A game-based augmented reality navigation system to support makerspace user education in a university library Free

As early as 1873, public libraries hosted social gatherings on the theme of sewing or crafting, and in the 20th century, libraries formally set up dedicated spaces for patrons to rent tools or do creative crafting (Good, 2013). In recent years, an increasing number of academic libraries have been provided techno-savvy spaces that are generally regarded as makerspaces to play a significant role in serving the learning needs of university students (Haruna and Kiran, 2023). Makerspaces are places where participants may work together to create and co-create knowledge and physical or digital products (Mersand, 2021). In other words, a makerspace is a learning space for creative activities, providing resources, such as crafts, new technologies, tools or other resources supporting users’ creativity needs (Britton, 2012) to enrich design thinking (Sheridan et al., 2014) and to expose students to implement and practice the application of knowledge related to science, technology, engineering and mathematics (Martinez and Stager, 2013). With the evolution of library functions, Noh (2015) proposed the concept of library 4.0, defining the concept of “infinite creative space”, the critical point of which is to enable students to not only extend the knowledge learned in a class but also develop the ability to create new knowledge (McComas, 2014). Therefore, providing hands-on creative services in the makerspace of a library has been considered an innovative patron service (Tan, 2019). However, makerspaces are still emerging services offered within academic libraries that need to pay much attention to development (Bell et al., 2023).

To enable users to make full use of a makerspace in a library for cultivating creativity, it is necessary to develop education on the use of a library’s makerspace (Maceli, 2019). However, due to the diverse equipment and applications in a library’s makerspace and the different learning needs of individual users, librarians often feel overwhelmed when conducting user education in a library’s makerspace. Osawaru et al.’s (2020) study indicated that some of the challenges encountered in academic libraries to adopt makerspace services are training of academic library staff, security of makerspace gadgets, poor funding, erratic power supply, high cost and the maintenance of equipment. Traditional library user education is usually conducted through librarian-guided tours and instruction, but with the development of information and communication technology, library user education is gradually moving towards a technology-based learning mode, such as the use of Web-based learning (Chen et al., 2022; Dewald, 1999), virtual reality (VR) (Lim, 2021; Xiao, 2000) and augmented reality (AR) (Chen and Tsai, 2012). For example, Chen et al. (2022) developed a Web-based inquiry learning mode with collaborative digital reading annotation system support to provide an innovative learning mode for effectively enhancing the information literacy of primary school students. Also, according to Xiao (2000), panoramic VR has the potential to improve Web-based library instruction by offering a combination of a physical tour and a Web-based virtual tour. This makes Web-based library instruction an effective tool, enabling students to access learning content remotely, as they can navigate, view, read and listen to the content. Additionally, game-based learning has also been widely used in library user education (Branston, 2006), and a few previous studies have confirmed it is effective in promoting learning motivation, sustained attention and interest (Domínguez et al., 2013; Rastegarpour and Marashi, 2012). Compared to traditional teacher-led instruction methods, game-based learning has a greater chance of successfully facilitating learners to engage in meaningful autonomous learning (Chen et al., 2023; Gee, 2003; Prensky, 2001; Shaffer, 2006; Van Eck, 2006). To guide games with good design, Chou (2016) proposed Octalysis gamification framework with eight core drives, including “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” as the design consideration for the learning functions and gamification mechanisms (Toasa et al., 2020).

Tan (2019) proposed four key elements in which a makerspace is regarded as a learning environment with high playability, high authenticity, development of nuanced knowledge and linking practice to representation. Accordingly, alternate reality game (ARG) has been widely used in museums, scenic spots and urban exploration in recent years, such as Urban Treasure Hunt (Huizenga et al., 2009). This type of game transposes the escape game into an actual situation. The interactive plot is often integrated into the real and digital worlds, and the elements of solving puzzles and crimes are incorporated into the story context, making the players feel more immersive and making the game more exciting and topical (Donald, 2008). University libraries have used escape room games as an application for orientation and outreach programmes that can help students become familiar with library services and resources (Gregor, 2018; Wise et al., 2018). It is clear that the characteristics of ARG can be applied to makerspace user education and that AR, a technology that combines virtual and physical environments, can be effective in enhancing students’ interest and engagement in learning (Stylianidou et al., 2020). Donald (2008) suggested that AR games could aid library user education to facilitate students’ familiarity with library services; however, no research has been conducted on using ARG combined with AR for user instruction in makerspaces. Based on all of the above, developing technology-based learning modes with game and AR features in a library’s makerspace that allows users to learn by themselves excitingly and effectively makes it possible to reduce librarian workload and help users learn autonomously how to use a library’s makerspace.

Moreover, scaffolding is regarded as a procedure that offers cognitive and social assistance to enhance students’ problem-solving skills within an educational environment (Vygotsky and Cole, 1978). In general, epistemic scaffoldings are frequently used to guide technology-supported and explanation-driven inquiry (Sandoval and Reiser, 2004). Therefore, this study develops a game-based augmented reality navigation system (GARNS) that combines an alternative reality puzzle game and AR tools with scaffolding design based on the Octalysis gamification framework and epistemic scaffoldings to support makerspace user education in a university library, allowing learners to understand the use of makerspaces and related equipment through the process of solving puzzles with the support of AR tools and to compare the difference in learning effectiveness, learning motivation and perception of learners among the use of GARNS, Web navigation system and narrative guided tour with a librarian for makerspace user education.

Previous research has pointed out that AR can help students to explore the real world (Dede, 2009) and can increase students’ learning motivation, helping them to gain better observation skills (Sotiriou and Bogner, 2008). Dunleavy et al. (2009) indicated that AR has advantages in creating immersive learning environments that incorporate virtual and real-world environments to promote learners’ critical thinking and problem-solving skills. AR-assisted learning can also help students’ understanding, strengthen positive learning attitudes and satisfaction and facilitate learning outcomes (Akçayır and Akçayır, 2017). Dalili Saleh et al. (2022) indicated that university libraries are becoming fourth-generation libraries, and this requires the provision of the necessary facilities and equipment and the application of the latest technologies. One such emerging technology is AR, that can greatly assist library management and improve the librarians’ and users’ professional activities. However, the lack of a suitable native platform for the implementation of AR technology is the key challenge to the development of fourth-generation libraries (Dalili Saleh et al., 2022).

To support library user education in an elementary school effectively, Chen and Tsai (2012) proposed an augmented reality library instruction system (ARLIS) to assist students in correcting book permutations after learning about the Chinese library classification scheme. Their study showed that this learning model improved learning satisfaction and motivation. Henrysson et al. (2005) suggested that the proliferation of mobile devices provides an ideal platform for the development of AR and mobile devices have many advantages, such as portability, interactivity and stand-alone operability (Hwang et al., 2012). The use of mobile devices with AR enables learners to be immersed in the learning process and improves student learning outcomes (Chiang et al., 2014) and mobile AR can also make boring learning more entertaining (Ibáñez and Delgado-Kloos, 2018). Therefore, this study developed the GARNS using mobile devices for makerspace user education.

In recent years, in addition to using game-based learning in academic subjects, libraries have begun to use games instead of traditional instruction to enhance their patrons’ information literacy and problem-solving skills (Battles et al., 2011). Games have the potential to help learners comprehend the process and execution of a particular task, and they can even provide a more immersive experience that closely replicates real-world work (Reade, 2017). Gregory and Broussard (2011) recommended using game-based learning and emphasized the effectiveness of integrating game-based learning with libraries. Marcus and Beck (2003) compared the differences in learning performance between traditional instruction and game-based learning in library user education and showed that learners using game-based learning for library user education had significantly better learning performance than those using traditional instruction.

In game-based learning, ARG can potentially facilitate learners to engage more in learning or exploration because of their story-driven approach (Szulborski, 2005). Huizenga et al. (2009) designed a city treasure hunt game in Amsterdam, The Netherlands, which allowed learners to explore the city in depth through gameplay and storytelling; the results of the study confirmed that this type of learning method can effectively promote learners’ learning effectiveness, learning motivation and engagement. Donald (2008) also suggested using ARGs for library user education, where the real world is transformed into a digital gaming scenario so that learners are more immersed in the virtual story to familiarize library services. In the past, university libraries have used alternative reality games with escape room game features as an application for orientation and outreach programmes that can familiarize students with library services and resources (Gregor, 2018; Wise et al., 2018). Therefore, this study uses ARG features and concepts to design GARNS to supplement makerspace user education. By combining games with real-life scenarios in makerspaces, learners can learn about the environment and apparatus of makerspaces through self-learning from exploring and solving puzzles to promote the effectiveness of the makerspace user education and to enhance the learners’ learning motivation and immersive experience.

The GARNS developed in this study refers to the Octalysis gamification framework to design the gamified learning mechanisms. Among the eight core drives, Chen et al.’s (2023) study suggested that designing learning functions that provoke the core drives of intrinsic motivation, such as unpredictability and curiosity, social influence and relatedness and empowerment of creativity and feedback, in a game-based learning system might be more beneficial than eliciting the core drives of extrinsic motivation, such as development and accomplishment and ownership and possession, in promoting learning effectiveness. Three core drives, including the epic meaning and calling, development and accomplishment and unpredictability and curiosity, were considered to design the GARNS developed in this study for makerspace user education and are explained in detail in Section 3.2.

The study refers to the equipment of a makerspace in a Taiwan’s university library to design the proposed GARNS and teaching content. This study used the AR educational game editing tool “Mini Universe” to design and create the GARNS that can be operated using mobile devices. “Mini Universe” was designed with gamified learning mechanisms and adopted epistemic scaffoldings as its core design ideas, providing easy game mechanism settings and content editing functions to help users quickly design AR educational games with cognitive scaffolding and real-time feedback mechanisms that are compatible with teaching content to assist learners in exploration and learning. The AR educational game development tool of “Mini Universe” mainly offers two functions: scanning and using stickers. In scanning stickers, after scanning the corresponding object designed for a machine in the makerspace, additional information for understanding the machine’s operation can be superimposed on the machine. Users can view the extended supplementary information anytime during the learning process through the function. The functions of using stickers include a sticker analysis area and a sticker storage area. The stickers obtained will be saved and displayed in the sticker storage area. Learners can click on the stickers in the storage area and put them into the analysis area to analyse them with different game mechanisms to view information, solve problems and complete game tasks.

This study is based on the makerspace user education outline shown in Table 1 to design the learning content. Using the makerspace as a learning environment, an AR puzzle-solving game oriented towards problem-solving tasks was developed and evolved into the GARNS in this study. This study’s game task design concept that refers to the Octalysis gamification framework (Chou, 2016) allows learners to explore and solve puzzles in the makerspace according to the mission calling from monsters. Learners can collect clues by scanning stickers and use them to analyse and complete game tasks, thus learning relevant knowledge about the equipment operation. There is a total of four game tasks, each corresponding to the learning of a machine. The game task design has multiple paths to avoid interference or hitchhiking among multiple players. At the beginning of the game, learners will randomly receive the challenge sequence for the four tasks and complete them according to their order. To achieve good learning effectiveness, in addition to the Octalysis gamification framework, this study also uses clues, matching and combination as cognitive scaffolding designs for learning tasks:

• clues: learners conduct contextual analysis and inference thinking based on the information they obtain, such as associating the correct items according to the hints given during the game process;

• matching: learners can focus on observation and conceptual verification, such as finding the appropriate mould for a specific item; and

• combination: learners can correctly choose resources and think systematically, such as collecting materials and machines required to make items and combining them to complete game tasks.

In the game process, each learner will receive a “mission guide card” at the beginning, as shown in Figure 1. The top section of the card is the “monster gift recipe”, which prompts learners that each monster’s desired gift is made from a combination of machine, consumable, file format, computer (software) and mould set. This design refers to the core drive of epic meaning and calling that makes people believe that they are doing things more significant than themselves or that they were selected to do something according to the Octalysis gamification framework proposed by Chou (2016). Each sticker represents an item of the machine, consumable, file format, computer (software) and mould scattered throughout the makerspace and learners need to collect the required information and items to make the desired gift that each monster wants to be based on the clues provided by a sticker. Restated, a sticker’s clues can be regarded as epistemic scaffoldings that can guide learners to finish the game mission. The bottom section is the “monster challenge sequence”, with the different challenge order for each learner.

The “mission scorecard” is shown in Figure 2. When learners have collected all the clues that can make the desired gift, they must use the blue energy stone at the top to exchange for an “energy key” sticker, which allows them to combine the gathered items simultaneously and complete the mission. The purpose of the blue energy stone is to prevent learners from simply trying different combinations without actively looking at the content of the stickers to find the answer. Additionally, the blue energy stone can be used to exchange stickers for obtaining additional hints to prevent learners suffering from difficulty in the game, thus affecting the learning experience. The remaining energy stones will be counted in the score, and if they run out of energy stones, learners cannot continue exchanging for “energy keys” and hints. This design refers to the core drive of development and accomplishment that provokes people’s intrinsic desire to make progress, develop skills, reach mastery and ultimately overcome challenges, according to the Octalysis gamification framework proposed by Chou (2016).

During the ongoing mission process, learners use the scanning function to collect stickers (as shown in Figure 3) containing information and tools related to making the gift according to the epistemic scaffoldings design of assisted makerspace equipment learning. Learners can select stickers from the sticker storage area at any time and then click “use alone” (as shown in Figure 4) to view information related to machines, consumables, file formats and gift clues (as shown in Figure 5). Once learners have collected all the recipe items according to the mission guide card, they can use the energy stone to exchange for an energy key sticker and select all the recipe items and energy key stickers to “use simultaneously” (as shown in Figure 6). If the combination is correct, a successful combination screen and information will appear (as shown in Figure 7). If the combination is incorrect, learners must recheck or collect the required recipe items and exchange them for energy key stickers to try again. This design refers to the core drive of unpredictability and curiosity that makes people want to know what will happen next, according to the Octalysis gamification framework proposed by Chou (2016). After completing the mission, learners can return to the app login screen to continue the challenge with the next monster. Learners who complete the gift challenge of four monsters within the allotted time can finish the game. Through the puzzle game in the monster mission, learners can understand the environment, characteristics and usage of equipment in the makerspace.

This study recruited 24 research participants who were grade 11 students from a high school in Keelung City, Taiwan, with ages ranging from 16 to 17 years old. All the research participants were first-time participants in the library’s makerspace education. In this study, the research participants were randomly assigned to the experimental group with GARNS, the control group 1 with a Web navigation system or the control group 2 with a narrative guided tour with a librarian for the makerspace user education, including ten in the experimental group, seven in the control group 1 and seven in the control group 2.

This study adopted a true experimental design to experiment with gathering research data. Since a librarian-guided tour and a technology-based learning mode, such as Web-based learning, are the two most widely used learning modes for library user education (Dewald, 1999), a narrative guided tour with a librarian and a Web navigation system were used to compare with the GARNS developed by this study for makerspace user education. Based on the consideration, a total of 24 research participants were randomly assigned to either the experimental group supported by GARNS, the control group 1 supported by a Web navigation system or the control group 2 supported by a narrative guided tour with a librarian for makerspace user education; the research participants in all three groups received the same learning content. The experimental and control group 1 were offered an individual learning mode for conducting a makerspace user education through digital learning methods. In contrast, control group 2 was offered a group learning mode with a narrative guided tour by a librarian for conducting a makerspace user education. The experimental group used the GARNS by a handheld mobile device to learn how to use the equipment of a makerspace to design something through solving game missions and interacting with physical equipment to conduct a makerspace user education. In contrast, control group 1 used a Web navigation system by a handheld mobile device and control group 2 used a narrative guided tour with a librarian to learn how to use the makerspace equipment to design something to conduct a makerspace user education.

The experimental procedure of this study was divided into three stages, including the pre-test, experimental activity and post-test stages, as shown in Figure 8, which lasted a total of 90 min. In the 25-min pre-test stage, the researcher first explained the learning objectives and teaching mode and sought each subject’s consent to participate by filling out a paper consent form according to their wishes. Research participants who agreed to participate in the experiment completed the pre-test of the makerspace user education test sheet, which allowed the researcher to obtain background information of the research participants and to understand their prior level of knowledge about the equipment of the makerspace before conducting the experimental learning activities. In the 45-min experimental activity stage, a 10-min introduction and demonstration of the system functions was conducted before the start of the experimental activity for the research participants in the experimental group using GARNS and control group 1 using a Web navigation system for makerspace user education. For the experimental group, the description of GARNS included the introduction of the game rules and game flow, as well as the explanation and demonstration of the AR system scanning and game task operation functions to enable the research participants to grasp the functions of the system; for control group 1, the explanation of the Web navigation system included the structure of the Webpage and the way of searching for information; and for control group 2, the group was arranged to gather according to the time of the tour, and then the librarians would lead the group tours on the makerspace. All three groups were given 30 min to conduct learning activities using the makerspace. In the 20-min post-test stage, all three groups of learners were required to conduct a 10-min post-test of the makerspace user education test sheet and the motivated strategies for learning questionnaire (MSLQ) to compare the learners’ learning effectiveness and motivation of using the three different learning modes. In addition, to gain a deeper understanding of the learners’ experiences and feelings towards the three different learning modes of makerspace user education, 10-min semi-structured interviews were conducted with five randomly selected participants from each of the three groups, for a total of 15 learners in the three groups, to collect qualitative data to supplement the quantitative analyses. In other words, this study adopted a mixed research method, including quantitative and qualitative analysis, to examine the research questions of this study. Quantitative analysis is the dominant method, whereas qualitative analysis is an auxiliary method.

This study used the Moodle platform to design a Web navigation system for control group 1. The menu of the system interface was designed in two levels according to the learning content of the makerspace used in this study. The home page will list four learning items, including the FDM 3D printer tutorial, stereolithography 3D printer tutorial, thermal transfer printer tutorial and laser cutter machine tutorial (Figure 9), and learners can check off the completed items. The learning content is presented as a Web page with pictures and text descriptions (Figure 10) and learners can select the machine to view its introduction, including machine functions, usable consumables, file settings, parameter settings, notes and so on.

This study compiled the “makerspace user education test sheet” based on the teaching content of the designed makerspace and discussed it with the makerspace librarians to determine the appropriateness and validity of the test content. The test content is presented in a problem-solving manner, providing learners with simulated scenarios and goals to answer questions. The test types include knowledge, combination and sorting questions. This study conducted pre-tests and post-tests using this test sheet before and after the experiment to evaluate learners’ prior knowledge and learning effectiveness. The test content includes four types of machines, each with five questions. The pre-test and post-test will randomly select two types of machines to form a test sheet with ten questions for learners to avoid students answering based on memory.

This study used the MSLQ developed by Pintrich et al. (1991) based on social cognitive learning theory to measure the intensity of learners’ learning motivation during the learning process. The questionnaire is divided into three parts: value, expectation and emotion, and includes seven dimensions: intrinsic goal orientation, extrinsic goal orientation, task value, control belief, self-efficacy, expectation for success and test anxiety, with a total of 31 items. The Likert scale ranges from 1 to 5, with options ranging from “strongly disagree” to “strongly agree”. A higher score indicates better learning motivation and vice versa. The overall reliability of the questionnaire is 0.96 (Feiz et al., 2013), indicating good reliability.

At the end of the experiment, semi-structured interviews were conducted with the three groups of learners to understand their feelings, thoughts and suggestions on using the GARNS, Web navigation system and narrative guided tour with a librarian to conduct makerspace user education to supplement the insufficiency of quantitative data analysis. This study randomly selected learners from the GARNS group, Web navigation system group and narrative guided tour with a librarian group to conduct semi-structured in-depth interviews. The GARNS group consisted of four learners, the Web navigation system group consisted of three learners and the narrative guided tour with a librarian group consisted of five learners. A total of 12 learners were interviewed. Although the Web navigation system group and narrative guided tour with a librarian group did not use the GARNS for makerspace user education during the experimental period, the system’s functions and design concepts were explained to the interviewees of the two groups during the interview process. Additionally, learners in the Web navigation system group and narrative guided tour with a librarian group also experienced the GARNS after their respective experiments ended. Therefore, the three groups of learners understood their learning modes during the interview and provided feedback and opinions on the GARNS. The interview questions included: “Please explain separately whether the GARNS, web navigation system, and narrative guided tour with a librarian help you learn about the use of makerspace equipment. Why? What are the advantages and disadvantages of those learning methods? How can they be improved? Which one will motivate you more to learn?”.

This study analysed whether there were significant differences in learning effectiveness among three groups of learners: the GARNS group, the Web navigation system group and the narrative guided tour with a librarian group. The descriptive statistical analysis results of the pre-test, post-test and improvement scores of the three groups of learners on the “makerspace user education test sheet” are shown in Table 2. Before comparing the differences in learning effectiveness among the three groups, this study first examined whether there was a significant improvement in learning effectiveness within each group. Because the experimental sample size of this study was small, the Wilcoxon sign rank test, a non-parametric statistics method, was used to examine whether there was a significant difference in the scores of the “makerspace user education test pre-test” and the “makerspace user education test post-test” among the three groups of learners, to verify the effectiveness of the learning. The results are shown in Table 3. The test results indicate that the GARNS group (Z = 2.81, p = 0.005 < 0.01), the Web navigation system group (Z = −2.13, p = 0.033 < 0.05) and the narrative guided tour with a librarian group (Z = −2.38, p = 0.018 < 0.05) all showed significant differences in pre-test and post-test scores within the group. The post-test scores were all better than the pre-test scores. This result shows that learners could achieve significant learning effectiveness through the GARNS, Web navigation system and traditional narrative guided tour with a librarian for makerspace user education.

This study used the Kruskal–Wallis test to compare the improvement scores of the three groups of learners before and after learning to examine whether there were significant differences in learning effectiveness among the three groups. Before conducting the Kruskal–Wallis test, the Levene’s test for homogeneity of variances was first performed. The test results did not violate the assumption of homogeneity of variances (F = 0.42, p = 0.660 > 0.05), so the Kruskal–Wallis test was continued. The analysis results are shown in Table 4. The test results showed that there were significant differences in learning effectiveness among the three groups of learners (χ2 = 6.81, p = 0.033 < 0.05). Further analysis with the Bonferroni post hoc multiple comparison test showed that there was no significant difference between the GARNS group and the narrative guided tour with a librarian group (Z = 2.59, p = 0.212 > 0.05) and between the Web navigation system group and the narrative guided tour with a librarian group (Z = −1.14, p = 1.000 > 0.05). However, there was a significant difference between the GARNS group and the Web navigation system group (Z = 3.73, p = 0.036 < 0.05), and the GARNS group was significantly better than the Web navigation system group.

This study analysed whether there were significant differences in learning motivation among the three groups of learners. The descriptive statistical analysis results of the total score and scores of each dimension of the MSLQ for the three groups of learners are shown in Table 5. This study used the Kruskal–Wallis test to compare whether there were significant differences in learning motivation among the three groups of learners. Before conducting the Kruskal–Wallis test, the Levene’s test for homogeneity of variances was performed. The test results did not violate the assumption of homogeneity of variances (F = 0.72, p = 0.498 > 0.05), so the Kruskal–Wallis test was continued. The results of the Kruskal–Wallis analysis are shown in Table 6. The results showed that there were no significant differences among the three groups of learners in the total score (χ2 = 5.79, p = 0.055 > 0.05) and the emotion dimension (χ2 = 0.94, p = 0.626 > 0.05) of learning motivation. However, there were significant differences in the value dimension (χ2 = 6.10, p = 0.047 < 0.05) and the expectation dimension (χ2 = 6.70, p = 0.035 < 0.05) of learning motivation. Since there were significant differences in the comparison of the three groups regarding the value and expectation dimensions of learning motivation, a Bonferroni post hoc multiple comparison test was conducted. The results showed that there was no significant difference between the GARNS group and the narrative guided tour with a librarian group (Z = 0.38, p = 0.807 > 0.05) and between the Web navigation system group and the narrative guided tour with a librarian group (Z = −0.49, p = 0.560 > 0.05). However, there were significant differences in terms of the value and expectation dimensions of learning motivation between the GARNS group and the Web navigation system group (Z = 0.87, p = 0.048 > 0.05) and the GARNS group performed significantly better than the Web navigation system group.

The interview results found that most of the interviewees believed that the learning mode using the GARNS for makerspace user education could help them continuously judge, select and explore learning tasks. The challenges and peer competition inherent in games could help them improve their learning focus and exploring, and combining stickers to complete learning tasks could help them memorize practical knowledge and give them a more profound impression of the learning content. The interviewees who used the Web navigation system for makerspace user education stated that learning through mobile devices was very convenient, and the design of the learning content was very detailed and easy to understand. However, there was too much text, and learners may lose patience with the learning process if they are not interested in the content, leading to them easily forgetting the learning content. The interviewees who used the narrative guided tour with a librarian for makerspace user education stated that the librarian’s lead and explanation could allow learners to absorb knowledge quickly. For learners without relevant background knowledge of the makerspace, the librarian’s guidance could increase their knowledge and help them quickly understand the uses of various makerspace equipment. The disadvantage is that learning can only be done through the librarian’s explanation and visual observation, and it is often difficult to understand the functions and requirements of each equipment in detail due to the limitations of oral explanation.

The interviewees believed using the GARNS to support makerspace user education was challenging. The learning process can motivate learners through continuous experimentation and challenges, and learners can become more familiar with the tips and tricks to complete learning tasks as they progress through them, thus gaining a sense of learning achievement. In addition, the learning method of collecting and combining stickers to complete learning tasks is like assembling actual props by oneself, which can further immerse learners in the learning process and enhance their focus. However, some interviewees encountered issues with the system’s scanning, which affected the gaming experience. The interviewees who used the Web navigation system for makerspace user education preferred hands-on assembly and operation to deepen their knowledge of the equipment. They stated that using GARNS for makerspace user education is more attractive and helpful for learning than Web-based learning, which relies only on text. Some interviewees in the Web navigation group pointed out that there was too much text, which made it difficult to stay interested in the learning material. As a result, they tended only to browse the content quickly and were more likely to forget what they learned. They also found their level of focus and interest to be lower compared to learning through GARNS. The interviewees who used the narrative guided tour with a librarian method to assist in makerspace user education learning indicated a preference for learning through GARNS. In addition to being more attractive, GARNS can enhance learning interest through interaction, making it livelier than a narrative-guided tour with a librarian learning. However, some learners also expressed that, although GARNS learning would attract them, they still preferred to learn through a librarian’s guidance because it could quickly help them understand the learning content.

This study explores the effects of using a GARNS, a Web navigation system and a traditional narrative guided tour with a librarian to supplement the makerspace user education on learners’ learning effectiveness and motivation. Firstly, a learning effectiveness analysis was conducted on the pre-test and post-test of the three groups. The results showed that the post-test scores of all three groups were significantly higher than the pre-test scores, indicating that all three learning modes were effective. Furthermore, significant differences in learning effectiveness were found among the three groups. Bonferroni post hoc multiple comparisons tests showed that there was no significant difference neither between the GARNS group and the narrative guided tour with a librarian group nor between the Web navigation system group and the narrative guided tour with a librarian group, but using the GARNS was significantly better than using the Web navigation system for makerspace user education. This result is consistent with Chen and Tsai’s (2012) study, indicating that using the proposed ARLIS for library user education resulted in the same learning effectiveness as conventional librarian instruction as well as overcome the shortcomings of personal teaching skills of librarians that may adversely affect student learning effectiveness even though conveying the same learning content to all students. However, this result differs from Marcus and Beck’s (2003) research, which found that game-based library user education had significantly better learning effectiveness than traditional teaching methods. In addition, from the interview results, it was found that most interviewees believed that the learning mode of using the GARNS could help them continuously judge, choose and explore learning tasks. The challenge and peer competitiveness inherent in games could help improve learning focus. Exploring and combining stickers to complete learning tasks could effectively help learning retention.

In addition, the results of this study showed that among all learners, there was no significant difference in their overall learning motivation across the three groups; however, there was a significant difference in the value dimension, with the GARNS group showing significantly higher scores than the Web navigation group. This indicates that learners perceived the overall value of the GARNS to be significantly greater than that of the Web navigation system. Learners were more curious about the challenges presented by the GARNS, had a greater desire to achieve good results and perceived the importance and interest of this learning mode than that of the Web navigation system. Additionally, there was a significant difference in the expectation dimension, with the GARNS group having higher scores than the Web navigation group. This indicates that learners’ overall expectations for success with the GARNS were significantly greater than those for the Web navigation system. Learners were more likely to attribute their learning success to themselves rather than external factors, had stronger self-efficacy and had more substantial expectations for learning success with the GARNS. This result is consistent with Erbas and Demirer’s (2019) study, which showed that AR learning modes can help learners understand abstract concepts and enhance their self-efficacy. However, there was no significant difference among the three groups in the emotion dimension. The study suggests that some learners might find the GARNS challenging at the beginning of the game and feel nervous due to time pressure and difficulty level. Therefore, they do not significantly differ in the emotion dimension compared to the other two groups. In addition, some interviewees mentioned that they spent more time searching for the location of items at the beginning of the game and felt pressure if they fell behind schedule. However, some interviewees stated that the challenge of the GARNS would stimulate their learning motivation and desire to challenge themselves and that the constant attempts and failures could motivate them to learn. Moreover, learners could become more familiar with the tips to pass the task as they gained a sense of achievement. Almost all interviewees mentioned completing the task was their favourite part of the game process.

Finally, some limitations of this work merit further consideration. Firstly, the study referred to the equipment of the makerspace of the Dah Hsian Seetoo Library at the National Chengchi University Library in Taiwan to design the proposed GARNS and teaching content. The research results may not be overly inferred from the utilization and teaching of other makerspaces. Secondly, due to the library’s closure in the later stages of this study caused by the COVID-19 pandemic, the experimental time and sample size of this study were limited, thus restricting the generalizability of the research findings.

This study combines game-based learning with AR tools to develop a GARNS suitable for the characteristics of makerspaces, providing an innovative and compelling learning mode. The research results show that learners using the GARNS for makerspace user education achieve significantly better learning effectiveness and motivation in value and expectation dimensions than those using the Web navigation system. In addition, most interviewees expressed their belief that using GARNS for educating makerspace users could assist them in consistently evaluating, choosing and discovering educational tasks. The obstacles and competition among peers that come with gaming can enhance their focus on learning and exploring while combining stickers to accomplish learning tasks can significantly enhance learning retention, leading to a more profound understanding of the subject matter. This study promotes popularizing makerspace user education and diverse learning models.

Based on the implementation of experiments supporting makerspace user education using the GARNS, this study proposes suggestions for librarians to apply this system in library user education in the future. Develop a blended learning model by combining three different guide modes: the GARNS, Web navigation system and narrative guided tour with a librarian. Use the Web navigation system or narrative guided tour with a librarian to give learners a preliminary understanding of the makerspace, such as the types of equipment and devices, to help learners use the GARNS more quickly and deepen their learning of makerspace equipment use. At the same time, this approach can address Web-based learners’ inability to focus and remember due to too much text and enhance the detailed understanding of equipment for learners guided by staff. The GARNS developed in this study focuses on puzzle-solving game design, and the information presentation interface is mainly based on images provided as clues. If the information presentation method can be changed to 3D objects, it can display the details of machine-made products or the form of collected materials. This can increase learners’ impressions and interest in learning through 3D objects and also enhance learners’ immersion in the game.

Suggestions for future studies are as follows. Firstly, through the interview results, this study found that the three learning modes of GARNS, Web navigation system and narrative guided tour with a librarian, were all favoured by learners in the makerspace user education. Learners who already enjoy hands-on work, interaction with others and game experiences may prefer to use the GARNS for learning. Learners who are quieter and do not enjoy interacting with others may prefer to read and learn independently by using the Web navigation system. Learners who want new knowledge and prefer passive learning quickly may prefer to use the narrative guided tour with a librarian for learning. Therefore, considering that different learning styles will affect learners’ learning effectiveness and motivation (Atma et al., 2021), future research can incorporate learning styles into the background variables and explore the impact of the GARNS on learners with different learning styles’ learning effectiveness and motivation. Secondly, considering learners may come to the makerspace alone for learning, this study designed the GARNS as a widely applicable single-player challenge learning mode. However, this study found that the GARNS is challenging for some learners. Therefore, learners can work together to complete tasks if the GARNS can be expanded into a cooperative learning mode (Jong et al., 2012). For example, all cooperative members have their tasks to complete. However, to achieve the final learning task, learners must assist each other and learn and share their experiences or knowledge. In this way, high prior knowledge learners can help low prior knowledge learners complete learning tasks, which will help to reduce the prior knowledge gap among group members and improve the overall learning effectiveness of learners. Thirdly, to enhance the contribution and applicability of the research, it is recommended to include a more extensive and diverse sample encompassing various makerspaces in future work. Finally, to provide a comprehensive understanding of the factors influencing learning effectiveness, the study could benefit from considering additional variables, such as the users’ personalities. Taking into account individual differences in learning styles and preferences would provide valuable insights and contribute to a more nuanced understanding of the effectiveness of the proposed GARNS in different user contexts.

The authors would like to thank the Research Center for Chinese Cultural Metaverse in Taiwan for financially supporting this research under Contract No. 112H21.