Does everyone dream every night? (The answer might surprise you)

"I don't dream." I hear this constantly. Friends say it, people on Reddit say it, someone brings it up every time dream journaling comes up in conversation. Surveys back this up: up to 6.5% of people report that they never dream, and roughly 1 in 250 people say they can't remember a single dream from their entire lives.

I went looking for what sleep research actually says about this, and the short version is: they're almost certainly wrong. The longer version is more interesting. Once you look at what's going on, the question shifts from "do I dream?" to "why don't I remember?"

You dream. Every single night.

In 1957, William Dement and Nathaniel Kleitman ran one of the first rigorous studies on dreaming. They woke participants during different stages of sleep and asked what they'd been experiencing. 80% of people woken during REM sleep reported vivid dreams. Only 7% reported dreams when woken during non-REM sleep.

That study anchored decades of research to a simple idea: REM sleep equals dreaming, and everyone goes through REM cycles. But later work complicated things in a way I find pretty interesting. David Foulkes found that more than 50% of non-REM awakenings also produce dream reports when you change the question from "did you dream?" to "did you experience any mental content?" Sleep stage N1, the lightest phase, yielded dream reports 80-90% of the time.

Most people over the age of 10 dream four to six times per night. Each REM period gets longer as the night goes on. The first might last a few minutes, but the last one before you wake up can run half an hour. Add in non-REM dreaming, and that's a lot of time spent in some kind of dream state with nothing to show for it by morning.

The question was never whether you dream. It's whether your brain bothers to save any of it.

95% of it disappears before you get out of bed

Approximately 95% of dreams are forgotten by the time you stand up. Not gradually over the morning. Within minutes of waking.

The reason comes down to brain chemistry. During REM sleep, your prefrontal cortex, the part responsible for logical thinking and memory encoding, is running at reduced capacity. Meanwhile the amygdala and limbic system are firing hard, which is why dreams feel so emotionally intense. Your brain is generating vivid experiences while the filing system is mostly offline.

For a dream to stick in memory, something needs to interrupt the process. A brief awakening, even one so short you don't register it, gives the brain a window to encode what just happened. Without that interruption, the dream plays out and vanishes.

This is also why you remember more dreams on weekends. Longer sleep means more REM cycles and a more gradual wake-up, instead of an alarm yanking you out of deep sleep.

Your brain is physically different from the person who remembers every dream

In 2014, a team at the Lyon Neuroscience Research Center led by Jean-Baptiste Eichenlaub ran PET scans on 41 volunteers. Half were "high recallers" who remembered about five dreams per week. The other half were "low recallers," averaging one dream every two weeks.

The difference showed up in two brain regions. High recallers had more blood flow in the temporoparietal junction (TPJ) during REM sleep, non-REM sleep, and while awake. They also showed more activity in the medial prefrontal cortex during REM and waking hours.

The TPJ is involved in reorienting attention toward unexpected stimuli. High recallers are more reactive to sounds and disturbances during sleep. They wake up more often during the night, usually for just seconds at a time. Those micro-awakenings are what allow dream memories to get encoded.

So the person who says "I dream all the time" isn't necessarily dreaming more. They're waking up more, in tiny fragments, and their brain is better wired to catch what was happening right before each interruption.

The "hot zone" that generates dreams isn't where you'd expect

This is the study that surprised me most. For decades, researchers assumed dreaming was basically a REM phenomenon. Turn on REM, dreams happen. In 2017, Francesca Siclari and her team upended that assumption.

Using high-density EEG, Siclari found that dreaming, in both REM and non-REM sleep, correlated with decreased low-frequency activity in a "posterior hot zone" spanning the occipital cortex, precuneus, and posterior cingulate. When this zone was active, subjects reported dreams. When it was quiet, they didn't. Didn't matter what sleep stage they were in.

The researchers got accurate enough to predict in real time whether someone was dreaming. Specific dream content even mapped to specific areas: faces activated the fusiform gyrus, spatial settings lit up the parahippocampal region, speech activated language areas.

The upshot is that dreams aren't just a REM thing. A specific cortical network produces them across sleep stages. You're probably dreaming even more than the older model suggested, and not remembering any of it.

There is one condition that actually stops dreaming

Nearly every person who says they don't dream is wrong. They're just not remembering. But there is a rare exception.

Charcot-Wilbrand syndrome is a neurological condition caused by focal damage to the occipitotemporal region, usually from a stroke. People with this condition stop dreaming entirely. Not just forgetting dreams, but producing no dream reports even when woken during REM sleep. Their sleep architecture stays normal. The REM cycles still happen. But the network that actually generates dreams has been damaged.

Only a handful of well-documented cases exist. Mark Solms used them to argue that dreaming and REM sleep are generated by different brain mechanisms, which Siclari's later work confirmed. The condition also tends to come with an inability to visualize while awake. Try to picture your kitchen, and nothing comes up.

For everyone else, the hardware works fine. You're just sleeping through the dreams.

What changes when you start writing them down

This is the part that actually matters if you're reading this and thinking "okay, but I still don't remember anything." Dream recall isn't fixed. It responds to attention.

A study by Michael Schredl found that keeping a dream diary for just two weeks increased recall frequency in people who rarely remembered dreams. Henry Reed's earlier 12-week study showed the same pattern: dream recall is a skill, and it gets better with practice.

The mechanism is probably the intention itself. Telling yourself before bed that you're going to remember shifts something in how your brain handles the transition from sleep to waking. You're more likely to lie still for a moment, scan for fragments, catch whatever's fading before you reach for your phone.

Many dream journalers report a noticeable uptick within the first week or two. Not from dreaming more, but from catching what was always there. Going from "I never dream" to writing down two or three entries a morning can feel like a switch flipped, but the dreams were happening before. You just started paying attention.

What this means if you think you don't dream

You almost certainly do. Four to six times a night, every night. The brain regions that generate dreams are active across every sleep stage, not just REM. The only known condition that actually stops dreaming is an extremely rare syndrome caused by specific brain lesions, with a handful of confirmed cases in the medical literature.

What varies is memory, and memory varies a lot. Some people have brains that wake briefly and often, catching dream fragments along the way. Others sleep straight through and wake with nothing. Neither says anything about the quality of your sleep or what's going on inside your head.

If you want to start remembering: keep something to write on next to your bed, tell yourself before sleep that you're going to remember, and don't move for a few seconds when you wake up. Give it two weeks and see what happens.