Dream Research
Micro-dreams, mind-wandering and the blurred edge of dreaming
Daydreaming and dreaming share a brain network, a loosening of control, and a tendency to go somewhere you did not intend. The boundary between them is considerably less firm than the words suggest.
Estimates from experience-sampling studies suggest people spend somewhere between a quarter and a half of waking life not attending to what they are doing. Matthew Killingsworth and Daniel Gilbert's large smartphone study put the figure at around 47 per cent of waking hours.
Some of that is deliberate reflection. A substantial proportion is spontaneous — thought drifting somewhere without being sent.
The shared network
The most substantial link between mind-wandering and dreaming is neural.
The default mode network — medial prefrontal cortex, posterior cingulate and precuneus, and lateral parietal regions — is active when a person is not engaged in an external task, and is associated with self-referential thought, autobiographical memory, imagining the future, and considering other people's mental states.
It is active during mind-wandering. It is also active during REM sleep, and imaging studies of dreaming find substantial overlap with the mind-wandering pattern. Meanwhile the executive control network, which regulates and constrains thought, is reduced in both states.
This has led several researchers — Kieran Fox and Kalina Christoff among them — to propose that dreaming and mind-wandering sit on a continuum of spontaneous thought, differing in degree of constraint rather than in kind.
Ordered by decreasing constraint: directed thought (working on a problem) → deliberate mind-wandering → spontaneous mind-wandering → hypnagogic imagery at sleep onset → non-REM mentation → REM dreaming. Each step involves less executive control, more sensory disengagement, and more vivid imagery. The transitions are gradual, not categorical.
What distinguishes dreaming
Two things, principally, and both concern how the experience is held rather than what it contains.
Sensory disengagement. Mind-wandering happens against a background of intact perception — you know where you are, and external events can capture attention. Dreaming involves the thalamus gating sensory input, so the dream substitutes for the environment rather than overlaying it.
Loss of metacognition. A person mind-wandering can usually recognise that they are, at least on being prompted. A dreamer cannot — the failure to notice absurdity is the defining cognitive feature of dreaming, and it is why lucid dreaming is remarkable rather than routine.
Interestingly, mind-wandering also involves a partial failure of this monitoring. People frequently do not notice they have drifted until they are caught — reading a page and realising they have taken in nothing is exactly a failure to notice a change of state.
Micro-sleeps
The most literal blurring of the boundary, and the most dangerous.
A micro-sleep is a brief intrusion of sleep into wakefulness — typically one to thirty seconds — during which the EEG shows sleep, the eyes may remain open, and behaviour continues. They are common in sleep-deprived people and increase dramatically with time awake and at the circadian nadir.
The person frequently has no awareness of having had one. Someone driving through a micro-sleep may travel a considerable distance with no perception or control, and reports of drivers who cannot account for the last few miles are describing exactly this.
Dream-like imagery occurs during micro-sleeps, which is why sleep-deprived people report hallucinations and intrusive imagery during monotonous tasks. It is not metaphorically dreaming; it is dreaming, briefly, while apparently awake.
Local sleep
A more recent and stranger finding.
Vladyslav Vyazovskiy and Giulio Tononi showed in rats that after prolonged wakefulness, small groups of cortical neurons enter sleep-like slow-wave activity while the animal is awake and behaving. The rest of the cortex remains awake. Errors on a reaching task were associated with these local off periods in relevant motor regions.
Similar findings have since been reported in humans using intracranial recordings, with local slow waves during wakefulness associated with lapses of attention and memory.
This complicates the whole categorisation. Sleep and wakefulness are not necessarily whole-brain states; parts of the cortex can be asleep while the person continues to function and to believe themselves fully awake.
Why the continuum view matters
Practically, three things follow.
The functions may overlap. If mind-wandering and dreaming share a network and a mode of thought, findings about one may inform the other. Mind-wandering is associated with creative incubation, autobiographical planning and future simulation — all of which have been proposed as functions of dreaming.
Sleep-onset is a genuine resource. The hypnagogic window sits at the point on the continuum where constraint has loosened but awareness remains, which is exactly why it produces the creative effects the Paris Brain Institute study measured.
Wakefulness is less unified than it feels. Between mind-wandering, micro-sleeps and local sleep, the impression of continuous alert consciousness through a working day is substantially a construction. The default mode network is running a great deal of the time, and what it produces there is not so different from what it produces at night.
