This research is supported by the continued collaboration with a Southern State University, and a Southern Art Museum to introduce Augmented Reality immersive learning experiences to a broad audience. This study introduces iterative cycles of Augmented Reality (AR) user behavior data collection at an Art Museum in the Southern United States and outlines the types of Augmented Reality experiences, personalized learning applications, and user behavioral pathing captured by an embedded user analytics software development kit within an informal mobile learning (mLearning) application platform. AR is defined as information, content, images, and multimedia that has been programed and superimposed onto real-life content and locations (Milgram, Takemura, Utsumi, & Kishino, 2004; Mann, Furness, Yuan, Iorio, & Wang, 2018). Unlike Virtual Reality (VR), which creates a completely artificial environment, Augmented Reality uses the existing environment and overlays new information with the use of a mobile device, tablet, or smartphone. AR combines interactive virtual and programmable content with images, learning artifacts, digital spaces, physical locations, and digital objects mediated by the mobile device's responsive graphical user interface and view screen (Wojciechowski, Walczak, White, & Cellary, 2004). Building upon an adaptive learning and technology mediated instructional framework, where the AR system “learns” from student interactions and then adjusts the path and pace of learning and mediated content (Educause, 2017). The Instructional Design and Educational Technology Augmented Reality Transmedia Stor ytelling (IDET ARTS) application includes an embedded adaptive analytic data collection model developed by the researcher to allow the rapid adoption and implementation of static objects into rich learning objects and enable user behavioral analytic tracking within a physical environment with the added appearance of virtual elements mixed in with the environment. The IDET ARTS interface is a simple view screen model, where the user simply points their smart device at the mediated content. The mediated content is deployed based via a graphical user interface tailored to museum staff for rapidly tailored and adjustable augmented content and exhibit site overlays. A prototype of the application is available to download on iOS devices in the Apple iTunes App Store by searching IDET ARTS.

The use of mobile applications to facilitate the learning experience is defined as mLearning (mobile learning). mLearning is quickly becoming a bridge between formal, informal, and lifelong learning (Montalto-Rook, Asino, & Thanomsing, 2010). The combinations of powerful computing mobile devices and mLearning offer an attractive medium for use in museums Augmented Reality and AR mobile learning both expresses and changes user's relationship to physical and digital learning spaces. Virtual elements, including but not limited to audio, video, and multi-mediated forms of enhanced media can be synchronized and superimposed unto real places and objects. Augmented Reality both expresses and changes user's relationship to the real and virtual world. Real-world boundaries between the physical and digital world become intertwined, and intertextuality can be designed and simulated (Rhodes, 2015).

Personalized learning using AR has also been found to be effective and motivating. Researchers, Salmi, Thuneberg, and Vainikainen (2016) posit that concrete AR-learning sessions from informal sources and non-traditional learning environments is often most beneficial for the lowest achieving learners, despite the higher achieving learners also learning. This is underscored as an encouraging result for developing new methods for at risk learners, while also supporting high performing learners by receiving new, and better learning results (Salmi et al., 2016; Torres-Ruiz et al., 2018). AR integration underscores and highlights a form of leveraging content in a novel way that supports learning and increases learner's engagement with content and domain specific learning material.

There is limited empirical research investigating mobile Augmented Reality Smartphone apps in traditional museum informal learning environments (Tomiuc, 2014). True mobile Augmented Reality applications are now vastly different from ontology-based models with Augmented audio reality system for museums (Hatala & Wakkary, 2005). These antiquated and outdated models require bulky and cumbersome installations and costly technology step up. However, presently a robust mobile Augmented Reality application based immersive learning system simply requires users to download a researcher developed freely available application from a commercial App Store and then aim their device at the AR-enabled exhibits. Thereby drastically reducing user cognitive load and the required steps to access learning materials and that digital augmentations can help in conceptual development of knowledge (Yoon, Elinich, Wang, Steinmeier, & Tucker, 2012; Capuano, Gaeta, Guarino, Miranda, & Tomasiello, 2016M; Squires, 2018; He, Wu, & Li, 2018).

Participants were presented with directions at the museum site on how to download and utilize the AR application. The app for Apple mobile devices and tablets which when pointed to at the artwork or artifacts, not only offers added information about a work of art or artist or curatorial input, it also allows for data collection and analytics of visitor usage of the application. Visitors are allowed to use their own devices or check out iPads from the museum gift shop for use while in the galleries: See Figure 1 (a, b, c). Additionally, the interactives are in use in the Museum's “Creation Station” education wing and are repurposed for each upcoming exhibition. Data was collected via the app for three months. Participant application usage data and behavioural analytics were recorded in tandem with any self-reported open-ended survey feedback within the app collection interface. Participants opted into the research study via the IDET ARTS application interface and the app is available for free on the iTunes app store. The IDET ARTS platform tracked and reported participant behaviour, what AR content was viewed at the museum site, time-on-task, and average session duration of use on either an iPhone or iPad. A qualitative questionnaire, accessible within the app itself by pressing the app's sail icon, recorded concurrently the participants' responses and perceptions.

Figure 1. (a, b, c) AR Museum User Aiming Device at Enabled Artifact

The embedded IDET ARTS Google Analytics software development kit integration reported a total of 149 unique users between February and April over the course of the installation. Users on average viewed 4 - 5 AR enabled trigger programmed sites at the museum. As shown in Figure 2, users' average session duration varied per day.

Figure 2. App Analytics Report - Avg. Session Duration

As shown in Figure 3, on average users viewed 4.35 screens when the app was open. Users viewed a combined 674 content triggers at the museum over three months. More than half of users moved between the main AR view to the web browser screen via the app Sail icon. The web browser screen showcases what museum artifacts contain AR enabled story caches and website links.

Figure 3. App Analytics Report - User Behavior

As shown in Figure 4, the majority of users, 85.91%, were on iPhones with 14.09% of users opening the app on iPads.

Figure 4. Device Models

As shown in Figure 5, a majority of users returned to the app after opening, or completely exiting out of the IDET ARTS user interface.

Figure 5. App Analytics Report - New and Returning Users

Users matched with Unique User Identification (UUID) numbers showed that the AR museum users (n=149) viewed multiple AR overlays and interacted with the content by engaging with the overlaid matter and pressing the embedded application sail icon. The participant path also shows that the same users returned to use the app again more than 76% of the time. The majority of users (85.91%) accessed the AR with their iPhone; however, an increased number of users accessed the application via iPads (14.09%). This increase may have been due to the Museum implementing an iPad checkout program. A documented limitation with using smartphones is the small screen size; however, utilizing larger tablets and iPads can potentially help to mitigate this limitation. Based on the imbedded application analytics, data users viewed the AR exhibits’ interactions spending time to engage with the programmed AR interactions and audio storytelling features. Overall learners who are existing users returned to the app 76.4% and spent 149 users viewed AR artifacts 675 times with an average view rate of 4.35 AR artifacts per user. Users data collection shows engagement and immersion within the Museum exhibit AR tour informal learning environment. Moving through the museum at their own pace participants were free to interact with the artwork and exhibits that interested them and could further unlock content and digital information that enhanced their experience. The integration of mobile augmented reality with Museum artifacts and AR enabled programmed stories caches ultimately offered participants an enhanced experience.

Augmented Reality integration within a static learning environment is an effective engagement strategy. We can extrapolate from the data that users are engaging with Augmented Reality in informal learning environments, triggering multiple digitally enhanced artifacts, and returned to use the AR application consistently. While the content programmed must be rigorously curated for each specific learning environment, the enabled story caches enhance the user experience and interaction with the learning environment. Furthermore, with detailed imbedded application analytics, we are able to pinpoint and specifically measure participants’ use of the mobile AR application, their time on task, and how long each learning artifact was engaged with within the museum site. This measure helps to shed light on a user behavioral pathing and ultimately allow users to engage in a more personalized experience at the Museum. The ability to choose and engage with artifacts and art work that individually appeals to the user allows an increased level of interactivity, by enabling pacing options that facilitate user choice. That is the ability to simply points one's smartphone or tablet at work and listen to and engage with the curate content be hand the painting or artifact. This also enables user's choice by allowing freedom of movement with the informal learning space. Giving users the freedom of choice and the option to move from AR enable artifacts at their own pace, and without interruption unless they choose to move faster, slower, skip works, or linger longer at works that are based on their own personal interests. Based on the unique identifiers and device tracking data, we discover that users spend their time immersed in the storytelling components of the AR tour and mapping their movement through Museum exhibit site we can being to extrapolate a full picture of the engagement potential of Augmented Reality as a storytelling, personalized learning, and user experience facilitation input medium.

More research is needed with augmented reality specific data collection devices. Also, continued measurement of the iPad checkout program and whether the larger screen size is an ideal viewing experience for users that do not have access to an iOS device. The next iteration of this project will analyze the Augmented Reality specific functionalities related to user experience and the overall impact on learning and engagement comparing audio AR and Video AR overlay artifacts related to user experience, self-efficacy, and engagement. Further development will involve a fully-equipped Lab space within the Museum and will provide to the public the ability to not just receive an augmented and virtual reality experience, but design and build immersive AR/VR/MR experiences for their own use. Students and instructors will have the ability to learn and teach the fundamentals of XR, which brings all three Realities (AR, VR, and MR) content creation and design to other students, instructors, and the public. Further qualitative questioning is planned for future iterations and participant dialogues to be facilitated in a mixed-methods approach combined with open-ended survey data collection.

Acknowledgments

The author would like to thank Joseph Schenk, Director of the Art Museum of South Texas, Sara Morgan, Assistant Director, Deborah Fullerton, Curator of Exhibitions, Karol Stewart, Coordinator of Community Services, Sheri Emerick, Development Officer, Linda Rodriguez, Curator of Education, and Curatorial Assistant Angela Resendez.