Developing the ability to recognize signs that you are dreaming—"dreamsigns"—is critical to having lucid dreams.
This skill takes time and practice to cultivate. For some, it's easier and more convenient to narrow the focus of their induction training on purposely evoking a specific type of dreamsign…otherwise known as a “Reality Test”. Also, it's sometimes difficult to fully convince yourself you are dreaming, even when you suspect it, so a strongly programmed behavior such as "testing" your state of consciousness can come in handy. Moreover, increasing lucid dreaming requires a practice of reflecting on your state of consciousness while awake. Since indicators that you're dreaming tend to be absent when you truly are awake, reality tests can help you break down and reflect on the the assumptions that you are indeed awake.
Most Reality Tests Don't Work Consistently
Lucid dream explorers report various sorts of reality checks to practice testing which state of consciousness they are in. Some reality tests are sensible in that the dreamer attempts to elicit an act that would be impossible in the waking state, for instance, flying or walking through a wall.
A major problem with these tests, however, is that it remains unclear how reliably these tests result in true positives or false negatives. As an example, it is just as conceivable to dream that you cannot walk through a wall, leading to the false conclusion that you are awake. Attempts to walk through walls in dreams can be quite realistic (and potentially even painful if you charge through headfirst), whereas other times it occurs with ease. What you expect to happen (and possibly other factors) is likely to play a role in the outcome and conclusion of commonly used state tests, but this is not yet scientifically clear.
If a reality test does not result in a high rate of true positives, then their utility is limited and likely deters you from reaching lucidity as much as you'd like. So it makes more sense to use a type of reality test associated with reliably unstable features. This will make your induction regimen more efficient and effective.
Re-Reading Text in Dreams is Reliably Unstable
Dreamed perceptions are known to have unstable qualities since they are cut off from sensory inputs of the external material world. Particularly, the nature of printed text in dreams has been noted to be unstable in classical descriptions by several early 20th century authors (See LaBerge, Steiner, & Giguère, 1996). Recognizing this, Stephen LaBerge hypothesized that rereading text as a reality test would result most reliably in some sort of change to text upon re-reading (LaBerge & Rheingold, 1990). This was anecdotally supported by other lucid dreamers as well.
Changes upon re-reading to the text could be an entirely different word or phrase, a variation in a single letter, changes to the font or size of the text, the text could disappear altogether, and various other forms of perceptual instabilities.
In the mid 1990s LaBerge tested the instability of printed text in dreams when re-read with a group of oneironauts (LaBerge, Steiner, & Giguère, 1996). The text changed 75% of the time upon the first re-reading, and 95% of the time upon the second act of re-reading. The high rate of reliable changes to written text make the act of re-reading a promising candidate for being one of the more reliable state tests there is.
The only other state test that is likely to compare to the reliability of the re-reading state test would be the "Reality Test Button" of the Dreamlight (See http://lucidity.com/DreamLight_Validity.pdf), a biofeedback induction device that once existed in the 90's. Given no commercial knockoffs of LaBerge's original lucid dreaming mask have ever provided such a brilliant feature to reliably test your state of consciousness, it's still safe to say that the re-reading state test is most likely the best reality test currently available.
HOW TO DO A RE-READING STATE TEST:
How do I know I am not dreaming right now?
LOOK FOR DREAMSIGNS: Notice if any dreamsigns are present and consider whether you could be dreaming. Before concluding you are truly awake, conduct a Re-Reading State Test to confirm you are awake.
REREAD ONCE: Read some small printed text, look away at something else, and re-read the text to determine if it changed.
REREAD TWICE: If the text does not change the first time you re-read it, look away again, and this time imagine that the text will be different when you re-read it. Expect it to change. Re-read the text once more and determine if it changed.
REHEARSE: Since you concluded you are not dreaming, visualize what it’d be like to be dreaming anyway. Imagine you recognized a dreamsign, remembered you are dreaming, and how you would have enjoyed the lucid dream.
REMIND: Remind yourself to recognize a dreamsign the next time you are dreaming.
WHEN TO CONDUCT A RE-READING STATE TEST:
Conduct a re-reading state test during the daytime several times per day. For example, do one when you encounter something dream-like in your waking environment.
Though the re-reading state test, when properly practiced , has good utility, it's not clear how reliably it results in lucid dreaming, how it compares to other reality tests, how much it accounts for lucidity onset when using an integrated induction protocol, and how it compares to other induction techniques.
A recent publication by an Australian research group (Aspy et al., 2017) tested a different sort of reality test in which dreaming subjects attempt to breathe through their closed mouth, which was NOT successful in inducing lucid dreaming.
Until the rereading state test is researched more carefully, we won’t know its exact effects for certain, but experts like LaBerge seem willing to bet that the rereading state test is more effective that other reality tests attempting to perform the impossible. He discusses that this may have to do with the complexity involved with the process of constructing and reconstructing small printed text without any external visual input. He places this in the context of evolution since the written word has not been present in human history for very long.
What have your experiences been like with reality testing, or re-reading text? Get in touch and let me know your thoughts.