The increasing demand for electricity due to the electrification of heating/cooling systems in buildings and the transport system, along with the alarming rise in carbon emissions from conventional energy sources, such as coal, oil, and natural gas, compelled the world to seek alternative solutions. Despite the increasing integration of renewable energy, its availability is still unpredictable and restricted. Therefore, the relentless pursuit of innovative approaches to ensure sustainable and resilient energy systems must be continued ^[1]. The European Council established the European Union’s (EU) climate action strategies to achieve a climate-neutral economy with net-zero greenhouse gas emissions by 2050 ^[2]. By 2030, the EU set key targets, which include reducing greenhouse gas emissions by at least 40% compared to 1990 levels, achieving a minimum of 32% renewable energy share, and improving energy efficiency by at least 32.5% [3,4]^[3][4]. It is noteworthy that the prior strategies outlined by the European Council aimed to achieve a 27% increase in energy savings and a 27% share of renewable energy ^[5], which highlights the necessity of further efforts and advancements in implementing sustainable energy practices and technologies to meet these ambitious targets.

Achieving the aforementioned objectives necessitates diverse actors taking a wide range of measures, from enhancing energy efficiency in buildings and the further development of renewable energy technologies to nudging behavior change in energy use. Numerous research studies created significant insights into the technological challenges of balancing energy supply and demand ^[6]. These challenges encompass a wide range of factors, such as the capacity of energy systems to facilitate affordable access to energy services and ensure the security and reliability of energy supply ^[7].

While technological advancements play a vital role in providing sustainable alternatives, the success of our efforts ultimately hinges on our ability to transform human behavior ^[8]. Increasing awareness of energy-related practices in different sectors is a cost-effective approach to promoting sustainable energy transition in societies, yet changing human behavior remains a significant challenge ^[9].

The initial effort to enhance public consciousness regarding energy usage was established upon the premise that energy is “doubly invisible”, given that the quantity and influence of energy are abstract concepts ^[10]. The intangibility of energy renders it arduous for society to conceptualize it, whereas conventional energy bills are unable to completely articulate the intricate and dynamic nature of energy consumption patterns ^[11]. In recent years, there was a growing research interest in increasing the perceptibility of energy usage through the implementation of diverse strategies, such as providing visibility for energy use ^[12], and the real-time visual display of consumption levels ^[13]. For instance, in-home energy displays are effective in average savings ranging from 4% to 12%, with over 20% peak savings ^[14]. Nevertheless, in aiming for lasting behavior change, the effectiveness of this intervention primarily depends on user engagement. Accordingly, the visualization mechanisms should be thoughtfully designed to render the feedback readily understandable to users, and provide them with insights into their energy consumption patterns and routines ^[14]. Innovative promotion of energy applications involves an interactive, collaborative, and visual approach to maintain user engagement and allow service providers to adapt to users’ capabilities ^[15], which necessitates the design, development, implementation, and active use of innovative systems, technologies, and behavior change strategies ^[16].

Intervening in the social routines of people through traditional education programs is challenging, as they may not have the same reach and impact as modern media platforms. In recent decades, serious games, particularly cross-media-oriented and multi-player-involved role-playing games, emerged as a key approach to addressing this challenge [17,18]^[17][18]. There is increasing evidence that certain individuals are more receptive to receiving information and feedback presented in a fun and engaging manner. Therefore, implementing game-based and gamification strategies could be a viable solution to address this need ^[19].

Gamification refers to incorporating game elements, mechanics, and design principles into non-game contexts. It involves applying game-like features, such as points, badges, leaderboards, and challenges, to engage and motivate individuals, encourage desired behaviors, and enhance their overall experience ^[20].

With the swift progress of digital and mobile technologies, digital game-based learning emerged as a prominent approach that can overcome the constraints of time and location and enhance accessibility and engagement for learners ^[21]. Hamari et al. ^[22] introduced three key components of serious games. Firstly, initial motivational goals that encourage individuals to engage with the game. Secondly, psychological outcomes similar to traditional gaming environments, such as increased enjoyment, motivation, and engagement, and finally, desired behavior patterns established within and beyond the game context. Among other gamification models is the Octalysis framework developed by Argilès and Chou ^[23] with eight core drives: Epic Meaning and Calling, Development and Accomplishment, Empowerment of Creativity and Feedback, Ownership and Possession, Social Influence and Relatedness, Scarcity and Impatience, Unpredictability and Curiosity, and Loss and Avoidance.

Since teachers recognized the potential of serious games, the education was an early adopter of this approach. Serious games demonstrated high motivation levels and offer instant feedback, making them adaptable to learners’ skill levels. They also facilitate effective knowledge transfer and provide opportunities for repetition ^[24]. Moreover, it is worth noting that gamification was successfully implemented in various domains beyond education, including health (for general health, rehabilitation, mental disorders, and educating patients) ^[25], business (for motivating employees and retaining loyal customers) [26^[26][27],27], and government (for promoting public participation) [28,29,30,31,32]^{[28][29][30][31][32]}.

Gamification is also used to promote sustainable behaviors, adopt resource-efficient practices in the users’ lives, and mitigate climate change ^[33]. The role of serious games in developing 21st century skills that can lead to addressing the climate crisis is a potential asset ^[34]. In this context, some of the main developed games can be categorized as sustainability education (e.g., ‘Factory Heroes’, for improving the sustainability leadership skills in manufacturing ^[34]), transportation and air quality (e.g., ‘Mordor Sharper’, for incentivizing the carpooling system ^[35]), waste management (e.g., ‘WasteApp’ for increasing recycling ^[36]), energy-related applications (which are discussed in this research), and water conservation (e.g., for increasing community engagement for water-related event preparedness such as planning to conserve water during drought ^[36]). The Smart H₂O project, with the objective of establishing a positive feedback loop between water utilities and users, provides information about water consumption in almost real time and enables water utilities to develop strategies for water supply ^[37]. This project uses a gamification approach to encourage users to modify their water consumption habits, utilizing various incentives, such as virtual, physical, and social rewards to promote competition between users ^[38].

Web platforms also emerged to serve as a directory of serious games on sustainability, such as Games4Sustainability, where games centered around the United Nations’ Sustainable Development Goals are presented ^[39]. It also provides a classification of both digital and non-digital games based on their intended age range and learning objectives ^[40].

Serious energy games (SEGs) as interactive experiences that engage users to various energy applications are gaining prominence as an innovative approach, particularly in the realm of energy usage, distributed generation, and interaction with energy markets. An early effort in SEG is the PowerHouse ^[41], which was developed to motivate the reduction in energy consumption in households in the short term through simulation-based games. After the success of initial SEG in reducing energy consumption in households, researchers expanded their focus to explore the potential of this technique for other target groups and to incorporate real data into the games (e.g., Power Agent ^[42] and Power Explorer ^[43]), which allowed for more accurate simulations and personalized experiences for users. Most of these games apply the points–badges–leaderboards model, known as ‘four-square,’ which is prevalent in serious games ^[22]. This model incorporates game-like elements and rewards to encourage interaction, engagement, and the establishment of new behavior patterns. These elements motivate continued interest and interaction, while social comparison on leaderboards and social media platforms reinforces the desired behaviors ^[11].

2. Objectives and Applications of Serious Energy Games

The energy transition is crucial for achieving a sustainable future. Still, one aspect that is often overlooked is the complexity of educating individuals on the benefits, urgency, and intricacies of renewable energy development and energy efficiency ^[24]. A properly designed serious game serves as an experimental platform that can effectively address multiple factors and explore different variables. This approach provides the necessary abstraction and flexibility for experimentation, scalability, and innovation within the energy behavior change context. To take advantage of the findings and developments from a series of empirical research projects in the realm of SEG, it is beneficial to analyze their implementation of energy applications. These projects provided valuable insights and learning opportunities, including lessons from their outcomes and failures. An overview of the reviewed research projects is presented in Table 1. These projects were conducted between 2011 and 2021 and were sourced from scholarly publications. The table provides a comprehensive summary of the research studies, including details such as project name, type of game, study duration, medium of feedback, target group, study area, and study region. According to Wu et al. ^[18], SEG can be classified into three categories based on the degree of end-user engagement: education-oriented, simulation-oriented, and application-oriented serious games. However, it is important to note that these categories are not mutually exclusive, with significant overlap between them. The latter categories often build upon and incorporate elements of the former ones. For instance, simulation-oriented games also have educational potential, and application-oriented games can integrate both educational and simulation aspects. This study also adopts the same classification to categorize SEG in Table 1. The categories are defined as follows:

• Education-oriented serious games: games that focus on energy consumption aim to raise awareness and shape behavior by utilizing game technology and design principles. These games provide virtual experiences and data related to energy conservation, using simplified real-life complexities and offering immersive learning environments to develop critical thinking and motivation. However, transferring of knowledge from these games to real life may pose challenges.
• Simulation-oriented serious games: games that aim to guide users in reducing energy consumption and exploring renewable energy options. These games utilize real-life energy data and encourage energy-related discussions. Compared with education-oriented serious games, these games connect gameplay to real-life behaviors by suggesting home-specific efficiency improvements, reducing the gap between the virtual world and reality. However, in these games, the collected data are condensed rather than detailed, and calculations are not automatically calibrated.
• Application-oriented serious games: games that utilize real or real-time data to provide engaging and practical experiences for users in various domains. These games go beyond entertainment, serving as effective training, learning, and problem-solving tools. By incorporating real-world data, users can immerse themselves in simulated environments that closely resemble their field’s challenges, enhancing their knowledge and abilities. These games offer a dynamic and authentic learning experience, bridging the gap between theory and practice.

The duration of the studies varies, ranging from short sessions lasting as little as 30 min to long-term investigations lasting up to two years. These studies targeted diverse groups to examine the impact of serious games on energy-related behaviors and practices. The selection of these target groups is determined by multiple factors, such as access to the target group, the necessity of intervention for a specific group, access to smart technologies, and game mechanics. Notably, domestic energy users were the primary focus of most of these studies, highlighting the importance of engaging this particular group. However, less is known about the challenges of applying SEG for different populations and sectors. For instance, Mendez et al. [45]^[44] focused on the university campus since they found it to be an ideal starting point to promote interaction between energy users and the city to improve energy awareness. In some research studies, the target group is selected based on social criteria. Social housing, a vulnerable segment of the housing sector facing high financial pressures, is considered one of the groups most severely affected by fluctuations and increases in fuel prices. It is reported that the social housing population in Europe faces 2.5 times more difficulty adequately heating their homes than the general population ^[11]. Research was conducted to address financial concerns and improve housing conditions within the social housing sector by utilizing SEG [11,46,47]^[11][45][46]. While previous research explored the use of serious games for energy consumers, Polyanska et al. [48]^[47] specifically focused on the application of serious games for energy companies. Drawing on the framework proposed by Figol et al. [49]^[48], two distinct types of gamification were identified: external gamification, which focuses on increasing consumer loyalty and company revenue, and internal gamification, which aims to enhance the productivity of personnel. Considering this distinction, Polyanska et al. implemented this tool in the management of energy companies in Ukraine and reported that using gamification tools could facilitate the support of energy policy and promote the effective integration of Ukrainian energy companies into the EU energy market [48]^[47]. In some studies, participants were specifically chosen from a group of high-energy consumers based on the assumption that they would have a greater potential for energy savings [50]^[49]. Including high-energy consumers allowed for a more targeted approach to understanding the factors influencing energy consumption patterns and identifying effective strategies for reducing energy usage.

2.1. User Engagement and Information

Several research projects show that public perception of climate change does not fully align with the urgency of the issue, and the public gives a low priority to climate change policy making compared to other societal problems [24,67]^[24][66]. Some investigations suggested that endeavors to communicate and educate about sustainability have not met expectations. For example, while sufficient information were provided, the presentation methods were not convincing enough to effectively convey the message [68,69]^[67][68]. Efforts to raise awareness about energy consumption aim to reduce electricity consumption, which can lead to cost savings of 5% to 15% with little to no investment required [70,71]^[69][70]. This fact incentivized some researchers to investigate incorporating innovative energy-saving features [45]^[44]. Moreover, the literature emphasized the importance of engaging customers in smart grid initiatives [72]^[71]. As an example, whether deploying smart meters alone can effectively influence the energy-related behavior of residential customers is a topic of debate [73]^[72]. In this context, many studies provided evidence to support the effectiveness of integrating interactive interfaces alongside smart meter technologies [59,66,74,75]^{[58][65][73][74]}. This integration allows for the implementation of participatory and context-specific interventions that enhance users’ awareness of their energy consumption and empower them to take proactive steps towards reducing their energy usage. Specifically, in a study conducted by Wemyss et al. [66]^[65], the Social Power application was designed as a complementary tool for the smart meter rollouts in Switzerland and leveraging their added advantages. User engagement relies heavily on the provision of information to consumers. This information serves as a crucial element in shaping their energy consumption behaviors. Darby [76]^[75] presented one of the most widely accepted classifications for feedback information as direct and indirect. On the one hand, direct feedback involves presenting raw information from energy meters or display monitors, offering immediate and easily accessible consumption feedback. On the other hand, indirect feedback involves processing the data before presenting it to the user, often through energy bills, which can result in delays in providing feedback. Direct feedback allows for user control and comprehensive representation of energy feedback, while indirect feedback offers post-processed information. The distinction between direct and indirect feedback relates to the accessibility and latency of the feedback and the level of data processing involved before reaching the user [77]^[76]. Darby [76]^[75] suggests that a well-designed combination of direct and indirect (e.g., accurate billing) feedback system is essential for achieving long-term sustainable energy consumption behaviors. In comparison, energy bills provide limited information on energy consumption and lack actionable insights for users [78]^[77]. The lack of information on everyday energy consumption, primarily limited to monthly or yearly energy bills that offer only a general overview, makes it challenging for households to understand how and when energy is used in their daily activities [79]^[78]. Consequently, misconceptions can arise, such as underestimating the energy savings obtained through energy-efficient behaviors such as enhancing home insulation and using more efficient equipment while simultaneously overestimating the energy savings derived from curtailment behaviors such as turning off lights [80]^[79]. To enhance the engagement of feedback beyond gamification features, SEG can offer consumers various types of information, including:

• Simple information: this includes basic details about energy usage, such as current energy consumption levels or historical data.

• Conjunctive information: this type of information compares the consumer’s energy usage with that of similar households or benchmarks, allowing for better understanding and context.

• Tips/Advice: information in the form of tips and advice can help consumers identify specific actions to reduce their energy consumption and make more sustainable choices.
• Forecast information: forecasting provides consumers with insights into future energy demand and prices, enabling them to plan their energy usage more efficiently.
• Demand response (DR) and statistics: this type of information involves sharing grid and/or market data such as the system balancing status, peak demand periods, pricing structures, and other statistical information related to energy consumption.

From another perspective, Wu et al. ^[18] categorized the information provided in SEG for user engagement based on their level of education into four levels:

• Level 1: visualization of energy consumption to improve end-user understanding.
• Level 2: delivery of energy-related knowledge to the end-users.
• Level 3: delivery of energy-related knowledge with a feedback mechanism to prompt behavior change.
• Level 4: enhanced engagement and behavior change through multiplayer interactions or involving the end-users family and friends via social media.

In another classification by Zangheri et al. [81]^[80], information types in energy feedback applications are classified as real-time, appliance disaggregation, social comparison, historical comparison of energy consumption, energy consumption rewards, and energy efficiency advice. Games are well suited for conveying information as they provide a contextualized environment where various factors of influence can be compared.

2.2. Demand Side Management

In the domain of SEG, researchers pursued innovative solutions to tackle pressing energy demand challenges. DSM is defined by Gellings as “planning and implementation of those electric utility activities designed to influence customer uses of electricity in ways that will produce desired changes in the utility’s load shape” [82]^[81]. DSM has diverse effects on power systems, encompassing the electricity market, environment, power system operation, and reliability. In the electricity market, consumers benefit from incentive payments, while utilities experience reduced costs, decreased load losses, and increased system efficiency. DSM initiatives enhance economic dispatch, augment electricity market performance, and mitigate market risks. Regarding power system operation, DSM aids in maintaining voltage stability, easing transmission congestion by smoothing the load profile, optimizing preventive maintenance scheduling, and postponing the necessity for facility upgrades. DSM also facilitates the integration of renewable energy sources and enhances power system flexibility [83,84]^[82][83]. Applying the DSM activities can take the form of voluntary subscription programs. However, the limited awareness of residential customers about DSM programs resulted in relatively low engagement. In this context, AlSkaif et al. [85]^[84] proposed a system architecture that integrates gamification elements into energy applications to enhance the participation of residential customers in the electricity supply market and bridge this gap. The main applications for engaging energy users in DSM can be classified into three categories: energy efficiency, self-consumption, and DR. Table 2 presents the energy applications incorporated in various SEGs. These games primarily focus on energy conservation and optimizing electricity usage within homes. A popular technique employed in these games is the use of quizzes, which effectively raise awareness about energy-related topics. However, it is crucial to consider the limitations of quizzes when it comes to facilitating deeper learning. Quizzes primarily activate low-level learning capabilities, corresponding to the lower tiers of Bloom’s taxonomy [87]^[86]. These lower-level cognitive skills involve memorizing and replicating isolated pieces of information. Electricity is the most investigated energy carrier in the reviewed studies, as indicated by Table 2, though there is a growing interest in exploring the combination of different fuel types to maximize energy efficiency. This includes the integration of electricity and gas, among others. Despite this growing interest, none of the reviewed studies provided feedback specifically aimed at optimizing the energy mix. In the following subsection, the implementation of energy applications in serious games is discussed.

2.2.1. Energy Efficiency

Over the past years, the industry’s sustained and productive effort to improve energy efficiency resulted in manufacturing devices and appliances that significantly reduce energy needs. Despite these efforts, the impact of the Jevons paradox energy consumption is more likely to increase rather than decrease as a result of economically justified enhancements in energy efficiency [88]^[87], or the rebound effect reduced the expected gains from new technologies [89,90]^[88][89]. This suggests that advancements in energy efficiency technologies, while promising, are inadequate in isolation and may not be sufficient to reduce personal and collective energy consumption without concurrent changes in consumption patterns. Gamification are also widely applied in a more simple way to impact efficient energy appliance use. In these types of games, simple tricks and practices are employed to motivate energy users to engage in energy-saving behaviors such as turning off lights, reducing the use of power-intensive appliances, and closing windows [57]^[56]. The increase in energy efficiency was a prominent aspect observed across a wide range of SEG studied. In research aiming for energy efficiency activities in the game, the objective is to engage users with energy-saving tasks. As users progress in the game, they encounter new energy-saving activities and challenges in the game and their daily lives, depending on whether the game incorporates real data. The challenges considered here mostly include informative texts and tips, quizzes (such as multiple-choice questions), and photo uploads. These challenges are designed to provide information, test users’ knowledge, and encourage active participation in energy-saving behaviors within the game environment. Recent studies utilized SEG to investigate occupants’ preferences within smart building infrastructures [47,53]^[46][52]. These games aim to encourage users to lower their energy consumption by considering various factors, such as thermal comfort, indoor air quality, lighting comfort, and general satisfaction. By incorporating occupant preferences, SEGs provide a platform for understanding the complex dynamics between energy efficiency and occupant comfort while balancing the two. Additionally, SEGs often emphasize the use of more efficient appliances. Players are encouraged to select energy-efficient appliances and devices within the game environment or in real life, which promotes the adoption of energy-saving technologies. This approach emphasizes purchase behavior and often offers comparative analyses between the energy usage of users’ appliances and the nominal average demand value for such appliances. The primary objective of these games is to promote the benefits of replacing energy-intensive appliances with more efficient alternatives, thereby encouraging energy-saving practices [85]^[84]. Another approach to enhancing awareness of energy consumption among appliances is by providing disaggregated information at either the appliance level, using smart plugs, or at the room level [91]^[90]. Identifying appliances that consume a significant amount of energy allows one to consider replacing them with more energy-efficient alternatives or adopting strategies to reduce their usage [77]^[76]. In many studies, smart thermostats were utilized to gather raw data concerning the heating or cooling demand within buildings. Heating, ventilation, and air conditioning (HVAC) systems play an important role in providing thermal comfort by regulating indoor temperature. To address building insulation, some research focused on assessing the HVAC systems’ energy efficiency by comparing energy consumption against average values [44]^[91]. However, this approach fails to address the intricate relationship between HVAC systems and building insulation. Evaluating buildings’ insulation as a means of energy conservation presents a multifaceted and intricate challenge. Its complexity stems from the requirement of comprehensive assessments and interventions that extend beyond the simple replacement of appliances, making it a critical focus area for energy-saving initiatives and SEG.

2.2.2. Photovoltaic Self-Consumption

Self-consumption of photovoltaic (PV) energy entails using the electricity generated from photovoltaic systems by the power producer, contracted associates, or private household systems without injecting it into the grid. Given that PV technology is the leading contributor to distributed power generation, it presents a significant opportunity for promoting self-consumption practices [92,93]^[92][93]. In combination with local storage, system owners may optimize revenues by participating in energy markets, typically via aggregators. A significant obstacle to achieving self-consumption in households is the mismatch between PV power generation and actual demand. Since a considerable portion of power production occurs when residents are away from home for work or other daily activities, the estimated potential for self-consumption without storage or DR measures ranges from 17% to 44%, depending on factors such as household size and exposure to irradiation ^[94]. PV electricity production follows the sun’s course during the day, typically resulting in lower feed-in of PV power during morning and evening hours. However, demand peaks tend to occur during these times, creating a disparity between high demand and low PV power feed-in. Optimizing PV system capacity with the demand as a constraint leads to placing half of the system facing East and the other half facing West ^[95]. PV technology faces challenges competing with wholesale electricity prices. Still, self-consumption is gaining traction since the decreasing costs of PV generation are approaching or reaching parity with retail prices. SEG can play a significant role in encouraging users to install solar panels and addressing disparities in PV generation and demand profiles, and increasing the share of self-consumption. These games can enhance customers’ knowledge about the importance of self-consumption, provide incentives, and empower them with self-control to increase their participation in self-consumption practices [85]^[84]. In a study by Rai and Beck ^[96], it was found that serious games can effectively bridge the information gap and empower citizens to overcome informational and perceptual barriers, facilitating the widespread adoption of solar energy in residential settings. In this context, Papaioannou et al. ^[97] designed an IoT-based framework for decreasing energy waste in public buildings. The architecture also includes a solar power microgeneration forecast based on weather predictions and historical weather data. This feature aids in minimizing the daily energy load of the building by prompting players to time-shift energy-consuming actions to periods when the net energy balance (microgeneration minus consumption) is maximized, particularly in scenarios where energy storage is unavailable. Some studies worked on the use of gamification related to the panels’ installation (e.g., Ouariachi et al. ^[24] developed a game in which the users learn how solar panels can be effectively placed on farmlands, or Olszewski et al. ^[98], presented a social gamification platform for stimulating the photovoltaic panels’ installation) and technical issues (e.g., Salim et al. ^[99] developed a serious game for improving the understanding and stakeholders’ decision-making ability for end of life management of PV panels).

2.2.3. Demand Response

The stability of energy grids depends on the continuous balancing of energy supply and demand. The integration of intermittent renewable energies, such as wind and solar, which are heavily influenced by factors such as weather, makes the task of balancing the energy grid increasingly challenging ^[100]. In this regard, an important challenge is educating individuals about the complexities of balancing supply and demand in different locations and conveying that larger locations present unique complexities ^[24]. It is essential to emphasize how individuals can actively contribute to the balancing process by aggregating distributed resources. Individuals can collectively form virtual power plants by pooling together smaller-scale renewable energy sources and DR capabilities. These aggregated resources can then be integrated into the grid to assist in balancing the fluctuations of intermittent renewable energies. Accordingly, policymakers and market participants recognize the significance of demand-side flexibility, particularly through DR mechanisms, to address these challenges and efficient electricity systems. In this context, intermediaries, such as suppliers and aggregators, offer DR programs to retail customers in the energy market through voluntary participation. These programs involve a contractual agreement outlining legal and technical criteria for implementing and verifying DR and incentives to encourage customer participation ^[15]. Generally, DR programs are classified into incentive-based and price-based categories ^[101]. To promote DR programs among consumers, various strategies are employed, such as time-of-use pricing, critical peak pricing, variable peak pricing, real-time pricing, and offering critical peak rebates. Additionally, power companies implement direct load control programs to regulate energy usage by cycling appliances such as air conditioners and water heaters during peak demand periods [85]^[84]. One incentive for using gamification for promoting DR programs is explained by Konstantakopoulos et al. ^[102]. Implementing DR programs is typically based on contractual agreements between utility providers and consumers. However, these contracts lack the flexibility to accommodate dynamic changes in occupant behavior and preferences, leading to discrepancies in demand expectations. To address this problem, they developed an approach that incorporates a gamification interface that enables building managers to interact with occupants and allows retailers and utility companies to utilize dynamic and temporal data to customize DR programs based on observed or predicted conditions. This approach enhances the adoption of more dynamic protocols for DR ^[103]. From the reviewed papers, only three applications concerned DR in their games: EnerGAware (Energy Cat) [51^[50][51],52], EnergyElastics [59]^[58], and Social Mpower [63]^[62]. In the Social Mpower application, the researchers incorporated load shifting as a strategy in addition to energy-saving tips and quizzes. Users were encouraged to shift their electricity loads from periods of high demand to off-peak periods. This involved tasks such as running the dishwasher, oven, washing machine, or tumble dryer during times when overall electricity demand is lower. The scarcity of DR-related games can be attributed to several factors, including fixed energy costs, the lack of information regarding off-peak periods, and the absence of incentives to highlight the potential impact of shifting energy usage. As a result, serious games in this domain are less popular than those centered around energy conservation ^[104]. Lampropoulos et al. ^[15] identified five objectives for the integration of gamification techniques in DR applications:

• Educating users about commercial offerings, including DR programs and self-consumption schemes.
• Raising awareness about energy usage through advanced metering infrastructure and consumer interfaces.
• Driving adoption of smart grid technologies and smart appliances.
• Encouraging active participation in DR programs and self-consumption schemes through incentives.
• Influencing behavioral changes measured by key performance indicators.

In a study by Gnauk et al. ^[100], gamification is used for demand dispatch. The demand dispatch system aims to encourage consumers to meet their energy demand with flexible options and maintain their engagement over the long term. They used gamification to develop an intrinsic motivational framework for consumers to explore their consumption habits enjoyably and interestingly, establishing a deep commitment to the program. The demand flexibility system proposed in this study consists of three steps: submission of new flexibility by the customer in the definition phase, utility-side review through multiple rescheduling runs, and ultimately reaching the dispatch phase. After evaluating this approach, users exhibited increased motivation and incentive to actively participate in the program.

2.3. Social Connection

The inclusion of social connections in SEG can enhance the enjoyment and attractiveness of energy applications. This social dimension not only adds an element of fun but also creates a sense of community and encourages positive energy behaviors [85]^[84]. Incorporating social elements and online interactions into gamification approaches can harness the significant potential of social connection to elevate energy-related experiences. This integration holds promise for fostering positive energy behaviors, promoting knowledge sharing, and facilitating collective problem solving, ultimately contributing to enhanced energy experiences and improved outcomes. Recent literature paid significant attention to the crucial role that social ties play in SEG. For instance, social ties were recognized as powerful tools for increasing player engagement and promoting behavior modification. In the PEAR project, players establish teams with friends and encourage cooperative energy-saving acts, demonstrating the influence of peer pressure and competition in encouraging sustainable habits ^[105]. Similar to this, the Powersaver project makes use of household-based teams, where players track and lower their household’s energy use and then compare their results with other teams ^[106]. In today’s digital learning environments, social learning is a trend that is amplified by such processes. According to Bandura’s social learning theory, peer observation and utilizing social experiences can result in significant behavioral changes ^[107]. When players, in their neighborhood-based teams, observe peers making energy-saving decisions or implementing best practices, they are more inclined to emulate such behaviors. The concept of collective awareness of energy use is incorporated in many SEGs and defined as an attribute found in communities or teams that help them solve collective action problems [61]^[60]. Without collective awareness, individuals may disregard community norms and fail to understand the impact of their actions. In communities with collective awareness, members take synchronized actions to achieve desirable outcomes for shared resources. Several SEGs are designed to educate energy consumers about resource allocation, electricity prices, and grid sustainability. Some games go a step further by incorporating social connections to enhance the learning experience. One such game is Social Mpower, which aims to prevent a collective blackout. Players achieve this by individually reducing their energy consumption and coordinating their actions in synchronization with others [61,62,63]^[60][61][62]. Different methods are developed for users’ communication in SEG with social connection elements such as in-game messaging, chat or discussion forums, and team or group communication. Alskaif et al. [85]^[84] proposed to enable users’ communication by linking the application to social media or developing a private web-based or mobile-based platform. In the Social Power project, the household participants were assigned to one of two teams: either a collaborative team where citizens in the same city try to reach a fixed, collective 10% electricity savings target together or with a competitive team that tries to save the most electricity in comparison to the other city. For social connection, the users designed a blog and Facebook page as a place for participants to interact, share experiences, and cooperate to build a creative understanding of how to save electricity at home. Players could find more detailed information about the energy-related topic of the week, post tips, offer suggestions, ask questions, and could cheer on their teammates. However, few participants were engaged in these traditional communication channels, and more interest was shown in the app challenges. In another study, Kashani and Ozturk [59]^[58] created a gamified platform for promoting energy-saving behavior using a mobile application that users can access via their Facebook accounts. The application requests permission to access the user’s friend list, allowing them to invite friends to participate in the game’s challenges. Through this feature, the application facilitates the creation of a social network, enabling users to share information and engage in friendly competition. Additionally, the application presents other users’ energy-saving activities, creating an educational and competitive environment that encourages individuals to learn about energy-saving solutions. In some studies, the social connections among users of SEGs were limited to the social sharing of their achievements and challenges, typically involving users sharing their progress, scores, or energy-saving accomplishments on social media platforms, fostering a sense of competition and community engagement. However, in other studies, the focus goes beyond mere social sharing. It extends to creating an energy community, in which users actively cooperate and share resources, particularly in the context of shared renewable energy sources, e.g., a co-owned PV installation shared between the residents of the community ^[104]. The body of research consistently underscores the capacity of both competitive and cooperative game mechanics to promote social engagement in the context of energy-related activities.

Collaboration and Competition

The surge in online gaming and the proliferation of online connections among players paved the way for the application of collaboration within the realm of SEG. This paradigm shift involves moving away from solely focusing on individual scores and achievements and embracing the concept of collective scores. It also entails transitioning from fixed user roles to user-adaptive roles, allowing players to adapt their roles based on the specific energy challenges dynamically. By leveraging the power of collaboration, SEG can foster a collective reduction in energy consumption during peak hours, leading to significant impacts on the overall energy system [85]^[84]. However, collaboration is still one element that is missed in most SEGs ^[108]. “We Energy Game” ^[24] is an excellent example of how serious games can utilize cooperative mechanics in a simulation environment. In this game, five different roles are defined as production (project leader responsible for energy production), people (the citizens), planet (responsible for clean energy production), profit (responsible for measuring the profit made by the different projects), and balance (responsible for the network operation). To obtain a positive balance, players must work collaboratively to achieve the total score for the chosen town. To do so, they must strategically use various energy sources, each offering a point value for each role. Players place these sources on a map to accumulate points and must balance the positive and negative scores to find the optimal solution with a mix of all available sources. To foster social influence within the system, Alskaif et al., proposed applying competition by enabling the users to compare their performance with other customers of similar household size and friends, neighbors, or the average household [85]^[84]. Normative feedback can be incorporated into collaborative game designs to showcase the collective performance of a group and promote energy-saving behaviors. Grevet et al. ^[109] developed a social visualization system for energy-saving behavior within a scalable society. This system allows for establishing society-wide goals and aims to encourage collective energy behavior. The social visualization provides unidimensional and multi-dimensional comparative feedback, enabling participants to compare their energy-saving efforts within their group and across different groups. In another study by Muchnik et al. ^[110], the authors explored the potential for competition and collaboration within saving energy applications. As a means of fostering competition, users were able to compare their energy usage with that of their peers, including friends on social networks or the average user. To ensure that such comparisons were meaningful, the authors recommended providing filters allowing users to compare their energy usage with others in similar types of buildings or apartments. Regarding collaboration, the authors defined it as an energy-saving tip exchange whereby users could share tips and utilize tips from other users. In many studies, the normative feedback approach is employed to nudge users to change their behavior ^[111]. One effective aspect of this approach is the ability to compare residents’ energy usage with their peers in similar buildings, often presented through graphical representations. This comparison created social pressure among peers (peer pressure), motivating residents to embrace energy-saving practices [112,113,114]^{[112][113][114]}. The effectiveness of normative feedback is enhanced when it conveys that a significant majority of other users of the application already adopted the desired behavior ^[115]. In SEG, the normative feedback approach is commonly implemented through leaderboards, allowing users to compare their total energy consumption or energy-saving performance with others or the community as a whole. However, it is crucial to acknowledge that implementing the normative feedback approach without considering the necessity for personalization to establish a more precise basis for comparison can lead to ineffectiveness, as demonstrated in several research studies. For instance, in their study, Kim et al. ^[114] underscored the significance of considering the households’ demographic characteristics and any specific underlying health conditions when considering energy consumption to ensure fairness within leaderboards.

2.4. Personalization

The idea of personalization (tailoring the gaming experience to meet the individual preferences and characteristics of each player’s individual preferences and characteristics) in SEG to enhance their effectiveness and impact is receiving increasing attention in recent studies. A few of the available tools for energy conservation guidance often exhibit a generic approach, where the advice is presented in a “one size fits all” manner applicable to everyone ^[116]. In this context, different approaches are implemented for personalizing the game environment in both the design and implementation stages (Table 2). Achieving this objective necessitates a comprehensive understanding of the user’s needs, expectations, typology, habits, and comfort level to provide timely, relevant, and personalized feedback [54,117]^[53][117]. Implementing behavior change theories in a serious energy application is an approach used in some studies for personalization. Muchnik et al. identified four key behavior aspects to address when developing an energy application ^[110]. Firstly, the “foot-in-the-door” technique should be used to show users that they already care about energy conservation. Secondly, users should be involved in competitions; however, it is important to consider the unexpected influence of the “boomerang effect” on low-energy consumers ^[118]. Thirdly, the application should provide users with feedback on their successes and failures. Lastly, information should be presented in an easy-to-understand manner. Several theories and models regarding human behavior were developed to affect meaningful changes in users’ habits. Mendez et al. ^[119] identified three of them to reduce household energy consumption. The first, TTM ^[120] classifies the process of behavior change into six stages, including pre-contemplation, contemplation, preparation, action, maintenance, and termination. The model was utilized across multiple domains to encourage behavior change, including promoting water ^[121] and energy conservation ^[122] among residential customers. Alskaif et al. [85]^[84] defined the requirements for energy-related behavior change and implementation of energy-related strategies in various stages based on the 5-stage TTM model, as depicted in Figure 1. These strategies include the knowledge process, learning how to use and interact with a platform for promoting energy applications, adopting new energy consumption behaviors, observing the outcome of the user’s actions, and providing incentives to sustain the new energy consumption behavior. The second model, the Fogg model ^[123], emphasizes the convergence of motivation, ability, and prompt elements as necessary conditions for behavioral change. Based on this approach, Wendel suggested an iterative cycle for behavior change based on four steps: gaining insights into strategies for modifying behavior, establishing target users, desired actions, and intended results, creating a user interface that aligns with user scenarios, and evaluating outcomes and enhancing the product based on feedback ^[115]. Based on Wendel’s proposal, Kim et al. ^[114] developed an eco-feedback design to promote energy-conserving thermostat adjustment behaviors. The four steps outlined above were implemented as follows: reviewing eco-feedback strategies that were utilized by previous research, reviewing existing literature on residents’ behaviors regarding programmable thermostats to identify the target behaviors, creating an interface that aligns with users’ primary tasks and behavior scenarios, and developing a test and refinement approach to enhance the overall user experience. The third model, the theory of planned behavior ^[124], posits that an individual’s behavioral intentions are shaped by three key factors: attitude, subjective norms, and perceived behavioral control. Based on the theory of planned behavior, Mendez et al. ^[119] proposed an energy-saving behavior model for smart homes combined with gamification elements. In a few studies, users engaged in the application development process through co-creation and co-design sessions. This helps researchers to gain insights into users’ needs, preferences, and goals. For example, the EnerGAware project invited social tenants to develop its application. Initially, the project team collected the users’ ideas and inputs related to the initial game concepts. The same group was then utilized to test early game prototype ideas. Finally, the group was employed as the pilot for the deployment and testing of the application [54]^[53]. While the co-design and co-creation approach showed promise in various contexts, its application in developing SEG is still a work in progress, particularly due to the intricate nature of the development process for digital games. In addition to various approaches for delivering personalized feedback to users, another effective method involves providing personalized information on demand. One such method is the ability to forecast future energy consumption, enabling users to anticipate their energy needs and make informed decisions accordingly. Another aspect involves clustering users based on their energy consumption profile, allowing them to be categorized into groups with similar energy usage patterns. This clustering helps identify user segments and tailor feedback strategies accordingly ^[125]. Moreover, this method provides disaggregated data on individual users by breaking down energy usage by specific appliances, time periods, or activities. One study that stands out for its robust methodology in the area of energy consumption personalization was conducted by Fraternali et al. [53]^[52]. The authors utilized different data collection sensors, such as smart plugs and smart meters, to determine the activities of energy consumers, including sleeping, cooking, and resting. They also measured the comfort level and energy behavior of buildings and stored the extracted values in a database with a reference timestamp for use by the recommendation engine and other components of the application [53,54]^[52][53]. An adaptive in-context action recommendation feature was developed in their framework that computes actionable energy-saving suggestions. The recommender clusters users into categories and utilizes their activity patterns to generate energy-saving recommendations specific to their context. Furthermore, it prioritizes energy-saving recommendations based on the user’s context, activity, and suggested action’s potential impact. To overcome the cold start problem, the recommender is equipped with general rules that apply to all classes of users and buildings and are used to initialize the feedback loop between the recommender and the user. Changing energy behavior can potentially negatively impact the quality of life, as individuals may have different priorities to consider [45]^[44]. In this regard, Fraternali et al. [53]^[52] classified the comfort level of the users into two categories: visual and thermal comfort. Visual comfort is determined by the human perception of luminance within a building and is assessed using the Kruithof graph, as described by Fotios ^[126]. Meanwhile, thermal comfort is determined by the human perception of indoor temperature and is evaluated using the predicted mean vote index, as outlined by Cigler et al. ^[127]. As the users progressed in the application, the researchers observed changes in their comfort level. According to Csikszentmihalyi’s concept of “flow’’, optimal experience and engagement occur when the challenges of an activity match the users’ skills ^[128]. Translating this into the SEG context, the concept of flow could ensure that the challenge levels correspond to the user’s skills, promoting their active engagement and sustained interest in the game (see Figure 2). For instance, if the game’s energy-saving challenges are too easy for experienced users, they may lose interest due to boredom. Conversely, if the tasks are too difficult for novice users, they may become frustrated and disengage. Implementing a dynamic system that tailors the level of challenges to the players’ skill levels may optimize user engagement. As such, the game can effectively guide the user towards sustainable energy behaviors while keeping them motivated ^[129]. The “flow” can be maintained by dividing it into four main categories: engagement, learning outcomes, usability, and user experience.

• Engagement: If the difficulty of a serious game is balanced with the users’ abilities, they are more likely to remain engaged. When the game is too easy, it can result in boredom, and the users may lose interest. On the other hand, if the game is too hard, it can lead to anxiety and frustration, which also decreases engagement.
• Learning outcomes: The level of challenges directly influences the learning outcomes. An optimal difficulty level in a game encourages deep learning and fosters intrinsic motivation. It creates a more rewarding experience for the users, increasing the chances of returning to the game and absorbing more knowledge about energy management and savings.
• Usability: Balancing game difficulty can also enhance the usability of a serious game. If the users perceive a game to be within their skill level, they are more likely to understand and utilize the game mechanics. This perception of competency enhances the user experience and makes the game more accessible and enjoyable ^[130].
• User experience: Overall, ensuring the right level of difficulty contributes to a positive user experience. A game that is appropriately challenging enhances satisfaction, promotes longer play times, and can increase the desire to play again. All of this contributes to a more enjoyable and effective serious game. By ensuring that “flow” (optimal play state) is achieved, a significant contribution can be made towards mitigating user participant attrition and fostering sustained long-term engagement in the game.

Categorization of users is a widely used strategy for providing personalized feedback in SEG. In Table 3, different categorization techniques that were implemented in previous studies in this field are described. The Bartle player type taxonomy [86]^[85] proposed the classification of players into four categories, each characterized by distinct propensities and motivations. Hedin et al. [64]^[63] designed the Piggy Bank game for energy saving based on this categorization to make the game appealing to all player types. While the developed game in this study considers various player types, it is concluded that individuals classified as “Achievers” may exhibit superior performance and greater enjoyment when engaging with this application. Frankel et al. ^[131] presented five categories of market segments and the main characteristics of energy customers. They presented this framework to assess how providers can deliver energy efficiency opportunities to the market. Ponce et al. ^[132] applied this classification to design an intelligent expectation interface for adopting smart thermostats, and Mendez et al. ^[119] applied the same approach using a gamification structure. In another classification by Peham et al. ^[133], energy target groups are divided into three categories: early adopter, cost-oriented, and energy-conscious. Some studies utilized personality traits as a categorization technique, which has proven effective in delivering personalized feedback in SEG. Among the various categorization approaches, the big five personality traits emerged as the most frequently employed method. Personalization plays a crucial role in elevating the effectiveness and impact of SEG. Incorporating behavior change theories, co-design and co-creation sessions with potential users, and personalized feedback strategies contribute to a comprehensive and user-centric approach. Categorization techniques, such as player types, market segments, and personality traits, further enhance personalization, allowing for tailored experiences that resonate with specific user groups. The emphasis on user comfort levels and the consideration of diverse priorities ensure that energy-related activities align with users’ needs and does not compromise their quality of life.