The Neural Correlates of Consciousness

Although philosophers, cognitive scientists, and clinical neuroscientists often use “consciousness” in different ways, the understanding of the concept that is most fundamental and most pertinent to clinical neuroethics is associated with the neurologists Fred Plum and Jerome Posner. ^{Footnote 1} They divide consciousness into two components: wakefulness or arousal and awareness of self and environment. Wakefulness is mediated by the brain stem ascending reticular activating system (ARAS) and its projections to the thalamus. Awareness is mediated by networks involving not only the ARAS and its projections to the thalamus but also projections from the thalamus to distributed networks in the cerebral cortex. Thalamocortical and cortico-cortical connections are necessary for awareness. Patients with locked-in syndrome (LIS) have intact activity in these same regions, but a lesion in the ventral pons of the brainstem disrupts connections between the central and peripheral nervous system and causes near-complete or complete paralysis. ^{Footnote 2} Those in a vegetative state (VS), or unresponsive wakeful state (UWS), have intact connections between the ARAS and thalamus but no integrated thalamo-cortical connections. The activity in their brains sustains sleep-wake cycles and other autonomic functions but not awareness. Patients in a minimally conscious state (MCS) have partly integrated thalamo-cortical connectivity corresponding to intermittent periods and levels of awareness. Functional connectivity between and among the brain regions that mediate wakefulness and awareness may be damaged or disrupted by infection, stroke, or traumatic or anoxic injury. The area of the brain affected by the injury and the extent of the injury can influence the diagnosis of the disorder as well as the prognosis. ^{Footnote 3} General anesthesia causes unconsciousness by suppressing the neural mechanisms mediating arousal and awareness. ^{Footnote 4} Narcotics and sedatives can impair awareness by disrupting the cortical networks mediating it. Through different neural mechanisms, all of these unintended and intended events and actions cause changes in the brain that partly or completely disrupt consciousness.

Neuroscientists investigate the neural correlates of consciousness, the brain mechanisms that underlie it. It is at this level that they can explain when there is wakefulness and awareness, as well as how these two states of consciousness can be generated, sustained, disrupted, or irretrievably lost. Three main theories have been proposed to account for these mechanisms. Neural synchronization theory claims that consciousness exists because of the synchronization of integrated neural activity as it changes across different brain regions. Consciousness comes about from the dynamic and fluctuating rhythms of neural activity. ^{Footnote 5} The global neuronal workspace theory says that consciousness arises from neural activity distributed across the entire brain. A key component of this theory is the prefrontal-parietal network and how it responds to stimuli inside and outside the brain. ^{Footnote 6} The information integration theory says that consciousness is generated and sustained by connected networks of neural information in thalamo-cortical networks and networks across cortical regions. As the degree of integrated information in these networks increases, consciousness emerges. As the degree of integrated information decreases, consciousness fades. ^{Footnote 7} The degree of disruption determines the extent to which awareness is diminished. The information integration theory may be the most intuitive of the three theories I have described, and I use it as the basis of my normative assessment of conditions involving varying states and degrees of consciousness. What all of these theories have in common is the idea that consciousness is a graded property. The neural correlates of consciousness are neither fully on nor fully off but maintain a resting potential prior to their inhibitory or excitatory action. Brain injury causing dysregulation in or between these two types of neural activity, as well as neuromodulation techniques or drugs to regulate them, can affect the phenomenal quality of awareness and the subject’s response to it. These factors can influence whether consciousness is something that a person values or disvalues.

Awareness is the component of consciousness with normative significance because awareness involves the subjective quality of experience and how a person is affected by it. Following philosopher Ned Block, awareness can be divided into phenomenal consciousness and access consciousness. He defines the first as “experience” and says that “the phenomenally conscious aspect of a state is what it is like to be in that state.” Access consciousness consists in information processing and its “availability for use in reasoning and rationally guiding speech and action.” ^{Footnote 8} Access consciousness involves not only the content of our experience but also how we respond to or use it. Specifically, it involves the cognitive capacity to report this content to others and possibly also to make decisions about actions that should or should not be done to us. In some cases, a patient may have the capacity to process information about the content of her awareness yet lack the capacity to report it because she is unable to communicate verbally or gesturally. She may have the cognitive capacity to process the information but lack the sensorimotor capacity to express it. This occurs in cases of complete locked-in syndrome, in which patients lack even voluntary eyelid movements, and in the minimally conscious state, in which cognitive or sensorimotor impairment may preclude communication. There may be enough integrated information in the brain for one to have and reflect on one’s experience, but not enough to clearly and reliably express one’s attitudes and wishes about this experience to others.

There are important ethical implications of the relation between phenomenal and access consciousness. ^{Footnote 9} A patient who is aware and cognitively able to process information about her experience yet unable to report it may be worse off than one with a lower level of awareness who lacks this cognitive ability. Unlike the second patient, the first could become frustrated and suffer from the inability to report her experience. Yet it is possible that a patient at this level could use this ability to her benefit through a device such as an electroencephalogram (EEG)–based brain-computer interface (BCI) that would compensate for her motor impairment. It could allow her to express her thoughts through a signal-processing algorithm in a computer linked to her brain through electrodes placed on the scalp. Deep brain stimulation to the region of the brain mediating motor functions associated with language expression might also allow one to communicate thoughts and wishes about medical care. These interventions could provide full access consciousness and enable the patient to control whether her phenomenal consciousness continued or discontinued as a beneficial or harmful experience for her. It will be instructive to explore the ethical implications of the relation between phenomenal and access consciousness by examining cases involving intact and impaired awareness.

Intraoperative Awareness

Suppose that a patient unexpectedly becomes aware during surgery despite being anesthetized. Due to the falling concentration of the anesthetic, he experiences pain and probably also anxiety and panic from awakening when he should be unconscious. He may also form and retain a traumatic memory of the experience. Unintended intraoperative awareness is the experience of sensory perceptions during surgery. It often also involves explicit recall of the experience. ^{Footnote 10} Many patients are unable to report being aware during surgery because, in addition to the anesthetic, they are given a neuromuscular blocking agent that causes paralysis. This is necessary to prevent patients from moving and interfering with the intubation and ventilator that breathes for them under general anesthesia. Although not all patients explicitly recall this experience, it may be encoded and stored as an implicit memory that could affect them outside the sphere of their awareness long after the surgery. In the hypothetical case at hand, a device such as the bispectral index (BIS) or end-tidal anesthetic-agent concentration (ETAC) monitor detects awareness by recording activation in cortical brain regions. A higher dose of the anesthetic is administered, and the patient quickly returns to unconsciousness for the remainder of the surgery and then gradually regains consciousness. Postoperatively, he does not explicitly recall being aware.

Was the patient harmed by the experience? If the monitoring devices indicated that he was aware for some period of time, then we could infer that he was harmed during that period, despite the lack of a report from the patient. Suppose that the monitors failed to detect awareness, and he experienced pain and panic for the duration of the surgery. Would he have experienced even greater harm? How could this be known if there is no evidence of awareness based on a subjective report from the patient or objective data from the monitoring devices?

If the patient became aware during surgery when he was being cut and cauterized and did not expect to be aware, then it seems that he was harmed by the pain, anxiety, and panic he felt. The harm would be greater if the monitors did not detect the brain activation indicating awareness and a higher dose of the anesthetic was not administered, because he would have perceived pain and suffered longer. Again, though, it would be difficult to know this without a report from the patient. He could lack access consciousness because the anesthetic could impair the cognitive capacity for the information processing necessary to report his experience as he was gradually emerging from unconsciousness. Or he might have had this capacity when he became fully aware but was unable to exercise it because of an inability to move or speak. Lack of access consciousness would not mean that he was not phenomenally conscious and not experiencing the effects of intraoperative awareness, even though there was no objective evidence for this. More importantly, lack of access consciousness would cause additional pain and suffering and additional harm because it would prolong his awareness. Access consciousness in the form of reporting awareness would limit the harm to the period between the time when he first became aware and the administration of the additional anesthetic sending him back into unconsciousness.

Anesthesia monitoring devices measure not the subjective quality of consciousness but only the effects of anesthesia on the neural correlates of consciousness. Because there is no direct causal connection between the neural activity recorded by these devices and a patient’s experience, these devices cannot capture what it is like to be aware. It is also possible for a patient to be aware without this awareness eliciting corresponding information on the monitors, given that the transition from unconsciousness to consciousness is gradual. Some patients postoperatively deny being aware during surgery despite following commands and carrying on a conversation with the surgeon and anesthesiologist. This is due to the amnestic effects of anesthesia, which shows that retrospective oblivion is no proof of unconsciousness. There is no objective way to directly detect awareness and to know whether or at what level a patient is aware, much less to know what it is like for her to be aware. Consciousness can only be indirectly inferred from observation, neuroimaging, or electrophysiology, and these are fallible indices of neural and mental processes. Lack of objective signs of awareness does not prove that a patient is unconscious. Yet without such evidence it would be difficult to assume that a patient had some level of awareness. The neurological criteria and the patient’s verbal report that would be necessary to make a claim about awareness may not be available.

One way of allowing and confirming a nonverbal and nongestural report would be to set up an EEG-based BCI linked to the patient’s skull that would allow bidirectional feedback between the patient’s brain and the computer. ^{Footnote 11} The patient could use neural signals to translate information about his experience from brain activity mediating pain perception, anxiety, and panic through the computer to the surgical team. BCIs typically operate on visual feedback, which a completely paralyzed patient with closed eyes would lack. But auditory feedback could allow the patient to manipulate the interface and make it clear that he was aware, in pain, or in a distressed state. This would be a way of enabling the patient to exercise the capacity to report his experience, which the paralysis had precluded. Without this technique, only the patient would have privileged “access” to his own experience and would suffer in silence, unbeknownst to anyone but himself. BCIs and other neural prosthetics might also allow patients with neurological disorders to reliably communicate their experience of living with such disorders and their wishes regarding life-sustaining treatment.

Disorders of Consciousness and Locked-In Syndrome

Different types of severe brain injury can result in a coma, in which there is a complete loss of wakefulness and awareness and complete unresponsiveness. Some comatose patients regain complete consciousness, usually within two to four weeks after injury. Later recovery is also possible. Others never regain consciousness and eventually are declared brain-dead. Still others progress from a coma to a vegetative state. According to neurological criteria, a persistent vegetative, or unresponsive wakeful, state becomes permanent 3 months after an anoxic brain injury or 12 months after a traumatic injury. This classification means that the patient effectively has no chance of becoming aware. It is supported by studies showing that there have been no instances of these patients improving to awareness during a five-year observation period. ^{Footnote 12} All of the patients either remained vegetative or died. Other patients with these injuries progress from a coma to a minimally conscious state. Until imaging studies by Adrian Owen and Steven Laureys helped to achieve diagnostic clarity in distinguishing disorders of consciousness, many patients who were covertly conscious were misdiagnosed as vegetative. ^{Footnote 13}

One ethical question that has emerged from diagnostic clarity in distinguishing the MCS from the VS (or UWS) is whether it is preferable to be in the first state than in the second. ^{Footnote 14} This question may strike some as incoherent because an unresponsive wakeful patient does not have the conscious capacity to compare what it is like for her to be vegetative with her being minimally conscious. But it is coherent to objectively compare the lives of two distinct patients with these two conditions. Many would answer the question by stating that it is clearly preferable to be minimally conscious than vegetative. But just because one patient is aware and the other is not does not mean that the first is better off than the second. Awareness as such is value-neutral. What matters is what we are aware of, how it affects us, and what we can or cannot do in light of this content.

Unlike vegetative patients, those who are minimally conscious may have an active neural “pain matrix” and may be harmed by noxious stimuli. ^{Footnote 15} These patients may also suffer if they have the capacity for an emotional response to pain perception and the anticipation of continued pain. These aspects of the capacity to experience what it is like to be minimally conscious, and the lack of this experience in the vegetative state, suggest that those in the first state may be worse off than those in the second. Any pain experienced by a patient in an MCS may be alleviated by analgesia. But the felt quality of pain and the extent to which it can be alleviated cannot be captured by functional neuroimaging. It requires some form of a report from the patient about her first-person experience, which is not possible if she cannot communicate. On the other hand, because they have more integrated activity between the thalamus and cortical regions than vegetative patients, minimally conscious patients have a higher probability of recovering some degree of cognitive and motor function. The critical question is whether the probability will be actualized to the point of achieving functional independence. The probability and extent of recovery depend on the type and severity of brain injury. Those who have progressed to an MCS from an anoxic injury have poorer prospects of recovery because this type of injury typically results in diffuse axonal damage in the white matter of the brain. This damage makes any form of neuroplasticity or neural regeneration from any intervention highly unlikely. Patients in an MCS from a traumatic brain injury (TBI) have better prospects of recovery because the damage to neural networks is not as severe, and intact growth factors in these regions may respond favorably to different interventions. It would seem, then, that it is better to be in an MCS than a VS, at least when the MCS results from TBI. Again, though, this hinges on whether and to what degree these patients will recover and how they respond to being in a neurologically compromised state.

Despite advances in diagnostic clarity, there is still prognostic uncertainty about the MCS. Emergence from an MCS is defined in terms of reliable and consistent communication. Although it is significant, the ability to communicate is only one aspect of recovery from brain injury. There is considerable variation in the rate and extent of recovery, even among those with TBIs. Some of this may occur later in the sequence of neurological events from the time of injury. Joseph Fins points out that even the ability to communicate “is unpredictable and occurs in a rather open-ended fashion two to five years after injury.” ^{Footnote 16} One patient who progressed to an MCS after sustaining both a traumatic and anoxic brain injury experienced significant improvement in cognitive and motor functions after taking the drug zolpidem. ^{Footnote 17} Another patient in an FDA-approved clinical trial of the first use of deep brain stimulation for the MCS showed a significant degree of functional recovery. The patient’s condition improved as a result of electrical stimulation of the central thalamus to modulate a dysfunctional mesocircuit consisting of the thalamus, basal ganglia, and frontal cortex. ^{Footnote 18} Among other functions, the patient regained considerable verbal ability that involved much more than the simple “Yes” and “No” responses to questions that some minimally conscious patients have been able to elicit. ^{Footnote 19} More clinical trials testing psychotropic drugs and brain stimulation on patients with more integrated neural connectivity might lead to therapies restoring a greater degree of cognitive and motor function and greatly improving their quality of life.

But the prognosis for most of these patients remains poor. The results of the zolpidem experiment and the DBS study have not been generally replicated. In a recent study using thalamic stimulation in minimally conscious and vegetative patients, the technique “did not induce persistent, clinically evident conscious behavior in the patients.” ^{Footnote 20} For most MCS patients, any recovery is usually modest, and those who do recover remain profoundly disabled. Theoretically, any experimental intervention should be offered within five years of the injury. This is consistent with the 2013 report of the UK Royal College of Physicians National Clinical Guidelines on Prolonged Disorders of Consciousness. They state that “the majority (>60–72%) of reported cases of patients who have emerged from the MCS have done so by 2 years after injury, with a further 30% emerging at 2-4 years. Cases of MCS patients emerging after more than 5 years from injury have rarely been reported.” ^{Footnote 21} In light of this data, for those who have been minimally conscious for more than five years with no signs of functional recovery, palliative care may be more reasonable and humane than allowing families to enter them in clinical trials or leave them to linger for years in long-term care facilities.

A patient with enough awareness to understand his condition alone should decide whether continued life would benefit him and whether artificial hydration and nutrition should continue. Yet without any communication from the patient, this may not be objectively known. Analgesia can control pain but does not always relieve suffering. The patient may suffer from the feeling of being completely dependent on others. She may also suffer from the inability to reliably express her attitudes and wishes to others. Indeed, Fins notes that this is the main source of frustration and suffering for many minimally conscious (and locked-in) patients. ^{Footnote 22} A critical component of this frustration is that, in many cases, recovery of cognitive functions outpaces recovery of motor functions, both of which are necessary for communication. Distress could increase when the gap between cognitive and motor functions increases. Intuitively, a patient at the higher end of the MCS spectrum with a higher level of awareness and cognitive function would be better off than one at the lower end. But without a corresponding improvement in motor function, the first could be more distressed and in this respect worse off than the second. This sheds further light on the question of whether it is better to be minimally conscious than vegetative because vegetative patients do not have the neurological capacity for this negative experience. If significant recovery from a minimally conscious state were highly probable, then any pain and suffering or any inability to have pleasurable experience during the five or more years following injury in this state could be outweighed by the benefits of recovery at a later time. But the data from studies on these patients do not generally support this view.

Because minimally conscious patients cannot express their wishes, families act as their substitute decisionmakers regarding medical care. They may also give proxy consent to allow these patients to be subjects in medical research combining functional neuroimaging and interventions aimed at unlocking thalamo-cortical connections and developing such interventions into therapies for them. Families of loved ones with disorders of consciousness generally act in their best interests. As much as they intend to do this and struggle in their role as surrogates, their decisions are not always consistent with these interests. Nor do their actions always promote or align with reasonable treatment goals. In some cases, the family of a vegetative patient with diffuse axonal injury and no chance of recovery may insist on continuing life-sustaining treatment when palliative care is more appropriate. Even with diagnostic clarity and approximate prognostic certainty, some families may misconstrue wakefulness as a sign of imminent recovery and have unreasonable expectations about treatment and outcomes for the patient. In cases of minimally conscious patients, some families may mistakenly infer from the fact that their loved one has progressed from a coma to a minimally conscious state that he will continue to progress to complete or near-complete recovery. They may assume that the patient’s awareness means that he will recover many cognitive and motor functions. Families have the right to sustain hope in recovery and to advocate for patients. But the mere presence of awareness does not imply meaningful recovery. Nor does awareness alone justify family requests to continue artificial hydration and nutrition indefinitely.

Most people do not consider severe brain injury and resulting disorders of consciousness when formulating advance directives about life-sustaining or end-of-life care. Even if they did, variability in attitudes about living with such a condition and uncertainty about whether one would or would not adjust to it show how limited advance directives are for people with these disorders and their proxy decisionmakers. A BCI or similar device that tapped into linguistic and semantic processing regions of her brain might enable a patient with a relatively high level of awareness to communicate her wishes about care through the computer. Minimally conscious patients may have these wishes, though they are unknown or ambiguous to observers because they cannot clearly express them. BCI-mediated communication could clarify these wishes and help to resolve conflicts between families and medical teams or courts in determining what is in the patient’s best interests.

In the 2011 case of Re M, for example, the English Court of Protection ruled that it would be unlawful to withdraw artificial hydration and nutrition from a woman who had been in an MCS for eight years. ^{Footnote 23} Her family’s request that life support be withdrawn, which the court rejected, was based on a statement by the patient before her injury that she would not want to remain alive in a completely dependent condition. The expert evidence indicated that the woman was in unremitting pain. But if the main issue was complete dependence on others, and the patient was aware of this, then she could suffer even if analgesia relieved her pain. She was considered to have a relatively high level of cognitive function at the higher end of the MCS spectrum. This suggested that she had the cognitive capacity to have interests. But she lacked the motor capacity to express them. Specifically, this might have included an interest in whether the artificial hydration and nutrition keeping her alive should have continued or been withdrawn. A device that allowed her to reliably communicate could have resolved the dispute between her family and the court by enabling an unambiguous statement of her wishes. By compensating for motor impairment and enabling these patients to communicate when they cannot do this on their own, techniques such as BCIs may help these patients to realize their interests and can ensure that others promote rather than interfere with them. They may allow patients to process enough information to report their experience. This exercise of their access consciousness in making a report may give them some control over continuing or ending their phenomenal consciousness. As neurologists Leigh Hochberg and Merit Cudkowicz point out, technology that may allow communication of patients’ wishes and possibly decisions about life-sustaining interventions “by proclamation rather than proxy will be revolutionary” in achieving this goal. ^{Footnote 24} Although they make this point with respect to locked-in patients, it also applies to minimally conscious patients with higher levels of awareness and cognitive function. This still leaves the question of whether high-functioning MCS patients could meet the formal requirements of informed consent and how this could affect the realization of their basic wishes.

Unlike the MCS, in which there is variability among patients in the degree of awareness and cognitive and motor functions, those with LIS have full awareness and intact cognitive functions. But they lack most if not all motor functions and have complete or near-complete paralysis. Locked-in syndrome is not a disorder of consciousness but a neurological disorder in which a person is conscious while being imprisoned in an unresponsive body. The LIS may be worse than the MCS because those who are locked in may suffer more from being fully aware of their physical limitations. Their level of awareness may exacerbate suffering by allowing them to draw comparisons between their present disabled life and their preinjury abled life. Some locked-in patients can communicate through voluntary eyelid movements. This is how Jean-Dominque Bauby, locked in from a brain stem stroke in 1995, authored an account of his experience in The Diving Bell and the Butterfly. Those who are completely locked in without the ability to move their eyelids may suffer the most, because they cannot interact in any way with their environment. As noted, BCIs typically use visual feedback to enable subjects to translate thoughts corresponding to brain signals in motor cortices through a computer. This is not available to completely locked-in patients with no visual capacity. But systems using auditory feedback might allow these patients to express their attitudes about their quality of life as well as continued life.

Bauby, who was not completely locked in, emphasized that his injury left his imagination and memory unscathed. It was largely because of this, combined with support from his rehabilitative team, family, and friends, that he was able to construct meaning from his severely physically limited life. Neuropsychologist Niels Birbaumer and colleagues have reported that a majority of locked-in patients have indicated through simple affirmative responses to questions that they have a good quality of life and want to continue living. ^{Footnote 25} At least some of this may be attributed to family and caregiver support. Patients’ attitudes about their quality of life could be clarified and confirmed by BCIs or other devices enabling them to communicate. Nevertheless, not all locked-in persons express the same positive attitude about life and may fall into a suicidal depression secondary to the neurological condition. Tony Nicklinson, who was locked in and paralyzed below the neck for seven years following a brain stem stroke, requested to the United Kingdom Court to be allowed to end his life with the help of a physician. The court rejected his request in August 2012, just six days before he died from pneumonia at age 58. Communicating through a computer, which he operated by eyelid movements, Nicklinson stated that he wanted to end his “dull, miserable, demeaning, undignified and intolerable life.” ^{Footnote 26} This assessment is in stark contrast to the value Bauby attributed to his experience. Still, Bauby was locked in for only two years before dying from pneumonia. One can only speculate on whether he would have sustained his positive outlook after seven or more years.

Even when they are not completely locked in and retain minimal voluntary motor functions, some patients would, if able, report that their lives are not worth living and accordingly should end. Others would say that they experience a good quality of life and want to continue living. These differences would obtain among minimally conscious patients as well. Some patients, like Jean-Dominque Bauby, adjust to these conditions and value continued conscious life. Others, like Tony Nicklinson, do not adjust and find no value in it. One’s attitude about being aware of and experiencing the effects of severe neurological disorders may depend on the extent of the injury, the rate of recovery, which cognitive and motor functions are intact, and professional and social support. But there is also the factor of the will to live, or loss thereof. Combined with these other factors, the strength of one’s will ultimately determines whether one would want to continue living. The content of this mental state cannot be measured by any objective quantitative or qualitative scale. It may be the most significant factor in explaining the psychological differences among minimally conscious and locked-in patients in their responses to their conditions.

Drifting In and Out of Consciousness

One of the challenges in the care of terminally ill patients is balancing pain control with oversedation. Opioid analgesics and sedatives can manage pain, but often at the cost of awareness becoming cloudy, impairing the patient’s thought processes and making him incoherent. From the patient’s perspective, perhaps the greatest challenge is balancing pain control with the ability to interact with healthcare providers, family, and friends. Ideally, treatment will maintain this balance and achieve both goals. In many cases, however, analgesics or sedatives control pain but decrease awareness. Or the drugs do not interfere with awareness but are not of a sufficient concentration to control pain and prevent suffering. What many patients fear is not so much pain but more so the dulling effect of the drugs on the mind. An unconscious patient would not suffer from being deprived of the opportunity to interact with others because she would not be aware of it. Still, she could be harmed by the deprivation if she had a general interest in interacting with others until the end of her life. Another psychological factor in weighing the benefit and harm of palliative sedation is that the patient may experience existential angst in knowing that, by gradually eliminating consciousness, the drugs will usher in her extinction. There are many anecdotes of people who refuse to take pain medication because of its effects on their consciousness. Freud refused to take anything stronger than aspirin until the very end of his painful ordeal with oral cancer. Concerned about how opiates would cloud his consciousness, he reportedly said: “I prefer to think in torment than not to be able to think clearly.” ^{Footnote 27}

Not everyone with a terminal illness who is nearing the end of life has the same attitude about consciousness. Some who have suffered from cancer or other diseases for a long time may welcome a release from the awareness that causes them to feel pain and to suffer from anticipation of continued pain. They or their families may request or agree to palliative (terminal) sedation that will end their suffering by rendering them permanently unconscious. There is no longer any meaning for them in continued conscious life. Unconsciousness will release them from the angst of experiencing disease.

A case vignette can illustrate how the apparently conflicting goals of controlling pain and maintaining awareness can be balanced. A patient in a hospital ward has end-stage lung cancer. The pain is controlled by morphine, though the increasing dosage is causing her to drift in and out of consciousness. When she is sufficiently conscious, she tells the attending physician that she is concerned that the morphine will make her incoherent and unable to interact with her family. The physician informs her that he will alternate the intravenous morphine with a saline solution, the amount of which she can control with a handheld infusion pump. The sense of control this arrangement provides the patient is one factor that can help to diminish her pain because it can influence her perception and anticipation of it. The physician explains that alternating the two substances may be the best way to relieve her pain and enable her to remain coherent enough to converse with her partner and children. The patient believes that alternating the morphine and saline will relieve her pain, even though the saline is biochemically inactive. Because he has informed her that he is giving her a saline rather than a morphine infusion, the physician is not deceiving the patient. He need not also tell her that the saline is an intravenous placebo lacking the biochemically active properties of morphine. Any pain relief as a placebo response will be influenced by the patient’s expectation, which is based on her memory of the past effects of morphine infusions. ^{Footnote 28} Even if the physician withholds some medical information about the details of the infusion, this has to be weighed against the fact that the placebo analgesia is a form of palliative care. More importantly, it allows the patient to remain conscious enough to interact with her family while tolerating the pain from the disease. These actions are necessary for her to engage in a valuable activity as she approaches the end of her life. The physician makes this possible by enabling the patient to retain both her phenomenal and access consciousness.