literature by subject area (coauthored papers have been considered at multiple geographic locations). The sample studies were conducted in 80 different countries, with 48 countries contributing less than five studies and 19 countries contributing just one study. The table lists countries with more than five documents, the average year of publication, and the total citations. The results reveal that developed countries, such as China, India, and the United States, contributed the most. The average year of publication for these countries was before 2015-2018, indicating that they were the drivers of interest in investigating this topic. The growing interest in these markets was due to the continuous development of digital payment applications. For example, the development of facial recognition payment in China has led the world. However, despite the economic size of these countries and the existence of various types of digital payments, they remain mostly cash-based markets. This triggers scholars' interest in better understanding the security and privacy concerns related to digital payment technology. It should be noted that China has contributed the most to the prevalent literature on digital payment, while African countries have contributed the least. Additionally, further work may concentrate on comparing samples from various nations based on customers' perceptions of the adoption of various payment approaches [21].

Leading Journals
Between 1982 and 2021, 591 articles appeared in approximately 370 journals and conference proceedings. However, 353 of them had fewer than five publications, while 276 had only one. Two aspects of each journal were analysed: first, the number of publications and citations, and the average year of publication; and second, the journals' quality as defined by the Academic Journal Guidance (AJG) 2021. Table 2   The AJG 2021 was also used to assess the studies' quality. It assigns a quality rating to business and management journals and ranks them as 1, 2, 3, 4, and 4*. Four with asterisk (4*) represents the highest-quality journal, while 1 represents the lowest. The AJG rating is a critical factor in researchers' advancement in management and finance areas, and it is frequently used by researchers [31,43]. The result in Table 3 indicates that the overwhelming majority of studies on digital payment privacy and security has been published in notranked journals (401 studies). Surprisingly, only two studies were published in Grade 4* journals out of 591 documents. As shown in the table, scholars have been more interested in Grades 1 and 2 journal outlets.

Most-Cited Studies
In this section, the citation trends of the sample literature were analysed. Citation analysis is a widely used technique for determining the significance of existing literature. Citation analysis determines the popularity of an individual publication in the body of existing literature based on the number of citations of that publication from other published studies. The initial citation analysis revealed that out of 591 documents, 134 studies had more than ten citations and 39 studies had over 50 citations. As shown in Table 4, we found that the most popular studies within sample literature had more than 150 global citations, including Schierz et al. [4] and Mallat [46], who were cited by the majority of researchers in this field. Citation analysis demonstrates the overall thematic trend in research on this topic. Table 4. The most influential studies based on citation number.

Author/s Name Year Cited By Summary of Findings Source Title
Schierz et al. [4] 2010 594 The impacts of subjective norm, individual mobility, and compatibility are all strongly supported by the empirical findings. The impact of security is well-documented.
Mallat [46] 2007 494 The advantages of mobile payments differ from what adoption theories suggest and include time and location independence, availability, remote payment options, a lack of critical mass, and queue avoidance. Perceived risk, complexity, and premium pricing are among the main hurdles to adoption highlighted.

J. Strategic Inform Syst.
Dahlberg et al. [47] 2008 442 This study presents a framework consisting of four contingency and five competitive force variables and arranges mobile payment research around each. Contemporary studies best cover consumers' perspectives on mobile payments as well as technical security and trust. The effects of social and cultural variables on mobile payments and comparisons between mobile and traditional payment services are all topics that have yet to be researched.
Slade et al. [48] 2015 320 The nonusers' intentions to embrace remote mobile payments are highly influenced by performance expectancy, social influence, innovativeness, and perceived risk, but not by effort expectancy.
Kim, Tao, et al. [49] 2010 273 A conceptual model that identifies the factors that influence consumers' perceptions of security and trust, as well as the impact of these factors on the adoption of e-payment systems, is proposed.
Thakur and Srivastava [50] 2014 265 Privacy risk and security risk are found to be significant subdimensions of perceived risk. Internet Res. Au and Kauffman [51] 2008 257 This study examines a new use of technology that is gaining traction globally in conjunction with the wireless revolution: mobile payments. Although this technology application is likely to have complexities and surprises, we urge the reader to keep in mind that many of the same economic dynamics will be at play as they have been in the past with other financial services and associated technology applications.

Electronic Commerce Research and Applications
Morosan and DeFranco [52] 2016 228 Performance expectancy was found to be the strongest predictor of intentions to use near-field communication mobile payments, followed by hedonic motives, habit, and social factors. There are a number of key consequences for academics and industrial decision-makers.
De Kerviler et al. [53] 2016 174 Social benefits and hedonic, utilitarian, financial, and privacy threats are major drivers from the perspective of perceived value. The authors also look at the distinctions between the drivers of more common mobile buying behaviours and emphasize the importance of experience.

Von Solms and
Naccache [54] 1992 169 Blind signatures appear to be an ideal answer in light of the increased emphasis on protecting the privacy of user data and actions in electronic systems. This research, on the other hand, looks at a flaw in blind signatures, demonstrating how a perfect solution can lead to a perfect crime.

Comput. Secur.
Yang et al. [12] 2015 160 The main determinants of perceived risk are confirmed to be perceived service intangibility, perceived regulatory uncertainty, perceived technology uncertainty, and perceived information asymmetry, while perceived privacy risk, perceived financial risk, and perceived performance risk were found to have strong negative effects on acceptance intention and perceived value.
Slade et al. [55] 2015 158 The extended model explains more variance in behavioural intention, but performance expectancy remains the best predictor across both models.

J. Strateg. Mark.
After examining these articles, we reach the conclusion that the thematic assessment that has developed in the field of digital payment research is narrow and focuses on perceived privacy and perceived security risks as determinants of consumer adoption of mobile payments [4,46]. It has been stated that privacy risk and security are significant subdimensions of the overall perceived risk toward digital payment. In this regard, a major gap is found to be the scarcity of studies on the antecedents of digital payment security and privacy risks.

Co-Citation Analysis
Bibliographic coupling is defined as the connection between two referencing documents when they refer to the same work. When two works share a large number of citations, this indicates a strong coupling. Additionally, bibliographic coupling denotes the subject matters' similarity. If the list of references contains commonly cited research, the two authors are bibliographically associated. While there are several techniques for reviewing the intellectual structure of a field, such as co-citation and citation assessment, these methods fall short of identifying emerging themes [31,43]. Bibliometric coupling overcomes this constraint by identifying contemporary topics within a domain. The pres-ence of two publications in an article's reference list more than once can indicate a degree of similarity in the theory, methodology, or empirical discipline of each. The link strength between two documents, supplied by VoSviewer, was used to quantify the connectivity between pair references. Van Eck and Waltman [56] suggested that this metric quantifies the strength of the connection between each pair of connected references.
The co-citation analysis revealed that there are 166 pairs of documents that have been cited together at least ten times and only 105 pairs that have been cited together more than twenty times.
As illustrated in Figure 4, the strongest co-citation relationship exists between Chen and Yang [57] and Davis et al. [58]; the link strength between these publications is 27. The former introduced the planned behaviour theory, while the latter applied it in understanding the antecedents of digital payment adoption, including perceived risk and security. As shown in the figure, the second most powerful co-citation relationship exists between Davis et al. [58] and Venkatesh et al. [59], followed by Schierz et al. [4] and Venkatesh et al. [58]. These studies mainly evaluated factors influencing the intention to utilize mobile payment [60][61][62] as well as a mobile wallet [63,64], based on the unified theory of acceptance and use of technology (UTAUT) [24,48], the technology acceptance model (TAM) [65,66], and the planned behaviour theory [53,67]. The co-citation analysis revealed that there are 166 pairs of documents that have been cited together at least ten times and only 105 pairs that have been cited together more than twenty times.
As illustrated in Figure 4, the strongest co-citation relationship exists between Chen and Yang [57] and Davis et al. [58]; the link strength between these publications is 27. The former introduced the planned behaviour theory, while the latter applied it in understanding the antecedents of digital payment adoption, including perceived risk and security. As shown in the figure, the second most powerful co-citation relationship exists between Davis et al. [58] and Venkatesh et al. [59], followed by Schierz et al. [4] and Venkatesh et al. [58]. These studies mainly evaluated factors influencing the intention to utilize mobile payment [60][61][62] as well as a mobile wallet [63,64], based on the unified theory of acceptance and use of technology (UTAUT) [24,48], the technology acceptance model (TAM) [65,66], and the planned behaviour theory [53,67].

Co-Citation Network
We analysed the evolution of clusters over time to improve our comprehension of the evolution of studies on digital payment. The evolution of clusters demonstrates the emergence and development of research subthemes in the area. Figure 5 demonstrates the number of articles published in each cluster over time. It can be deduced that earlier research was more concentrated on cluster 1 (red). We examined the references of prior research (591 studies) and discovered that the sample literature cited 23,116 unique references. The number of articles cited more than once locally was 1304, and 47 studies were cited more than ten times. The term "locally cited" refers to the number of times a reference was cited within our sample of literature. This investigation exemplifies the thematic trend in research on the security and privacy of digital payment.

Co-Citation Network
We analysed the evolution of clusters over time to improve our comprehension of the evolution of studies on digital payment. The evolution of clusters demonstrates the emergence and development of research subthemes in the area. Figure 5 demonstrates the number of articles published in each cluster over time. It can be deduced that earlier research was more concentrated on cluster 1 (red). We examined the references of prior research (591 studies) and discovered that the sample literature cited 23,116 unique references. The number of articles cited more than once locally was 1304, and 47 studies were cited more than ten times. The term "locally cited" refers to the number of times a reference was cited within our sample of literature. This investigation exemplifies the thematic trend in research on the security and privacy of digital payment. Following an assessment of these works, we conclude that the theme series that has emerged in the area of digital payment literature is rather small, focusing on the UTAUT and the TAM developed by [58,59,68]. The research developed from the TAM was introduced by Davis et al. [58]. In this regard, it was found that there is a scarcity of theorical lenses on the cultural aspect of digital payment users. As a result, we argue that the ontological and theoretical frameworks in the current literature do not adequately address the intricacies and complexity of digital payment. The full map was too complicated to be understood; therefore, we followed Khatib et al. [31] and limited the map to the most influential studies in each cluster based on the average number of citations. More in-depth evaluation of these clusters is presented in the following sections.

Cluster 1 (Red)
As illustrated in Table 5, this cluster is the largest within the digital payment community, with studies focusing on two primary areas: the technology acceptance model (TAM) [58,69,70] and structural equation modelling [71][72][73]. The TAM illustrates the relationship between perceived ease of use, attitudes, and perceived usefulness toward technological innovation and acceptance behaviour [58,69]. This model received widespread support from the literature. Pavlou [74] argued in favour of the TAM as a way to increase consumer acceptance of electronic commerce. Additionally, the TAM was enhanced by incorporating subjective norms, individual mobility, and compatibility, which were statistically significant in the model [4]. Luarn and Lin [18] provided compelling evidence that the extended TAM is effective at forecasting users' intentions to adopt mobile banking. Researchers advanced the technology acceptance model by introducing TAM2 [59] and TAM3 [70]. It has been suggested that the TAM explained only 40-60% of customer behaviour intention, leaving nearly half of the relative factors unaccounted for. Following an assessment of these works, we conclude that the theme series that has emerged in the area of digital payment literature is rather small, focusing on the UTAUT and the TAM developed by [58,59,68]. The research developed from the TAM was introduced by Davis et al. [58]. In this regard, it was found that there is a scarcity of theorical lenses on the cultural aspect of digital payment users. As a result, we argue that the ontological and theoretical frameworks in the current literature do not adequately address the intricacies and complexity of digital payment. The full map was too complicated to be understood; therefore, we followed Khatib et al. [31] and limited the map to the most influential studies in each cluster based on the average number of citations. More in-depth evaluation of these clusters is presented in the following sections.

Cluster 1 (Red)
As illustrated in Table 5, this cluster is the largest within the digital payment community, with studies focusing on two primary areas: the technology acceptance model (TAM) [58,69,70] and structural equation modelling [71][72][73]. The TAM illustrates the relationship between perceived ease of use, attitudes, and perceived usefulness toward technological innovation and acceptance behaviour [58,69]. This model received widespread support from the literature. Pavlou [74] argued in favour of the TAM as a way to increase consumer acceptance of electronic commerce. Additionally, the TAM was enhanced by incorporating subjective norms, individual mobility, and compatibility, which were statistically significant in the model [4]. Luarn and Lin [18] provided compelling evidence that the extended TAM is effective at forecasting users' intentions to adopt mobile banking. Researchers advanced the technology acceptance model by introducing TAM2 [59] and TAM3 [70]. It has been suggested that the TAM explained only 40-60% of customer behaviour intention, leaving nearly half of the relative factors unaccounted for. Correlations between usefulness and behaviour are much stronger than those between ease of use and behaviour. Perceived ease of use may be a causal antecedent of perceived usefulness.
[77] 3185 E-service adoption Adoption of e-services is hurt mostly by performance-related risk perceptions, and perceived ease of use of the e-service reduced these risk worries.
[78] 997 The initial trust in mobile banking and intention to use The firm's overall reputation was insufficient to persuade customers to use mobile banking. The proportional benefits, trust proclivity, and structural guarantees have a significant impact on early trust in mobile banking. [18] 2516 Factors determining users' acceptance of mobile banking The data demonstrate that the extended TAM is highly predictive in anticipating customers' intentions to utilise mobile banking. [79] 11,821 The adoption of information technology innovation The study creates a tool for assessing an individual's different viewpoints for accepting a breakthrough in information technology. By concentrating on the characteristics that are most likely to affect system usage via both design and implementation tactics, the deconstructed theory of planned behaviour gives a more comprehensive account of behavioural intention.
[70] 6805 Extending of technology acceptance model into TAM3 The initiative built a comprehensive nomological network for information technology uptake and use of TAM3.
[59] 23,706 Extending the boundaries of the technology acceptance model into TAM2 Both cognitive instrumental processes (reported ease of use, demonstrability of results, output quality, and work relevance) and social influence processes significantly affect user approval (image, voluntariness, and subjective norm).
However, the TAM focuses on the positive aspects of a consumer's perspective when it comes to adopting new technology, and there are several factors (i.e., emotional, external, or environmental factors) ignored by this model and its extensions. Legris et al. [81] claimed that earlier research on this model is not entirely consistent and clear, and that crucial components are not included in the model; they must be incorporated into a bigger one that includes variables relating to both individual and social transformation processes. Similarly, Chuttur [82] concluded that the model lacks sufficient and rigorous research and that, despite its widespread use, there is disagreement about its theoretical assumptions and practical usefulness. Other user qualities (emotional, cognitive, and demographic elements) have been proposed to be taken into account as well, since they may function as a moderator of the link between TAM variables and technology acceptance.

Cluster 2 (Green)
Cluster 2 addresses the theme of digital payment adoption from a planning and behaviour perspective introduced by Ajzen [57], who proposed that perceived behavioural control and attitudes toward the behaviour subjective norms can predict intentions to perform different behaviours and that these intentions, in combination with perceptions of behavioural control, account for a substantial amount of variation in actual behaviour. As presented in Table 6, studies in this cluster have primarily evaluated the trust and risk that formulated the attitude of technology users addressed in the planned behaviour theory. Based on the framework of this theory, Venkatesh et al. [83] developed the unified theory of acceptance and use of technology model; it comprises four primary drivers of intention and usage and four important relationship moderators. This trend of studies highlights the same gap in the literature: customers' attitudes toward digital payment may be significantly influenced by individual cultural differences. The UTAUT2 was introduced by extending the unified theory of acceptance and use of technology [84], while other scholars evaluated the consumer acceptance of mobile wallets [64], mobile payment [50,85], and mobile banking [23] to predict the acceptance, adoption, and use of information technologies. The extensions presented in UTAUT2 resulted in a significant increase in the variation explained by behavioural intention (from 56% to 74%) and technological usage (from 40% to 5%).
[ 88] 1392 The drivers of intention to use mobile payment Compatibility with existing payment systems is not a significant factor in users' choice to accept it. Perceived simplicity of use and perceived usefulness are significant predictors of intention to utilise m-payment.
[89] 418 To assess the relative significance of several elements in the adoption of a new system of mobile payment The user's age introduces significant changes in the proposed links between third-party effects and the payment system's ease of use, between perceived trust in the system and its ease of use, and between perceived trust and a favourable attitude toward the payment system's use. [90] 219 The acceptance of mobile payment in virtual social networks The suggested behavioural model was changed accordingly, demonstrating that prior experience improves the likelihood of use.
[91] 828 The determinants of customer adoption and intention to recommend mobile payment Social influence, innovativeness, performance expectations, perceived technical security, and compatibility are all expected to have a major indirect and direct impact on mobile payment acceptance and the intention to suggest these technologies. [55] 294 The possibility of a new customer technology adoption paradigm, as well as its extension with trust and risk frameworks Although the extended model explains a greater proportion of the variance in behavioural intention, performance expectancy remains the greatest predictor in both models.
[50] 467 To investigate the functional link between mobile payment usage intention, perceived risk, and adoption readiness When the proposed model was evaluated, five of the six hypotheses were found to be fully supported, while one was found to be moderately supported. The invariance test revealed significant variation between users and nonusers.
[84] 9158 UTAUT2 In comparison to UTAUT, the extensions offered in UTAUT2 resulted in a significant increase in the variation explained by behavioural intention (from 56% to 74%) as well as technological use (40 percent to 52).
[85] 703 The drivers of mobile payment adoption While personal characteristics, social influence, and behavioural beliefs all play a role in determining mobile payment service acceptance and use, their effects on behavioural intention vary throughout stages. How diverse uncertainty leads to distinct perceived risk dimensions, which impede mobile payment adoption Perceived service intangibility, perceived regulatory uncertainty, perceived information asymmetry, and perceived technological uncertainty have all been confirmed as significant predictors of perceived risks, whereas perceived privacy risk, perceived financial risk, and perceived performance risk have all been shown to have a significant negative impact on perceived value and acceptance intention.
[92] 841 Continue to use mobile payment The primary factor determining trust is the quality of the service, but the primary factor affecting satisfaction is the quality of the system. The quality of information and services has an effect on flow. Trust, flow, and contentment all contribute to the intention of mobile payment users to continue using it.
[23] 1592 UTAUT and task technology fit model Social influence, task technology fit, and performance expectations all have a substantial impact on user adoption. The match of task technology with performance expectations has a substantial effect.
Similar to the findings of the cluster 1 studies, researchers in this cluster concentrated on customer acceptance of digital payments, while paying less attention to merchant adoption. Nonetheless, consumer adoption (as defined by the aforementioned models) has remained a highly-researched area of study. However, the publications' findings on certain variables remain equivocal. According to Oliveira et al. [91], all factors of social impact, innovativeness, performance expectations, perceived technology security, and compatibility exert a substantial indirect and direct effect on mobile payment uptake and the desire to suggest this technology. However, in a ground-breaking study conducted by Kim et al. [88], all respondents said that compatibility was not the key factor in their decision to use mobile payment. Additionally, reputation of the platform and firm is a strong predictor of customer trust in digital payments [93], although Kim et al. [78] observed that reputation as a corporate trait did not attract individuals to mobile banking. This inconsistency may be explained by the fact that the needs and expectations of adopters differ significantly across user groups. As a result, scholars must strive to balance the varied interests of various groups of individuals. Hence, service providers can tailor their offerings to fit the task requirements of various groups, hence increasing user adoption of technology innovation [23]. Furthermore, adoption models may need to be adjusted to account for country-specific differences [94]. Kim et al. [78] reported that an individual's disposition to trust is shaped by their cultural upbringing. Countries' successes or failures in adopting mobile payments vary according to their infrastructure, urban-rural disparities, and regulatory synthesis between the banking and telecommunications sectors.

Cluster 3 (Blue)
Cluster 3 is the smallest one in the sample literature that takes the research a step further by focusing on security and privacy risks [10,[95][96][97]. Customer privacy and security are essential factors in digital payment applications which aim to give customers confidence in the application company by maintaining privacy and security on every element of the customer. Consumers face substantial vulnerabilities in online settings, including identity theft and information exploitation. For example, businesses may utilise consumers' data for marketing reasons without their agreement or even disclosure. This is unsurprising, given that a person's online distinctive and valued identity is comprised of a combination of financial and personal data. Risk perception is a significant predictor of acceptance of novel technologies. It has commonly held in the literature that perceived security and privacy are both expected to have a major indirect and direct effect on mobile payment acceptance and the intention to use this technology [91,95,[97][98][99]. Khalilzadeh et al. [100] argued that risk and security have the most substantial impact on the behavioural intention of customers toward digital payments. Luo et al. [96] found that eight different risk facets, including privacy and security, are salient antecedents to innovative technology acceptance. However, short message service, rapid response code, magnetic security transmission, and near-field communication systems can be employed for digital payment. The question remains whether risk perception toward different payment systems are different. For example, Ghezzi et al. [101] suggested that mobile payment solutions must deal with more privacy and security issues than in e-commerce and electronic payment.
Regarding security and privacy, as shown in Figure 6, various research works have approached it from the behavioural aspect of customers and it has been found to have significant impact (direct/indirect) on intentions to use and recommend digital technologies [91,102]. As past studies have demonstrated, trust can thrive and is necessary via the internet while dealing with an unpredictable and unsafe environment. In retail situations, trust between the trustor (a customer) and the trustee (a seller) is a significant incentive, since it improves repurchase intention and minimises perceived risk. In general, users are comfortable sharing generic and nonspecific information such as preferences but are less comfortable supplying sensitive information, such as account or credit card details. Additionally, whether or not clients transfer personal and financial data online is contingent upon their faith in a particular online supplier. As summarised in Table 7, researchers in this cluster have examined a varie sues affecting mobile payment service markets [47,101], developing frameworks t into account all players [51]. Academia appears to be restricted in terms of what logical benefits it can provide practitioners, except firms engage academic researc gressively. According to the report, confidence in a payment vendor, service p institutional context, and technology component has a significant role in forming services [76,103]. Diverse payment scenarios previously required physical tok As summarised in Table 7, researchers in this cluster have examined a variety of issues affecting mobile payment service markets [47,101], developing frameworks that take into account all players [51]. Academia appears to be restricted in terms of what technological benefits it can provide practitioners, except firms engage academic researchers aggressively. According to the report, confidence in a payment vendor, service provider, institutional context, and technology component has a significant role in forming trust in services [76,103]. Diverse payment scenarios previously required physical tokens and their combinations, network traffic algorithms based on software, trust, and creation of various securities. This is perhaps the most fertile ground for collaboration and academic study. Lu et al. [103] established a model of client decision making based on trust in the setting of nonindependent third-party mobile payment services. Table 7. Summary of studies in cluster 3.

Author/s # GS Citations
Focus of the Study Summary of the Findings [47] 1060 Mobile payment The paper provides a paradigm comprised of four unforeseen and five competing force elements.
[101] 101 The drivers of mobile payment applications The case studies aided in the comprehension of the primary diffusion drivers: Despite the numerous benefits associated with these services, severe inhibitory factors and adoption barriers continue to limit user uptake.
[103] 698 The usage intention of mobile payment Trust, in conjunction with positive and negative valence variables, has an effect on behavioural intention both directly and indirectly. These effects on employees and students have considerably varying magnitudes.
[ [95] 110 The use of m-wallets Perceived utility and perceived simplicity of use have a substantial effect on user satisfaction and desire to use m-wallets in the future. Perceived security has a considerable influence on customer happiness, while grievance resolution mitigates the influence of perceived security on the desire to continue using m-wallets.

Author/s # GS Citations
Focus of the Study Summary of the Findings [97] 1352 The usage intention of internet banking The findings corroborate several of UTAUT's hypotheses, including performance expectation, social influence, and effort expectancy, as well as the importance of risk as a greater predictor of intention.
[10] 53 Mobile wallets We then provide a novel approach to secure mobile wallets and protect the privacy of mobile users by incorporating digital signature and pseudoidentity techniques.
[99] 791 The usage intention of mobile banking The factors that have the most influence on people's willingness to use mobile banking services are social risk, social norms, and utility. When it comes to their sense of usefulness, female respondents were more impacted by ease of use than male, while male respondents were more influenced by relative advantage.
[ 21] 138 Satisfaction with mobile wallets There is a strong correlation between mobile wallet users' perceptions, preferences, and satisfaction. Additionally, the data demonstrate the effect of consumers' perception, happiness, and preference on mobile wallets adoption in India.
[106] 561 The usage intention of mobile banking Initial trust is mostly determined by structural assurance and information quality; however, perceived utility is greatly influenced by information quality and system quality. Initial trust has an effect on perceived usefulness, and both variables are associated with the desire to utilise mobile banking.
[96] 1271 Emerging IT artefacts Risk perception, which is comprised of eight distinct dimensions, is a significant predictor of new technology uptake. Apart from previous research, the findings give empirical support for the use of personal characteristic variables in assessing the adoption of developing IT artefacts.
[64] 748 The validation of a complete consumer acceptance model of mobile payment The model validates the conventional function of technology adoption factors. The users' attitudes and intentions are impacted by perceived security and trust. Demographics have a significant moderating influence on the correlations between the variables, as demonstrated by the extended model.

Discussion and Future Research
This section provides critical discussion on the topic and provides suggestions for future research. A large number of studies on digital payment have discussed it from the technical and user acceptance and adoption standpoints, with some attention to perceived security and privacy risks. In line with the common theories (i.e., the TAM and the UTAUT), it has been reported that perceived security has a positive effect on behavioural intention [102], while the effect is negative for perceived privacy [13,102]. Despite this contradiction in the direction of the relationship between subattributes of risk, some studies used risk as a general construct in their research model [102,[107][108][109]. Hence, future work might extend the existing models through multigroup analyses based on subdimensions of risk, including service-related risk, platform-related risk, network-related risk, and devicerelated risk. Additionally, cultural differences could be another driver of contradictory findings of prior studies. Kim et al. [78] reported that an individual's disposition to trust is shaped by their cultural upbringing. Dahlberg et al. [47] also suggested that social and cultural variables might have a significant effect on digital payments adoption. In this regard, customers who espouse cultural values of uncertainty avoidance would have a strong intention to use digital payments services, reducing the uncertainty of the usage of digital payment services, which would also reduce the fear of information security breaches. Moreover, consumers from low-uncertainty-avoidance cultures are likely to take more risks compared to individuals from high-uncertainty-avoidance cultures. According to several studies, culture plays an important role in the adoption of new technologies [25].
In addition, individuals in smart cities are reported to have limited adoption of digital payment for security and confidentiality concerns [11]; future research, therefore, could evaluate the security assurance and confidentiality of financial technology transactions as they are proportional to the user's convenience and the availability of the service itself. Furthermore, the complex inter-relationship between risk attributes that consequently influence individual behaviour is still unclear. For instance, Morosan and DeFranco [52] reported that perceived security influences the general privacy perception, while El Haddad et al. [13] reported this association through perceived financial fears. In addition, other studies [102,107,108] did not consider the inter-relationship between risk attributes. We therefore argue that understanding the inter-relationship between risk attributes can be another area of interest in further research.
Moreover, some prior studies evaluated the antecedents of digital payment risk, such as quality [109], ease of use [13], technology risk [110], software awareness [110], financial incentives [108], and Fintech service knowledge [111]. There are several factors that influence risk perception yet to be examined, such as reputation, as users may depend on belief indicators such as reputation to establish their initial trust in digital payment providers, and the consideration of financial transaction service providers to explore models of service improvements that comply with government rules or international standards to enhance compliance and information security culture. Furthermore, whilst the majority of studies concentrated on retail, concern and risk will differ depending on the industry. This is a critical topic for future research.

Conclusions
Globally, the proliferation of mobile phones has paved the way for digital mobility systems that benefit a variety of organisations. To strengthen the field of digital payments security and privacy, it was necessary to conduct a literature assessment and become aware of knowledge gaps, thereby defining an agenda for researchers and practitioners. Thus, this study updates a previous review of the literature published in Scopus-indexed journals to assess the field's intellectual development. The investigation adds to the literature by making the following significant findings: (1) China contributed the most to the prevalent literature on digital payment risk, while African countries contributed the least. (2) Because digital payment research has primarily focused on the unified theory of acceptance and use of technology and the technology acceptance model, these findings highlight a research gap in that there are few theories addressing the cultural aspects of digital payment users. Thus, we conclude that the theoretical and ontological frameworks available in the current literature do not adequately address the nuances and complexities of digital payment.
(3) Numerous studies have examined digital payment from both a technical and user acceptance standpoint. Researchers have a plethora of topics to choose from when it comes to the antecedents and determinants of digital payment risk.
For practitioners, our study summarises and organises existing research findings according to a set of criteria. We also offer a comprehensive framework for digital payment risk, highlighting the factors that must be taken into consideration when developing digital payment services. According to our review, practitioners should direct technical development toward increased collaboration with consumers and merchants. Additionally, our findings indicate that in order for mobile payment services to flourish, their business models must develop away from exclusive proprietary solutions and toward cooperative and standardised solutions. Newly added technological drivers draw the attention of merchants toward the influence of technology and its skilful usage on adoption behaviour. Newly added strategic activities of addressing security and privacy concerns makes a case for the importance of innovative-technology-related risk in customers' adoption.