Specifics of training in virtual reality
Specifics of training in virtual reality
Introduction
It is worth noting that even with the relatively high cost of VR equipment and the complexity of setting it up, we see how every year new models of VR glasses enter the market and hundreds and thousands of different simulators are created for consumer needs, and every developer in this field, as well as most of their clients, are confident in the effectiveness of this method.
Also, unfortunately, the implementation of a virtual reality simulator in many companies ends with several demonstrations to employees or guests of the company at exhibitions and presentations, which has nothing common with training. In fact, in my experience, ordering and creating a simulator, the question how effectively knowledge is absorbed or consolidated is asked not so often, and for some reason there is absolutely no need to prove the usefulness of this technology, not to mention whether it will be more effective in this case than traditional methods training. I’ll try here to raise several important problems of virtual learning and answer what my team learned, solving these problems based on the mistakes we made along the way.
About us
We’ve come an interesting path in creating virtual reality simulators, and we had the opportunity to create a few tens full-fledged simulators for students of several hundred educational institutions.
Thousands of students aged 14 to 18 have completed our interactive virtual reality training sessions. Our simulators were built into the traditional education model, which allowed us to obtain data from subjects, adapt a real school curriculum to VR and analyze a large amount of data from passing.
We used action analysis and testing to better understand how a person feels, what causes discomfort, and what actions will be useful and which are harmful and interfere.
We have also implemented several projects in augmented reality. Despite the fact that AR has great potential, at the moment we are more focused on VR, as our main goal is to teach something or consolidate already acquired knowledge, which we managed to do the best in virtual reality.
In this article and this blog, I would like to share with you how we analyze thousands of passing of VR simulators in order to clearly show you, not only visually but also with specific numbers, that virtual reality technologies are not only about the “Wow” effect, which if performed well, it will provide you with vivid impressions, but also an effective method of acquiring and consolidating knowledge.
It’s also needed to say that in the past we weren’t a company producing games and had nothing to do with either training or simulators in general. Everything we did — we did for the first time, virtually blindly, and much more had not been done by anyone except us.
We made a lot of mistakes, I would say a lot. We paved this path by making mistakes in the development, training approach, choice of equipment and much more, in fact, this is the main reason why I decided to share with you what we learned useful from these mistakes.
It’s worth clarifying that the actual use of virtual reality can be divided into several main areas:
Games
Actually the locomotive of virtual reality with the greatest hopes, and releases like half life alex perhaps justifies not only the potential, but also the real application of VR games and the B2C segment in general.
Marketing
As a sales element to attract attention to a problem or demonstrate the potential of something. This technology showed itself especially well here due to the strong user involvement in the content transmitted to him and the “wow” effect, especially if the person wasn’t previously familiar with this technology or has little experience with it.
Simulators
I pay the most attention to this section, because this is the main direction of our activities.
Next, I will mention a few obvious points that we identified, but in future articles I will analyze them in more detail and provide more analytical data and specific facts.
Where to start learning in virtual reality?
The process of learning and learning in a new environment can take some time. First, you must let the person get used to the new environment, nothing is required from him, then you can start teaching how to use the controllers and turn on the interactive for everyone, while the first 2-3 tests shouldn’t take more than 3 minutes. These aspects consider the understandable features of the prospects, which in the future will need to be adjusted to him. Starting with optimal control mode, sitting or standing position and overall tolerance of the vestibular system.
Trying to teach something in these first moments won’t turn out to be anything good, and if in real life an unfamiliar environment can help in mastering decision-making, so this is a completely different story, because most of the time you will be confused and your brain is busy trying out all the possible combinations of buttons on controllers and adaptation to the new environment.
While first meeting with virtual reality, a person doesn’t learn anything.
The process of learning and learning in a new environment can take some time. First, you must let the person get used to the new environment, nothing is required from him, then you can start teaching how to use the controllers and turn on the interactive for everyone, while the first 2-3 tests shouldn’t take more than 3 minutes. These aspects consider the understandable features of the prospects, which in the future will need to be adjusted to him. Starting with optimal control mode, sitting or standing position and overall tolerance of the vestibular system.
Trying to teach something in these first moments won’t turn out to be anything good, and if in real life an unfamiliar environment can help in mastering decision-making, so this is a completely different story, because most of the time you will be confused and your brain is busy trying out all the possible combinations of buttons on controllers and adaptation to the new environment.
While first meeting with virtual reality, a person doesn’t learn anything.
Next, when a person is adapted to the new environment, we can begin training him.
There are a number of key points in this transition:
1. At the first meeting stage, you must adapt the controls to individual characteristics that won’t change throughout the whole training, otherwise it simply confuses the person; this also applies to the library for working with objects, dialog boxes, etc. For example, if you go through 4 out of 5 scenarios interacting with an object using a beam, and in 5 you ask to take it with your hand or pull it telepathically, this will confuse the person, waste extra time and affect decision making
2. Learning new management and interaction skills must be done consistently. It doesn’t matter if there are several or just one, give the information consistently. First, let the person walk around, look around, then along the script or linear plots, gradually add new interaction functions according to the degree of complexity, giving audio prompts, duplicating text, graphic animation and highlighting the necessary objects. If possible, it’s better to develop a training scene.
How to understand that the simulator is effective?
The chances of getting it right the first time are very low. Most likely, 3-4 months after implementation, you can collect basic information from the passing and eliminate the main defects, but the cycle of testing and data analysis may take more than one quarter.
The chances of getting it right the first time are very low. Most likely, 3-4 months after implementation, you can collect basic information from the passing and eliminate the main defects, but the cycle of testing and data analysis may take more than one quarter.
An effective simulator is a long and labor-intensive process.
One of the most long-term aspects is the implementation of a data collection and output system in the simulator.
For example, try to assemble origami and transfer these mechanics to a virtual reality environment. In real life, we’ll give a person instructions where it’ll be shown how to assemble the desired figure step by step. Now we put on glasses and try to assemble it in VR.
Here the testing stage is usually divided into two parts:
- A person performs this operation in a simulator, after which he does the same in real life. Of course, here you can add the number of attempts, time, etc. for analysis, we use ANOVA and SEM at these stages, dividing people into groups with training using video, text and audio lectures, at the same time at this stage we can understand whether VR is effective or another the method is more effective, but in general the process is very clear and is no different from testing standard simulators and standard educational methods.
2. It’s a completely different story when we often don’t have the opportunity to assemble origami in real life, or all the information that we visualize is pure theory. For example, a scientific experiment, or a simulation of an extreme situation, or equipment when working with which we can’t make a mistake.
The difficulty with this training, unlike the previous method, is that we can’t recreate the simulation we created in real life, which makes it difficult to effectively create facts to confirm the operation of the simulator.
And what if we don’t want to consolidate or test the theoretical knowledge we’ve already acquired, but teach something new?
Return to the same origami figure, show step by step in a virtual reality simulator how to assemble it, and ask to repeat it, assuming that the subject does not have the opportunity to assemble it in real life.
We have approximate information from the theoretical part that it’s assembled in such and such stages and in n time. Here, our primary task is to remove, if possible, all the effects of virtual reality that will distort the real action in life. Firstly — remove everything unnecessary. Our movement, working with objects and much more is one way or another different from a VR helmet and 2 controllers, so immersion and, if possible, the use of only those functions that virtual reality can reflect. For example, if in the learning process it’s necessary to walk a path and not teleport, then we give the opportunity to pass, if this doesn’t play any role, we transfer the person ourselves to a given point. If, when assembling a figure, it’s important for you to bend a sheet of paper with hands and practice fine motor skills — then we simulate hands, giving the opportunity to bend the sheet yourself and provide tactile feedback using vibration, or use leap motion for hand tracking or oculus quest, rather than bending it automatically when aiming the beam and pressing a button.
Only after this work, the next stage of testing begins—data collection and analysis.
Here we must understand what information is important to us first of all. For example, if you need to assemble a crane in n time — we analyze how long it takes a certain number of people to assemble it in a simulator, divide it into groups and analyze what errors were made in different versions of interactive solutions — if these are errors caused by the virtual environment, we eliminate them — or we give a condition not to count this is a mistake.
In a similar way, we deal with errors caused by the subject’s failure to master the material. Obtaining this information is the most valuable data in this entire system. This allows us to find a weak point in the entire training model and focus our attention on this problem. There can be thousands of variants of such errors, and to many of them the approach is usually individual; the solution can be either in the form of increased repetition or in the reworking of the teaching model.
Choosing the Right Equipment
Choosing the right equipment for a specific goal is extremely important, and the first step is to determine which glasses are best suited for your situation.
Choosing the right equipment for a specific goal is extremely important, and the first step is to determine which glasses are best suited for your situation.
The decision will depend on your budget, number of participants, time devoted and organization of the testing site.
If there are quite a lot of people, for example more than 100, who need to be tested in a short time, then it’s better to use autonomous shooters, dividing groups of 20–30 people. There are several models on the market like HTC Focus, Oculus Go, Quest, Pico or Google Daydream, with which you can visually show a video or simple animation, if you want, you can even add interactivity for selection or wrong answer, just try not to reload the text of your application, and make it large for readability, as in most cases autonomous models are not of excellent quality, often no larger than 2880 x 1600, which makes it difficult to read. Don’t forget to take care of the management system, such as script administration and concurrency systems. This solution is also well suited for introducing the environment and demonstrating advertising and marketing offers.
Perhaps in some cases, like retail, where you need to determine which product a person’s gaze is directed at first, or when placing an emergency stop button for equipment in the simulation, you need to place it in the place where a person will notice it first. Here it will be extremely useful to use equipment with eye tracking capabilities like htc pro eye or Varjo.
Think about what you will teach. Here you need to consider several important questions: what amount of detail is important to you, because if you are working with small objects and text, then a resolution of less than 2000 × 2000 by eye will cause difficulties, and if the key in your simulator is the use of your hands, such as working with a screwdriver, It is better to use technology to track them rather than controllers, this also applies to movement where wires can interfere with turning, although this problem can be solved by proper placement of equipment in the room, such as hanging wires and using wireless adapters.
Conclusion
It’s hard not to notice that equipment manufacturers are constantly improving and new players are entering the market offering their own unique solutions. The same can be said about content developers, whose communities are constantly developing unique and often open libraries for working in the virtual reality environment, which allows to reduce the time for implementing complex mechanics while focusing on creating new functions, which ultimately leads to constant improvement of the software.
The difficulty in creating an effective simulator lies in visualizing the user of a fictional or recreated virtual world, in eliminating the defects of the still imperfect, although constantly improving virtual reality technology, trying to convey interactivity and the effect of presence.
Developers need to show maximum creativity in order to constantly deceive the brain, which should be maximally involved in the virtual environment, forgetting about the program’s simulation of reality.
All this brings us to the fact that we have working technologies that are excellently developing at a rapid pace, but we have large gaps in the modernization of previous training systems to work together with a new and as yet little-studied environment that has already entered the market and even in this form shows good results. Not to mention putting training in a virtual environment into a separate category, because the rules for working with it are just beginning to be created and these methods are in many ways unique and have their own nuances that we have to formulate and confirm in practice.
