Keywords

1 Introduction

Today’s technologies are rapidly advancing, causing a revolution in the nature of warfare. These dramatic changes demand quicker and more accessible training methods for preparing our military forces for combat. As a result, Virtual Environments (VEs) have been implemented as a supplement for traditional training methods. According to Schroeder [1, 2], VEs are “computer-generated display[s] that allow or compel the user (or users) to have a sense of being present in an environment other than the one they are actually in, and to interact with that environment” (p. 25). These environments can accommodate individual and team training through transportable and cost-effective means [3]. While there has been much progress in developing training that prepares warfighters for their military operations, there is an urgency to explore the benefits of VEs for Soldiers transitioning from military to civilian life. As of September 1, 2014 the total number of major limb amputations due to battle injuries, between the years 2000 and 2014, is 1,573 [4]. These individuals are returning home with new disabilities that involve embracing many changes (i.e., physical, psychological, and emotional). Fortunately, emerging technologies in the area of Clinical Virtual Reality are being researched in hopes of addressing the challenges these Veterans and their families face in their daily lives.

Past research suggests that virtual environments and avatars are viable means for clinical assessment and intervention [5–8], especially for achieving enhancements in the area of psychomotor skill development [9]. In addition, research findings indicate that the manner in which the virtual representation, or avatar, of an individual is presented can change his or her behaviors and beliefs [10]. This leads to the question, “Does an avatar’s appearance effect an individual’s ability to learn?” According to Baylor [11], the ability to blend our real physical characteristics with a virtual avatar has potential for motivational and affective impacts. This paper aims to address this question by providing an overview of three types of virtual avatars (i.e., generic, highly recognizable subject matter expert (SME), and doppelganger) and current research conducted on their applications. In addition, a use-case scenario is provided to illustrate the benefits of applying this virtual approach to physical therapy. The findings of this research will be used as a theoretical foundation for further investigation on the effects of an avatar’s appearance on a learner. Finally, this research will expand the knowledge base of several training domains, such as the military, rehabilitation, high performance athletic training, etc.

2 Avatars

An avatar is any entity that becomes a representation of its user; an icon, a pawn in a game board, or a computer animation are examples of avatars even though they do not necessarily resemble or behave like their user [12]. A 2-dimensional (2D) or 3-dimensional (3D) computer animation that represents a human in a VE is referred to as a virtual human or agent [13]. Agents differ from avatars in the manner in which they are controlled; an avatar is typically controlled by a human, while an agent is controlled by computer algorithms [13–15].

Recent research efforts have outlined the importance of accurately modeling human behavior cues onto 2D and 3D models in VEs [16–18]. Baylor [11] argues that the appearance of an avatar is a vital element for promoting motivation for learning. In addition, research conducted by Jeremy Bailenson and colleagues at Stanford University suggests that individuals change their behaviors according to the appearance of their avatar [10, 12]. Appearance for the purposes of this paper will be operationally defined as the physical attributes of an avatar represented in the VE. The subsequent sections will provide an overview on the three different avatars of interest. Images were developed and customized in house for use in this paper.

2.1 Virtual Human

As previously stated, a virtual human is a computer-generated model that appears as a human in a VE. It is a generic representation that does not represent any specific user. These entities are typically customizable; a user can change certain features of the avatar (e.g., hair color, eye color, clothing, etc.). Figure 1 provides an example of a virtual human. Research indicates that a virtual human that is the same sex, race, or similar behaviors as their users provoke learning [19, 20].

Fig. 1.
figure 1

Example of a virtual human

Full size image

2.2 Subject Matter Expert

SME, also known as a domain expert, is an individual that is highly knowledgeable in a specific topic area. For example, a SME on the topic of American football can be a coach, a football player, or an athletic trainer. These individuals are authorities in their field of study. Figure 2 displays an example of an American football SME avatar. Limited research is found on the effects a SME avatar has on an individual’s learning outcomes.

Fig. 2.
figure 2

Example of an American football SME Avatar

Full size image

2.3 Doppelganger

The word doppelganger is German for “double” or “look-alike” [21]. For the purposes of this paper, a doppelganger is defined as a virtual representation of the self [21–23]. Often times in video games, a player can design his or her avatar to resemble themselves. For example, the Nintendo Wii allows individuals to design their “mii,” an avatar, to look like them. These avatars are typically controlled by the player. Once an avatar resembles its user, but it is not controlled directly by the user (meaning it is controlled by a different user or computer algorithms), then it is consider a doppelganger. Figure 3 displays an example of an individual (right) and his doppelganger (left). Figure 4 is an addition example of an individual (left) and his doppelganger (right). Research has shown that individuals that watched their doppelganger exercise expressed and acted upon their intention to exercise as well [22]. This suggests that doppelgangers may invoke a sense of motivation in their users.

Fig. 3.
figure 3

Example of a doppelganger and the represented individual (Authorization and consent to use the photograph in this figure were obtained from the live model)

Full size image
Fig. 4.
figure 4

Example of an individual and his doppelganger (Authorization and consent to use the photograph in this figure were obtained from the live model)

Full size image

3 Application Domain

Physical therapy or physiotherapy, often referred to by its abbreviated acronym PT, is a health care profession that seeks to bring about positive changes or enhancements to physical impairments. Physical therapists are licensed professionals that provide intervention strategies that teach their patients various techniques to reduce pain, restore function, and prevent disability [24]. They operate in a variety of settings, including hospitals, private practices, schools, nursing homes, as well as sports and fitness facilities.

Improving the quality of life for motor rehabilitation patients depends mostly on three factors, which are intervention, task oriented training, and repetition [26]. However, several issues in regards to traditional PT have emerged, including time and accessibility. To take full advantage of their treatment options, patients must travel to a clinic or medical office. This implies that once a patient returns home the ability to access the support of their therapist is limited [26]. Patients also report feeling boredom and fatigue when undergoing traditional PT [25, 27]. Fortunately, virtual reality (VR) has successfully integrated within the domains of medicine, rehabilitation, and psychology to offer the multimodal environment needed to individualize patients’ treatments [25, 26]. Morie et al. [27] suggest that virtual worlds may provide the patient the desired constant accessibility to their treatments, as well as the ability to visualize their progress. Boredom and fatigue seem to mitigate with the use of VR for therapy; Research has suggested that VRs provide enjoyment from the participant [25].

The following use-case scenario depicts a fictional scene for the primary user group (i.e., physical therapy patients), capturing how the end-users are expected to interact with the proposed virtual approach to PT. The scenario provides insights on key elements of the system, including its purpose, functional capabilities, and safety measures. Furthermore, it sets the stage for comprehending the benefits of using avatars and virtual technologies for physical therapy that support the rehabilitation of Soldiers, athletes and others.

3.1 Use-Case Scenario

Fred is a 36 year-old former U.S. Coast Guard Soldier. He is the husband of Chloe and a father of 4 (Jack, Isabella, Christian, and Desiree). After 16 years of military service he was injured in battle and recently underwent total Anterior Cruciate Ligament (ACL) reconstruction surgery. Although the doctors feel that the surgery went exceptionally well, Fred is experiencing severe pain and discomfort when attempting to straighten the knee out 90 degrees.

After receiving daily 1 h sessions of traditional PT for three weeks, Fred seems to be making little, if any, progress on getting his knee beyond 45 degrees. His physical therapist believes it is Fred’s lack of motivation and unwillingness to complete the exercises that are preventing him to progress.

After examining the x-rays and magnetic resonance imaging (MRI) results of Fred’s surgically repaired knee, the doctors and therapists agree that there is nothing physically impeding Fred’s ability to straighten out his knee. Therefore, they recommended that Fred undergo a series of Virtual Physical Therapy (VPT) sessions, with the intention to alter Fred’s state of mind and provide him with the necessary motivation to improve the movement of his knee.

During the VPT sessions, Fred interacts with his doppelganger. The doppelganger highly resembles Fred’s physical persona and has undergone the same surgery as Fred. It illustrates to Fred the different exercises needed to straighten his knee. The avatar takes Fred through the before, during, and after phases of the surgery in hopes of eliciting motivation and guiding Fred to improvement. Fred enjoys the VPT sessions because it incorporates his three P’s (i.e., portable, personal, and practical). Fred is able to access this training at home and at the therapist’s office. The software also captures his progress and allows Fred to review it. The therapist is also alerted when Fred has completed assigned exercises.

3.2 Limitations

Several limitations were identified throughout this investigation. As the aesthetics of avatars becomes more human-like, the theory of the uncanny valley should be explored. This theory, coined by robotics professor Masahiro Mori [28], posits that our affinity towards human-like avatars increases to a degree and then a sense of revulsion appears. Although it has been widely cited, there is still much to explore. Current research is being conducted on producing a set of heuristics for avoiding the uncanny valley [28, 29].

Another limitation found in the research is accurately modeling behavior using virtual avatars. The mobility of an agent is directly connected to believability [18] and believability is important for training transfer. Research focusing on the appearance of avatars will need to take into considerations the limitations in the mobility of the avatars chosen. Future investigation should look into determining the best suited platform for incorporating virtual avatars for the purposes of physical therapy.

4 Conclusion

Clinical Virtual Reality, specifically in the area of rehabilitation, is a relatively new area of research and development. Many researchers have focused on incorporating VEs or VR into healthcare professions, however, there is still limited research on the area of utilizing virtual avatars for therapy. Future research should focus on investigating the impacts of virtual therapists and the effect an avatar’s appearance can have on the patient. Ultimately, this use-case scenario provided in this paper conveys the need for examining how the manipulation of an avatar’s appearance affects the psychomotor learning obtained by the patient.