How can you up your chances of having a lucid dream?
By reminding yourself you want to just as you're falling asleep, either as a verbal statement or idea: "Tonight when I dream, I want to realize I'm dreaming." That's the single most important thing, other than simply getting enough sleep. For any sort of dream recall or influencing of dreams, or for lucidity, simply getting enough sleep is one of the most boring pieces of advice, but one of the most important. When you deprive yourself of sleep, you are getting a lower proportion of REM. We go into REM every 90 minutes through the night, but each REM period gets much longer and occupies a larger chunk of that 90-minute cycle each time. So if you're only sleeping the first part of a normal eight hours of sleep, you're getting very little of the REM sleep you could.
Beyond that, if you check on whether you're actually awake in a systematic way during the day, you'll eventually find yourself doing this in a dream, and that can make it likelier that you will have lucid dreams.
You can do this by identifying something that is consistently or usually different from your sleeping and waking experience. Lots of people find they can't read text in a dream, that if they see text it's almost always garbled or hieroglyphics or doesn't make sense or it's fuzzy. People who can read in a dream will still report that the text is not stable; if they look away and then back, it says something different or there's no longer any writing there. So trying to read something in a dream is a good test for lots of people. Others find that things like light switches and other knobs that are supposed to turn things on and off work normally in their real world and don't do what they expect them to in a dream.
If you work out one specific check and then ask yourself, does everything look logical, you'll find yourself doing that in a dream. Some of these techniques are successful in as many as 10 percent of people in the course of a week for a few studies.
What are less effective ways of controlling a dream?
People who decide that they want to alter their nightmares or solve a problem through lucid dreaming have carved out an infinitely more difficult path—not that it's impossible but there's a lot more hard work and a lot less chance of success that way.
When lucidity was getting press in the 1970s, people were thinking it's a great way to end nightmares and have problem-solving dreams. But it turns out that lucidity takes a lot more effort and happens more unreliably than other forms of dream control. The study where I had students select real-life problems within their ability to solve—with strong motivation, in one week half dreamed about the problem and one fourth dreamed an answer to their problem, and that's much higher than you'd get for lucidity techniques. In transforming-nightmare studies, that rate is higher and happens quicker than it does for lucidity. So approaching these goals by almost demanding that the dream do what really you can do much better awake is not the smartest approach.
What about controlling someone else's dream—is this possible?
Occasionally there are some ways that one might influence someone else's dream content ahead of time via waking suggestions or during sleep via sensory stimuli that are impinging on the dreams.
The auditory seem to things work best, such as water or a voice saying something. Very strong stimuli wake us up. You want it to get in some narrow threshold where it gets detected by the brain and processed but it doesn't wake you up, and then there's a shot at it getting incorporated into the dream.
In his research on lucid dreams, psychophysiologist Steve LaBerge tested a dream light that sleep subjects wore on their faces that detected REM and flashed a low-level, red light during that phase. He found that it often got incorporated into people's dreams—they saw a pulsing red glow. If you combine that with the suggestion that when you see the flashing red light you know you're dreaming, you can promote lucidity.
Magnetic input is being done in the waking state to improve depression and to halt psychomotor seizures. If you can influence mood awake, it would seem you could influence the mood of a dream. We will get more precise about what we know about different brain areas and targeting magnetic signals toward them.
Lastly, we can image the brain well enough awake or asleep to know things like: there's an unusual amount of motor activity; or this person is probably doing mathematical calculations right now; or this person is processing incoming language or speaking or writing or is very likely sad or very likely happy. And we will probably get better at that. We can already do more things with animals: If you've trained rats in a maze, during REM sleep they look like they're dreaming the maze—they show the same pattern of firing left-right turns. That's done by sinking needle electrodes into their brains, which we obviously don't do to humans. But we may get good enough at imaging nonintrusively from the outside to see a lot more about the content. That's not directly controlling a dream, but it's one of the things that you might want to know if you were trying to control dream content.