Research explainer · Dreaming
Why Do We Dream? Answers from the Research
You wake from a dream that felt completely convincing. Your childhood home was somehow your workplace, an old friend appeared, and an ordinary conversation turned into something impossible. Why would a sleeping brain create that experience?
The honest answer is that researchers have several promising explanations, but no single established answer. Dreams may reflect processes involved in memory, emotion, and imagination. Showing that these processes happen during sleep, however, is different from proving that the experience of dreaming performs a necessary function. Here is what the research can tell us, and where the questions remain.
Dreaming happens beyond REM sleep
Rapid eye movement, or REM, sleep is closely associated with vivid dreams. But dreaming also occurs during non-REM sleep, including the transition into sleep. REM describes a physiological sleep stage; dreaming describes an experience. They overlap, but they are not interchangeable. [1]
Researchers study dreams by waking volunteers and asking what was happening in their minds just before the awakening. Brain recordings add another layer. In a 2017 study, reports of dreaming in both REM and non-REM sleep were associated with changes in electrical activity in a region toward the back of the brain. [2]
That finding helps identify brain activity accompanying dream experiences. It does not explain why dreams exist, establish their purpose, or let researchers read every detail of a sleeping person’s private story.
Dreams may reflect the processing of memories
Sleep supports learning and memory. One idea is that dreams give us a partial glimpse of the brain revisiting and reorganizing experiences during sleep. The dream may reveal that processing without necessarily being what makes it work.
In an overnight study published in 2019, people learned a virtual maze before bed. Clearly recognizable maze content in their dream reports was associated with better performance the next morning. This replicated an earlier finding from a nap study. [3]
The distinction matters: participants were not randomly assigned to dream about the maze or not. The results connect dream content with later memory performance, but they do not prove that dreaming caused the improvement.
A 2025 experiment adds another piece. Thirty-four participants recalled an emotionally negative memory from their distant past before a nap, or completed a comparison writing task. Independent judges found that the recalled memories appeared in subsequent dreams, usually as fragments rather than exact replays. Participants’ own ratings did not show the same content effect, although their dreams were more negative after the memory task. [4]
This shows that a waking manipulation can influence dream content. Because the study did not measure subsequent changes in the original memory, it cannot establish that the dreams repaired, strengthened, or updated it.
Emotional processing is a promising theory
Dreams can carry strong feelings, so researchers have asked whether they help us respond to emotional experiences.
A study published in a 2020 issue of Human Brain Mapping found that people reporting more fear in their dreams showed lower emotional and brain responses to threatening images while awake. This is consistent with an emotion-regulation theory, but the relationship was observational. It does not mean frightening dreams are always helpful. [5]
In a 2024 study, participants viewed emotional pictures before and after a night of sleep. Those who recalled dreams showed a pattern of retaining negative pictures at the expense of neutral ones, alongside reduced emotional reactivity. The authors interpreted the findings as evidence for a role in emotional memory processing. [6]
There is an important caveat: remembering a dream is not the same as having one. People who reported no dream may still have dreamed and forgotten it. Comparing dream recallers with non-recallers therefore cannot, by itself, isolate the effects of dreaming. These findings are intriguing, not proof that dreams provide overnight therapy.
Dreams may help generate new ideas
Another possibility is that dreaming encourages unusual associations. A 2023 controlled experiment used spoken prompts during light sleep to encourage dreams about a particular theme. Participants in the sleep-and-incubation condition later performed better on creativity tasks related to that theme than comparison groups. [7]
This moves beyond simply noticing that creative people remember dreams. Still, the result concerns a specific laboratory procedure and theme-related tasks. It does not establish that every dream solves problems, that the benefit depends only on conscious dream experience, or that interrupting your own sleep will make you more creative.
Forgetting dreams is part of the research puzzle
Some mornings bring a detailed story; others leave only the feeling that something happened. A 2025 study followed 204 adults using morning reports and sleep measurements. Dream recall was associated with factors including attitudes toward dreams, mind-wandering tendencies, and sleep patterns. Remembering the content also differed from merely knowing that a dream had occurred. [8]
This is one reason scientists must interpret dream reports carefully. A blank morning report cannot reliably tell us that the sleeping mind was blank. Nor do these studies establish dream recall as a score for healthy sleep.
What to take from the research
Dreams can be interesting material for reflection. You might notice familiar worries, relationships, or memories, without assuming that a particular image has a fixed meaning. The studies discussed here do not validate a universal dream dictionary or a way to predict future events.
The most useful conclusion is modest: dreaming is a real, measurable aspect of sleeping brain activity, closely connected to waking experience. Memory, emotion, and creativity are productive research directions. Whether dreaming itself is necessary for these benefits, and whether different dreams serve different functions, remains unresolved.
Key takeaways
- Dreaming occurs in REM and non-REM sleep.
- Memory and emotional-processing findings are promising, but association does not prove a dream’s function.
- Forgetting a dream does not establish that you did not dream.
- There is no need to decode every dream to appreciate what research is revealing.