Why Do We Dream?

You spend roughly two hours dreaming every night.

That’s over 700 hours a year. Over a lifetime, it adds up to years of experience — vivid, emotional, strange — that you mostly can’t remember by morning. Your brain devotes enormous resources to something that, on the surface, seems to produce nothing.

There must be a reason. Scientists have been trying to find it for decades. Two theories have risen above the rest. Both are probably partly right.

Theory 1: Threat Simulation

In 2000, Finnish cognitive scientist Antti Revonsuo proposed that dreams exist to rehearse responses to danger.

His argument: across most of human evolutionary history, life was genuinely dangerous. Predators. Rivals. Unfamiliar terrain. The individuals who survived were those who responded fastest and most effectively to threats. And the best way to prepare for a dangerous situation is to have already practiced it.

Dreams, Revonsuo argued, are that practice. The brain runs simulations — you’re being chased, you’re falling, you’re in a confrontation — to rehearse fear responses, threat detection, and escape strategies. The simulation is imperfect (you often can’t run, the attacker changes form, the rules of physics don’t apply), but the emotional training still works.

This would explain why negative dreams are so much more common than positive ones. Roughly 70% of dream content involves some element of threat, failure, or conflict. Your brain isn’t pessimistic — it’s running drills for the scenarios that actually kill people.

It also explains why you wake up from nightmares with your heart already racing. The body responded to the simulation as if it were real. That’s the whole point.

Theory 2: Memory Consolidation

The other leading theory focuses on what sleep does to memory — and dreams are a window into that process.

During REM sleep (the phase where most dreaming occurs), the brain is doing something important: replaying and reorganizing the day’s experiences. New information gets linked to existing memories. Emotional weight gets recalibrated. Connections that don’t matter get pruned; connections that do matter get strengthened.

Dreams, under this theory, are the visible surface of that process. When you dream about a conversation from yesterday mixed with a memory from three years ago in a setting that doesn’t exist, you’re watching your brain sort filing.

This explains why people often dream about whatever they were working on intensely. Scientists studying a problem will dream about the problem. Musicians will hear music. Athletes will practice skills in their sleep. The brain is rehearsing — not for threats, but for competence.

It also explains why sleep deprivation is so cognitively devastating. You’re not just tired. You’re skipping the filing.

What We Don’t Know

Both theories explain some of dreaming. Neither explains all of it.

They don’t fully account for why dreams are so strange — why your high school is also your current office, why your grandmother is suddenly a stranger, why the rules shift mid-scene. If the brain is just running training simulations, why are they so incoherent?

One possibility: coherence costs energy. A perfectly realistic dream simulation would require enormous processing power. The brain may be taking shortcuts — recycling old settings, merging characters, skipping causal logic — because the emotional and mnemonic content matters more than narrative accuracy.

Another possibility: the strangeness is the point. Some researchers think dreams deliberately violate expectations to force new associations — to connect things the conscious mind would never link. That the surrealism is a feature, not a bug.

The Honest Answer

We don’t fully know why we dream. We know it happens, that it’s associated with REM sleep, that it involves emotional processing and memory consolidation, that disrupting it has measurable costs. We know the content is shaped by waking experience and emotional preoccupations.

But the why — the evolutionary reason a brain would devote so much nightly processing to this — is still genuinely debated.

Which means that every night, you’re participating in something science hasn’t fully explained yet.

That’s not a bad way to fall asleep.