Interceptive motor skills are complex tasks which require spatial and temporal accuracy. For intercepting a fast moving object, the movement of the end-effector needs to be initiated and controlled in such a way to end at the right place and time to intercept the object and return the object to a specific location. This has to be done with one single action and mostly within an extremely short period of time. Therefore, an interceptive motor skill perhaps demands more control of movement compared to non-interceptive motor skills. Generally, these interceptive motor skills are acquired and practiced task-specifically in the real-world, by just repeatedly performing the motor skill on the training-field. However, the opportunities to train in virtual reality (VR) are developing fast.
Virtual reality (VR) environments are artificial settings simulated as real life (RL) in which a person can play or train a certain task, and are increasing in popularity in our nowadays lives. The development and impact of these systems is large. The start of training simulators being used for (complex) tasks was at the end of the Second World War (Gopher, 2012). In the beginning, this mainly consisted of of 2D images/digital games, primarily focused on fine motor skills (Wiemeyer & Hardy, 2013) . Later on, bigger companies made it possible for the public to use VR systems (Xbox Kinect, Wii, google cardboard) which utilize whole-body movements for playing games. With the development of these new interfaces, the opportunities to implement VR in sports became bigger and bigger.
A VR environment offers multiple benefits. According to Connoly et al. (2012) […] “modern theories of effective learning suggest that learning is most effective when it is active, experiential, situated, problem-based and provides immediate feedback” (p.661). Generally, everyone goes through the different stages when acquiring a motor skill, with a decreasing amount of errors and variability over the three stages of learning. The first being the cognitive stage, marked by a lot of large errors and a high variability. The second is the associative stage with fewer and smaller errors, where one acquires the basics of the motor skill and becomes more consistent. The third is an autonomous stage where one can perform another task at the same time with very small variability between attempts (Magill & Anderson, 2000). The key ingredients from Connoly et al., (2012) are all present in a virtual environment and can provide help with acquiring learning for all the stages. The information from the VR environment could contribute to identifying cues, train of actions, and generating spatial memories which can all be repeated in training and therefore could enhance the RL performance (Rose, Attree, Brooks, Parslow & Penn, 2000).
Sports scientists and coaches are continuously trying to improve the interceptive motor skills of athletes. RL training requires repeating the similar movement lots of time to reach perfection. The repeating can lead to overuse injuries, so therefore VR could be a good training method for them too. It is well-known that practice leads to improvements, but what kind of training can result in the highest improvement of the motor skill? Transferability from task A to task B, where a transfer between ‘identical elements’ could occur between different tasks, is crucial to enhance these motor skills (Thorndike, 1913). The identical elements may be ‘physical, anatomical or biomechanical, perceptual or conceptual’ (Schmidt & Wrisberg, 2000). For many years, this explanation of transferability enabled by shared elements was dominant. Henry (1968) proposed that “the highest amount of transfer to the on-field performance will be derived with training in a way that most closely resembles the specific motor skill in the specific environment” (p.328). Rienhoff et al. (2013) for example demonstrated a transfer from a basketball throwing task to the task of throwing darts, where skilled basketball players scored higher on accuracy during a novel dart throwing task compared to less skilled players. However, nowadays transfers (task A to task B) can not only be acquired in the real world, but also from a virtual to a real world. In order to create a successful VR training, the training must provide the chance to obtain information in a way that is similar to the real world.
So far, research about the transfer of VR, mainly focused on rehabilitation. Different studies indicated that VR is beneficial when a person has to regain their ability to perform specific movements (Anderson, Annett and Bischof, 2010; Gomoll, Pappas, Forsythe and Warner, 2008). Moreover, Calatayud et al. (2010) showed that VR could also help to further improve the execution of a specific movement. In the study of Calatayud et al. (2010), they showed an improvement in laparoscopic performance after VR training. Laparoscopic surgery is a medical procedure that is difficult to train in RL. Additionally, VR is used for teaching pilots and soldiers particular motor skills, so they could train dangerous tasks in a safe environment. (Rose et al., 2000). Potential explanations for these improvements could be an enhanced eye-hand coordination, visual attention and decision making (Green & Bavelier, 2003). These factors are also important in sports. Previous research indeed showed promising results of athletes who used VR specific training to improve motor skills. A transfer occurs from a virtual to a real environment in static sports, like darts. Tirp et al., (2015) for example showed an improvement in throwing accuracy after 150 throws on a virtual dartboard where they had to focus on the bulls eye.
However, VR has currently only been studied in static, non-intercept games and not in fast-paced intercept motor skills. Earlier research already focused on the perceptual transfer during table-tennis (Sohnmeyer, 2011). VR training is nowadays rarely proposed in the motor learning setting (Tirp et al., 2015), while it could be an interesting training tool for athletes. Athletes could repeat a certain motor skill multiple times in multiple ways, which is in specific circumstances difficult in the real-world setting, and therefore makes VR a possible training tool for athletes (Tripicchio, 2012; Rose et al., 2000). Another benefit of training in a VR environment is the possibility to adjust the training to the individual. The context of the environment could be manipulated in a specific way to focus on the constraints of the athlete (Tirp et al., 2015). In this way VR could possibly contribute to an improvement of the on-field performance. VR training can also be used in situations when it is hard to train in the real-world setting, because it is too dangerous (in case of injuries for instance), too difficult to control or too expensive (Rose et al., 2000). One current example of this is Max Verstappen in Formula 1 racing and his race simulator practices. The VR training gives him the opportunity to train his cognitive, visual and motor skills that are transferable to a real life racing track. Therefore, it is important to look at the effect of VR on a fast-paced intercept task.
During fast-paced intercept games, subjects need to intercept and return a fast moving object with a single action. “The athlete has to control the right amount of force, has to intercept the ball at the right place and time and has to project it to the desired location” which require spatial and temporal accuracy (Van der Kamp, Rivas, Van Doorn & Savelsbergh, 2008). In order to achieve a successful fast-past interception, visual anticipation is an important determinant. This can occur due to a constant interaction between the ventral and dorsal stream in the brain. The ventral stream provides a representation of the environment and the dorsal stream uses this information to initiate an action (Magill & Anderson, 2000). Researchers mentioned the importance of intercepting a real, actual object for getting the interaction between the ventral-and-dorsal pathway, in which the visual anticipation of experts was better when making bat-ball contact in criquet (Mann, Abernethy and Ferrow, 2010). Visual anticipation is important for the visual-motor skills. Based on the identical element theory of Thorndike (1913), the similarity based approach of Henry (1968) and the ventral and dorsal association of Mann et al. (2010), the question arises intercepting motor skills can improve without actually hitting the ball in real life.
If VR shows to be an effective way of training in fast-paced intercept sports, it would give a lot of opportunities to improve motor skills away from the field. VR may offer opportunities to create individual training settings, optimizing the specific constraints of the athletes (Tirp et al., 2015) or to replace real-world training in case of an injury. This will be studied by training four groups of novice table-tennis players: a real life training group (RL), which will show the effect of regular on-field training, a virtual-reality training group (VR), which shows the effect of training in a virtual reality environment instead of a real life environment, a group who will perform both type of training (VRRL), which shows the effect of training in virtual reality on top of regular training and a control group (CO), who will perform no training at all. All of the participants will be playing with their non-dominant hand since this will equal out the tennis table playing history in all participants. In this way, VR will be measured as both a replacement of on-field training, which could be beneficial in case of injuries, and an addition to on-field training, which could be used to further improve the performance. Performing both types of training increases, the amount of training. Training in a virtual reality could be less physically demanding and therefore the training volume could be increased without risking too much strain on the body.
The aim of this study is therefore to look at the effect of VR as a training tool for fast-paced interceptive ball-sports, as both a replacement and/or an addition to RL training. Our first hypothesis is that the improvement in the RL training group will larger than the improvement in the VR training group. The second hypothesis of this study is that the VRRL training group will have a larger improvement than the RL and VR training groups since they are performing double the amount of training.