Most people who try VR for the first time are unaware of technical flaws that would be obvious to some experienced engineers and developers. If the VR experience is functioning as it should, then the user should be overwhelmed by dominant visual stimuli and feel as if he is inhabiting the virtual world. Minor flaws may be subtle or unnoticeable as attention is focused mainly on the targeted experience (as considered in the definition of VR from Section 1.1). Some parts might not be functioning as designed or some perceptual issues might have been neglected. This might result in an experience as that not as good as it could have been after performing some simple adjustments. Even worse, the flaws might cause the user to become fatigued or sick. At the end, such users are usually not consciously aware of what went wrong. They might blame anything, such as particular visual stimuli, a particular experience, the headset hardware, or even the whole concept of VR.
This problem can be mitigated by training specific users and developers to notice common types of flaws. By developing a program of perceptual training, a user could be requested to look for a particular artifact or shortcoming, or to repeatedly practice performing some task. Throughout this book, we have seen the importance of adaptation in human perceptual processes. For example, if a constant stimulus is presented over a long period of time, then its perceived intensity diminishes.
Through repeated and guided exposure to a particular VR system and experience, users can adapt their perceptual systems. This is a form of perceptual learning, which is a branch of perceptual psychology that studies long-lasting changes to the perceptual systems of an organism in response to its environment. As VR becomes a new environment for the organism, the opportunities and limits of perceptual learning remain largely unexplored. Through active training, the way in which users adapt can be controlled so that their perceptual abilities and discrimination power increases. This in turn can be used train evaluators who provide frequent feedback in the development process. An alternative is to develop an automated system that can detect flaws without human intervention. It is likely that a combination of both human and automatic evaluation will be important in the years to come.