Judging by the article, it doesn't seem like the experiment supports the conclusion. The experiment demonstrates that applying the filters makes it more difficult to distinguish real objects from virtual objects, but it does not necessarily follow that this makes the experience more immersive than the unfiltered version. In general, a consistent experience is important to suspension of disbelief, but that is only one factor. Most people didn't have a problem "getting into" "Who Framed Roger Rabbit?" or "Space Jam". Believable interactions is another important factor - one that these visual filters will impair. Obviously, the effectiveness of this approach will vary from person to person.
In any case, I can't imagine a use case for this technique. Such an approach would make interaction with the environment (including walking) more dangerous and frustrating. Attempting to interact with the environment would likely result in increased stress as your mind fights to determine what is real and what is not. If you're forced to remain stationary and can't trust what you see, what's the point in augmented reality?