Consciousness Across Waking and Dreaming: A Unified Psychological and Cognitive Science Perspective

Part One

Part Two

The Connection Between Waking and Dreaming Consciousness

One of the most striking discoveries of late twentieth- and early twenty-first-century consciousness research is that the sharp folk-psychological boundary between waking and dreaming is, from the point of view of the underlying mechanisms, almost entirely illusory. The same core neural architecture that generates the stable, externally constrained, serially coherent, self-reflective consciousness we call “waking” is also responsible for the labile, bizarre, credulous, amnesic, and often deliriously vivid consciousness we experience in dreams. The difference is not the presence versus absence of consciousness itself, but a systematic reconfiguration of exactly those control parameters that cognitive science has spent four decades mapping: attentional gain, metacognitive monitoring, working-memory buffering, source monitoring, prefrontal inhibition of inappropriate associations, and the balance between top-down prior expectations and bottom-up sensory evidence. Dreaming is therefore not a breakdown of consciousness; it is a lawful, highly informative altered state of the very same consciousness studied in the waking experiments of Lecture 3.

At the phenomenological level, the majority of dreams (especially REM dreams) are experienced as fully conscious in the same sense that waking life is. The dreamer is not a passive spectator but an embodied agent who sees, hears, feels pain and pleasure, believes the events are really happening, and often experiences emotions more intensely than in waking life. Lucid dreams, in which the dreamer realises “this is a dream” while remaining asleep, provide the cleanest proof of continuity: the same subject who, moments earlier, was credulously fleeing a monster can suddenly achieve full metacognitive insight, voluntarily control the dream narrative, and even signal to the outside world with pre-agreed eye movements while posterior gamma power and prefrontal activation rise to waking levels. Large-scale surveys (Schredl & Erlacher, 2023) indicate that 55–60 % of people have had at least one lucid dream, and trained practitioners can achieve lucidity on most nights. The boundary between non-lucid and lucid dreaming is therefore graded, not binary, exactly as the boundary between inattentional blindness and full awareness is graded in waking paradigms.

Non-lucid REM dreams, however, exhibit systematic deviations that map one-to-one onto the parameters identified in waking consciousness research:

• Severely reduced working memory and attentional control
• Grossly impaired source monitoring and reality testing
• Hyper-emotionality with diminished regulation
• Acceptance of physically impossible events without surprise
• Profound anterograde and retrograde amnesia upon awakening

These are not random defects; they are predictable consequences of the neurochemical and connectivity profile of REM sleep.

High-density EEG, simultaneous EEG-fMRI, and serial-awakening studies (Siclarli et al., 2017; Wong et al., 2020; Siclari & Tononi, 2023) have established the following signature of REM dreaming:

• Posterior cortical high-frequency (beta/gamma) power is as high as, or higher than, waking levels, especially in the hot zone identified in Lecture 1.
• Prefrontal and dorsolateral prefrontal cortex show profoundly reduced low-frequency and gamma power, mirroring the deactivation seen in deep meditation, psychedelics, and hypnotic trance.
• The dorsolateral prefrontal cortex, anterior cingulate, and inferior parietal lobule (core nodes of the central executive and salience networks) are selectively deactivated.
• The default-mode network remains active but is decoupled from executive control regions, leading to unfettered narrative generation without critical evaluation.
• Acetylcholine is at peak levels, dopamine is elevated, noradrenaline and serotonin are at their lowest diurnal concentrations.

This neurochemical cocktail has direct functional consequences that align perfectly with the psychological models of Lecture 3:

1. Low noradrenaline → impaired prefrontal top-down control and source monitoring (Hobson’s AIM model, 2005; updated in Hobson et al., 2019).
2. High acetylcholine → enhanced bottom-up sensory gain and internal generation of perceptual content (consistent with predictive-coding accounts).
3. Dopamine surge → increased salience of internally generated representations and reduced reality testing (the same mechanism implicated in positive-symptom schizophrenia and psychedelic states).
4. Prefrontal deactivation → collapse of working memory, logical reasoning, and metacognition (exactly the regions whose disruption abolishes conscious access in masking and attentional-blink paradigms).

The result is a brain that retains the capacity for vivid, multimodal, high-complexity experience (high PCI, high posterior gamma) but has selectively disabled the very supervisory systems that, in waking, keep perception tethered to external reality and to a coherent self-narrative.

Global Workspace Theory (Dehaene, Changeux, Laureys) Dreaming is the perfect dissociation: the workspace is partially dismantled. Contents are generated and experienced vividly, yet they are not stably broadcast to the full frontoparietal network, and therefore cannot be held in working memory or reported reliably upon awakening. Lucid dreams restore prefrontal activation and re-establish global broadcasting, allowing deliberate control and reliable recall. The theory predicts, and experiments confirm, that dream contents that do make it into waking report were those that momentarily achieved higher workspace occupancy even in non-lucid sleep.

Recurrent Processing Theory (Lamme, Block) Posterior recurrent processing is not only preserved but enhanced in REM sleep. Local gamma-band synchrony in occipital, temporal, and parietal cortex is often higher than in waking rest. This explains why dream imagery can be hyper-vivid and hyper-real, supporting the claim that local recurrence in the hot zone is sufficient for phenomenal consciousness even when global broadcasting is impaired.

Higher-Order Thought Theories (Rosenthal, Brown, Lau) Standard HOT theories struggle with non-lucid dreaming: the dreamer experiences a rich world without any apparent higher-order thought that “I am now experiencing this.” Lucid dreaming rescues the theory by restoring explicit metacognition, but non-lucid dreaming forces either (a) the postulation of unconscious higher-order thoughts (Rosenthal’s move) or (b) a revision that allows first-order states to be conscious in the absence of metacognition when prefrontal areas are offline.

Illusionism (Dennett, Frankish) Dreaming is the illusionist’s dream come true. The dreamer is filled with the illusion of immediacy, presence, and reality, yet upon awakening discovers that the entire theatre was a functional fiction generated by the same brain that, in waking, generates the illusion of a stable external world. The difference is only in the strength and source of top-down priors.

Predictive-Coding / Active Inference (Friston, Hohwy, Clark, Seth) Dreaming is offline Bayesian model optimisation. With external sensory input clamped at a low level, the generative model is free to explore its own latent space, reducing prediction error against an internal dataset of memories and expectations. Bizarre dream content reflects the system’s attempt to explain away residual prediction error using the most probable internal causes available. Lucid dreaming occurs when the model infers, with sufficient precision, that the current data are best explained by the hypothesis “I am dreaming,” triggering a cascade of top-down changes that stabilise the dream body and environment.

Enactive / Sensorimotor Theory (O’Regan, Noë, Di Paolo) Dreaming appears to refute sensorimotor contingency theory because dream actions (flying, passing through walls) violate waking contingencies yet feel perfectly natural. Defenders reply that dream sensorimotor contingencies are internally consistent within the dream world: flying feels like flying-in-a-dream, not like violating physics. The theory survives, but only by granting that the relevant contingencies are those generated by the sleeping brain’s own efference copies, not by the external world.

Lucid dreaming is the single most important naturally occurring state for consciousness science because it is the only condition in which the subject can (a) enjoy full waking-like metacognition and volitional control while (b) remaining in a neurophysiological state (verified REM sleep) that normally produces credulous delusion. Real-time communication experiments (Dresler et al., 2021; Konkoly et al., 2024) have now achieved two-way interaction: experimenters flash coloured lights or speak words during verified REM, and lucid dreamers incorporate the stimuli into their dream and signal back with eye movements. This allows, for the first time, online interrogation of conscious content in a state where external sensory input is minimal and prefrontal activation is voluntarily modulated by the dreamer. Preliminary results show that prefrontal gamma and frontoparietal coherence rise dramatically at the moment of lucidity, exactly as GNW and HOT theories predict, while posterior hot-zone activity remains high.

Clinical and Philosophical Implications

• Schizophrenia and dreaming share the same neurochemical signature (high dopamine, low noradrenaline, prefrontal hypoactivation) and the same functional profile (delusion, source-monitoring failure, emotional amplification). Dreaming is a physiological model of psychosis (Hobson, 2004; Scarone et al., 2023).
• Nightmares and PTSD dreams reveal that the affective core of consciousness is preserved and even intensified when cognitive supervision is removed.
• The profound amnesia for dreams upon awakening demonstrates that phenomenal consciousness and episodic memory encoding are almost completely dissociable, echoing the dissociation seen in hypnosis and anesthesia.
• Philosophically, dreaming dissolves the privileged status of waking consciousness. If the same brain can generate a world that feels completely real yet is entirely confabulated, the realist intuition that waking perception is veridical by default loses force. Idealist, simulation, and illusionist positions gain indirect support.

Synthesis with Lecture 3 (Part-1)

The psychological and cognitive science of waking consciousness described in part three is not replaced by the study of dreaming; it is completed by it. Every parameter that varies continuously in waking experiments (attentional gain, metacognitive accuracy, working-memory load, source monitoring, prior strength, prefrontal engagement) varies even more dramatically, and more lawfully, between waking and sleeping states. Dreaming is therefore not an exotic exception but the most radical point on the same multidimensional state-space that includes inattentional blindness, subliminal priming, hypnotic absorption, psychedelic ego-dissolution, and meditative absorption. The entire manifold can be described by a small number of control parameters whose values determine whether the resulting consciousness is externally constrained and critically reflective (ordinary waking), internally generated and credulous (non-lucid dreaming), or internally generated yet critically reflective (lucid dreaming).

Waking and dreaming are not two kinds of consciousness. They are two stable attractor states of one and the same consciousness-generating system, differing only in the weighting of bottom-up evidence versus top-down expectation, and in the degree to which the supervisory, reality-testing, narrative-self system is online or offline. The study of dreaming thus confirms, deepens, and radically extends the core insight of modern cognitive approaches: consciousness is not a fixed property or a mysterious extra ingredient; it is a controllable, highly plastic process whose contents, structure, and very credibility are functions of the momentary configuration of the same neural and cognitive architecture that psychologists have been dissecting, with ever greater precision, for the past forty years.

Tanmoy Bhattacharyya