How Disorders of Consciousness Affect the Brain

Disorders of consciousness are conditions in which a person’s awareness of themselves or their surroundings is severely reduced or absent following brain injury. They range from deep coma, where the person shows no signs of wakefulness or awareness, to states in which someone appears awake but cannot reliably interact with the world. Misdiagnosis rates between these states run strikingly high, and emerging brain-imaging techniques are revealing that some patients who look completely unresponsive are, in fact, conscious inside a body that won’t cooperate.

The Spectrum of Impaired Consciousness

After a severe brain injury, consciousness doesn’t simply switch off and then back on again. It exists on a gradient, and clinicians recognize several distinct points along it. In coma, the person’s eyes remain closed, and there is no detectable sleep-wake cycle. In the vegetative state, now more commonly called unresponsive wakefulness syndrome (UWS), the eyes open and close in something resembling a sleep-wake pattern, but there is no reproducible sign that the person is aware of anything around them.¹ The person may grimace, groan, or move a limb reflexively, but these behaviors don’t appear purposeful.

The minimally conscious state (MCS) is where things get both more hopeful and more confusing. A person in MCS shows inconsistent but clearly identifiable signs of awareness: they might track a moving object with their eyes, reach for a cup, or occasionally follow a simple command like “squeeze my hand.”² Because these signs come and go, catching them during a clinical evaluation depends heavily on timing and the examiner’s skill. Emergence from MCS is defined by recovery of functional communication or the ability to use everyday objects correctly, though severe disability often persists.

A newer category, cognitive motor dissociation (CMD), describes patients whose bodies appear unresponsive but whose brains light up appropriately on neuroimaging when asked to perform mental tasks. Recent data suggest CMD occurs in up to a quarter of brain-injury patients who seem behaviorally unresponsive.³ These patients occupy what researchers sometimes call the “gray zone” of consciousness, and their existence has upended assumptions about who is and isn’t aware.

Why Misdiagnosis Is So Common

The gap between what a person with a brain injury can actually perceive and what they can show on a bedside exam is enormous. A systematic review estimated that roughly 37 to 43 percent of patients diagnosed as being in a vegetative state are actually in a minimally conscious state.⁴ That is not a small error margin. It means that for every three patients labeled as unaware, at least one may actually have some level of consciousness.

Part of the problem is that consciousness in these patients fluctuates dramatically. A study examining behavioral changes over a three-week period found that more than 80 percent of patients showed significant day-to-day fluctuation on standardized assessments. Among patients rated as conscious at baseline, a quarter were subsequently rated as unconscious at least once during the study window.⁵ If a clinician catches a patient during a low point, the assessment will underestimate their actual capacity.

Repeated standardized evaluations help. The Coma Recovery Scale-Revised (CRS-R), a structured bedside tool that examines auditory, visual, motor, and communication responses, is the most widely used instrument. But even with the CRS-R, a study comparing consensus clinical impressions against repeated formal assessments found that about 25 percent of patients who were actually in a minimally conscious state were misdiagnosed as being in an unresponsive wakefulness syndrome after a single evaluation. When assessments were repeated, the misdiagnosis rate actually rose to about 38 percent for those patients, suggesting that clinical consensus alone is not sufficient without the structured scale applied systematically multiple times.⁶

What Brain Imaging Reveals About Hidden Awareness

The landmark case that changed how researchers think about consciousness in brain-injured patients involved a woman who met every clinical criterion for vegetative state. When placed in an fMRI scanner and asked to imagine playing tennis, her brain’s supplementary motor area activated in a pattern indistinguishable from that of healthy, fully awake volunteers. When asked to imagine walking through the rooms of her house, a different set of brain regions lit up, again matching healthy controls. The fact that she could hear and understand the instructions, choose to cooperate, and switch between two distinct mental tasks on command was considered unambiguous evidence of conscious awareness, despite her outward appearance of total unresponsiveness.⁷

Since that initial case, researchers have explored different mental imagery tasks to see which works best at detecting covert command-following. A study comparing hand-squeezing imagery with tennis-playing imagery in patients with traumatic disorders of consciousness found that neither paradigm caught everyone. In patients who already followed commands on bedside exams, the hand-squeezing task detected covert responses more often, while in patients who did not follow bedside commands, the tennis task occasionally picked up command-following that the hand-squeezing task missed.⁸ The implication is that relying on a single imaging paradigm will miss some patients who are conscious, and combining approaches improves detection.

Beyond task-based imaging, resting-state brain scans and metabolic imaging are yielding diagnostic value. PET scanning of brain glucose metabolism has shown that global cortical metabolic rates average about 42 percent of normal in vegetative-state patients and about 55 percent of normal in minimally conscious patients. A threshold of roughly 45 percent of normal metabolism predicted the transition from vegetative to minimally conscious state, and the technique differentiated the two diagnoses with about 82 percent accuracy.⁹ A separate approach using glucose metabolic pattern analysis achieved similar differentiation, with hypometabolism concentrated in frontal and parietal cortex being a hallmark of the unresponsive state.¹⁰

Techniques using transcranial magnetic stimulation combined with EEG offer a more portable approach. A measure called the perturbational complexity index quantifies how complex the brain’s electrical response is when given a magnetic pulse. The idea is that a conscious brain produces a rich, diverse response, while an unconscious brain produces a simple, stereotyped one. This index can distinguish between unresponsive and minimally conscious patients and carries potential as a bedside diagnostic and prognostic tool.¹¹

What Goes Wrong in the Brain

Consciousness depends on a circuit that runs from the brainstem up through the thalamus and out to the cortex. Two systems are especially important. The ascending reticular activating system (ARAS), a network of brainstem nuclei, generates the basic state of wakefulness. Damage to the ARAS, particularly its upper portions connecting to the thalamus, strongly correlates with impaired consciousness. Studies of traumatic brain injury patients have found that the structural integrity of these tracts, measured by diffusion tensor imaging, shows a strong positive relationship with level of consciousness as measured by clinical scales.¹² A similar relationship has been documented in patients with intracerebral hemorrhage, where damage to the lower dorsal ARAS in the affected hemisphere was closely linked to impaired consciousness in the early period after the bleed.¹³

The central thalamus acts as a relay and amplifier, integrating arousal signals from the brainstem and distributing them across the cortex. Direct injury to the central thalamus or loss of input to it from widespread brain damage is associated with severe impairment of the brain’s ability to coordinate activity and maintain arousal.¹⁴

At the cortical level, the default mode network, a group of brain regions active during internal thought and self-referential processing, appears to be a key marker. Connectivity within this network drops in proportion to the severity of consciousness impairment, from healthy controls down through locked-in syndrome, minimally conscious state, vegetative state, and coma. The precuneus, a region near the back of the brain, shows particularly sharp differences: its connections are significantly stronger in minimally conscious patients compared to those who are fully unresponsive.¹⁵ More detailed structural analysis has confirmed that white-matter pathways connecting the posterior cingulate cortex and precuneus to the thalamus and to other default network regions are progressively more damaged as consciousness levels decline.¹⁶ The posterior cingulate cortex, in particular, acts as a central hub whose self-regulatory balance of excitation and inhibition appears to break down in the vegetative state.¹⁷

Who Recovers and What Shapes the Outlook

The cause of injury matters tremendously. Patients whose disorder of consciousness resulted from traumatic brain injury generally fare better than those whose injury was caused by oxygen deprivation (anoxic injury) or stroke. In comparative studies, traumatic brain injury patients tended to be younger, entered rehabilitation sooner, and achieved greater functional and cognitive recovery, while anoxic brain injury patients achieved the lowest levels of recovery overall.¹⁸ Even when matched for initial functional status, anoxic injury patients recovered more slowly, with physical recovery lagging behind cognitive recovery during inpatient rehabilitation.¹⁹

Late recovery does happen, but it’s uncommon and typically comes with significant limitations. In one study of patients with long-lasting vegetative state, about 20 percent of traumatic brain injury patients and a similar proportion of non-traumatic patients showed some late return of responsiveness. Of those, only about 12 percent further progressed to actual consciousness. Younger age and traumatic cause were associated with late recovery, but severe functional impairment was present in every patient who recovered, regardless of the cause.²⁰ A follow-up case series found that among patients who did recover responsiveness late, some regained the ability to follow simple commands, but none achieved independence in daily activities, and all showed lasting cognitive and behavioral difficulties.²¹

That said, a longer-term study tracking patients with disorders of consciousness from traumatic brain injury over a full decade found a more encouraging trend. More than two thirds eventually achieved independence in mobility and self-care, and about a quarter achieved independent cognitive function by the ten-year mark. Recovery was not front-loaded: the proportion achieving functional independence continued to increase between five and ten years post-injury, suggesting that ongoing monitoring and rehabilitation may provide benefits well beyond the early period that most clinical systems focus on.²²

Blood biomarkers are also emerging as prognostic tools, particularly after cardiac arrest. Neurofilament light chain (NFL), a protein released when nerve fibers are damaged, rises sharply in patients who go on to have poor outcomes. By 72 hours after cardiac arrest, NFL levels showed high prognostic accuracy, with concentrations above a certain threshold identifying patients with severe brain injury with excellent specificity.²³ The trajectory of NFL levels over the first few days, not just a single measurement, adds further information and may help clinicians distinguish severe brain injury from other causes of poor outcome.²⁴

Treatments Under Investigation

There is no drug that reliably restores consciousness, but a few interventions show promise. Amantadine, a dopamine-promoting medication originally developed for influenza and later used in Parkinson’s disease, is the best-studied pharmacological option. A landmark placebo-controlled trial published in the New England Journal of Medicine found that patients with severe traumatic brain injury who received amantadine during weeks 4 through 16 after injury recovered significantly faster than those on placebo over a four-week treatment period.²⁵ A recent meta-analysis, however, found that the overall pooled evidence across multiple studies did not show a clear long-term advantage for amantadine on functional recovery. When the analysis was limited to randomized controlled trials examining the first six weeks of treatment, a significant benefit did emerge, suggesting amantadine may accelerate early recovery without necessarily changing the endpoint.²⁶

Deep brain stimulation (DBS) targeting the central thalamus is one of the more dramatic experimental approaches. A meta-analysis of individual patient data found that cyclic DBS of the central thalamus acutely improved daytime awareness in patients with severe consciousness impairment and may increase the potential for ongoing recovery.²⁷ Researchers have also identified a candidate “sweet spot” in the lower thalamus, near the border with the midbrain, where stimulation appears most strongly associated with clinical improvement.²⁸ Supporting this approach, primate experiments using general anesthesia as a model for unconsciousness have shown that central thalamic DBS can restore both arousal and the complex patterns of brain activity considered signatures of consciousness, while stimulating a different thalamic site had no such effect.²⁹

Less invasive brain stimulation is also being explored. A meta-analysis of randomized clinical trials using transcranial direct current stimulation (tDCS), which delivers a weak electrical current through electrodes on the scalp, found a small but statistically meaningful improvement in CRS-R scores favoring tDCS over sham treatment. Patients who were male or already in a minimally conscious state saw greater benefit, and five or more stimulation sessions outperformed single sessions.³⁰

Multimodal sensory stimulation, which involves exposing patients to combinations of familiar music, voices, scents, and tactile input, has shown positive effects in patients who are already in a minimally conscious state. A systematic review found that negative outcomes from sensory stimulation programs were rare, and the approach appears to be low-risk.³¹ One controlled study observed higher scores on consciousness assessments during treatment phases compared to non-treatment phases, but only in the minimally conscious group, not in patients in a vegetative state.³²

The Ethical Weight of Covert Consciousness

The discovery that some patients who appear vegetative may actually be aware has created difficult ethical terrain. If a person can understand spoken language and respond with brain activity but cannot move, speak, or signal outwardly, the stakes of decisions made on their behalf become vastly more serious. A minority of behaviorally unresponsive patients have been shown to follow commands or even communicate through neuroimaging or electrophysiology, which raises hard questions about end-of-life decisions and quality-of-care standards.³³

The timing of these assessments introduces further complexity. In one reported case, covert consciousness was detected acutely in a patient who ultimately died after life-sustaining therapy was withdrawn, while another patient who showed no covert consciousness at the time of testing later recovered full communication and functional independence.³⁴ Neither result changed the clinical team’s decisions at the time, but in retrospect, each case challenges straightforward interpretations of what brain-imaging findings should and shouldn’t mean for prognosis.

Pain assessment is another fraught area. Patients who cannot communicate can still potentially experience pain, and distinguishing a reflexive grimace from genuine suffering is difficult. The Nociception Coma Scale-Revised (NCS-R) was developed specifically to evaluate pain responses in non-communicative brain-injured patients. The scale separates reflexive behaviors like groaning from higher-level responses like localizing the source of pain, and patients in a minimally conscious state consistently score higher than those in an unresponsive state. Scores on the NCS-R appear to correlate with cortical processing of nociception, suggesting it captures something real about the patient’s inner experience.³⁵

What Families Go Through

The burden on families of patients with prolonged disorders of consciousness is severe and persistent. A scoping review found that caregivers consistently scored high on standardized measures of depression and anxiety. Up to 60 percent of caregivers in one study met diagnostic criteria for prolonged grief disorder, a condition distinct from ordinary grief in which the sense of loss remains acute and disabling long after the initial event.³⁶ This grief is sometimes called “ambiguous loss” because the person is neither fully present nor fully gone, a situation that resists the closure that comes with a clear bereavement.

A longitudinal study found that while some aspects of caregiver burden do decrease over time, financial burden does not. Medical costs, lost income from leaving work to provide care, and the ongoing expenses of long-term residential or home care compound relentlessly.³⁷ A systematic review of the psychological literature identified loss and grief, changes in psychological wellbeing, perceived burden, and use of coping strategies as the four primary domains of caregiver impact, underscoring that these families need structured psychological support and not just medical updates.³⁸

Locked-In Syndrome and Why It Gets Confused

Locked-in syndrome is sometimes grouped with disorders of consciousness in public discussion, but it is fundamentally different. A person who is locked in is fully conscious. They can think, feel, and understand language. The problem is purely motor: damage to the ventral pons (a lower brainstem structure) destroys nearly all voluntary movement except, in the classic form, vertical eye movements and blinking. That narrow channel of control is how locked-in patients communicate, and some have written entire memoirs through eye-blink dictation.

The brain-imaging data confirm the distinction. Default mode network connectivity in locked-in patients is not significantly different from that of healthy controls, placing them at the opposite end of the spectrum from vegetative-state patients despite superficially similar presentations at the bedside.³⁹ The practical risk is that a locked-in patient, who has full awareness, could be mistaken for someone in a vegetative state if the examining clinician doesn’t specifically check for vertical eye movement or blinking to command. Given the misdiagnosis rates already discussed for vegetative and minimally conscious states, the danger of a locked-in patient being classified as unconscious is not hypothetical.

EEG-based classification systems can help here too. A framework classifying resting-state EEG into categories based on the dominant frequency peaks, ranging from delta-only patterns (reflecting complete disruption of the thalamus-to-cortex circuit) through normal alpha and beta patterns, has proven useful for distinguishing levels of thalamocortical function and correlating them with outcomes.⁴⁰ A locked-in patient would typically show a normal or near-normal EEG pattern, which flags them as someone whose cortex is working fine even if their body is not.