Dream bizarreness: why the impossible seems normal until you wake

In a dream, a dead relative appears and the dreamer is not surprised. A house has a room that has never existed and this seems unremarkable. A person is simultaneously your brother and your teacher. You are late for an examination in a subject you never studied, and the response is anxiety about being late rather than confusion about the premise.

The bizarre content is interesting. The uncritical acceptance is more interesting, and it is the defining cognitive feature of dreaming.

What counts as bizarre

Researchers have developed coding schemes to quantify this, and the categories are useful.

Discontinuity — a scene, character or object changes abruptly, or the dreamer relocates without transition.

Incongruity — an element is out of place. A person in the wrong setting, an animal speaking, an object with the wrong properties.

Uncertainty — the dreamer is unclear who someone is, where they are, or what is happening, and accepts the ambiguity.

Condensation — a character or place is simultaneously two things. This is one of the most characteristic and hardest to convey afterwards.

Studies applying these schemes find bizarreness in the substantial majority of REM dream reports, and considerably less in non-REM mentation — which tends to be more thought-like, mundane and coherent.

Why this explains what it explains

Almost every characteristic feature of dream cognition follows from that arrangement.

Failure to notice absurdity — the region that would flag it is not working.

Poor working memory within the dream — dreamers frequently lose track of what happened moments earlier, which is why dream narratives drift and why intentions formed in a dream are rarely carried out.

Inability to remember waking life properly — dreamers rarely reason from what they know awake. A person dreaming of a dead parent does not consult the knowledge that the parent died.

Emotional intensity out of proportion to content — limbic activity high, regulatory control low.

Confabulated explanation — the narrative frequently appears to be constructed after the fact to account for imagery and emotion that arrived first. This is the same left-hemisphere interpreter effect documented in split-brain patients, who confidently generate explanations for actions whose real cause they have no access to.

The lucidity connection

This is what makes lucid dreaming neurologically interesting. Studies of lucid REM find increased activity in exactly the prefrontal regions that are normally deactivated, along with increased gamma-band power.

Lucidity is, on this account, the partial reactivation of the critical faculty within the dream state — which is why it permits the recognition that would otherwise never occur, and why lucid dreams feel qualitatively different.

It also explains why reality checks performed inside dreams so often fail. The check is a behaviour; evaluating its result requires the faculty that is offline. Dreamers report seeing six fingers on their hand and concluding everything is fine.

Is bizarreness functional?

Two positions.

The by-product view holds that bizarreness is simply what happens when a generative system runs without supervision. It has no function; it is the signature of a brain doing something else with its critical faculties switched off.

The functional view holds that loosened constraints serve a purpose — that novel and improbable associations are useful for creativity and for integrating memories in flexible ways, and that a system which only ever produced coherent recombinations would find fewer connections.

The evidence that REM sleep favours remote associations and that sleep aids insight extraction offers some support for the second. It falls well short of establishing it.

A qualification on the data

How bizarre dreams actually are is harder to measure than it looks.

Dream reports are reconstructions, and the reconstruction happens with the prefrontal cortex back online. Some of the apparent incoherence may be introduced by the difficulty of rendering the experience in language — and, conversely, some may be removed, as the waking mind tidies fragments into a story.

Studies comparing reports collected immediately on laboratory awakening with home diary reports find differences in coded bizarreness, which suggests the method is affecting the measurement.

The uncritical acceptance, though, is not a reporting artefact. It is reported consistently, it is what distinguishes ordinary dreaming from lucid dreaming, and it has a clear neural correlate. Whatever else is uncertain, the brain that dreams is a brain that has stopped checking.