Much empirical research is devoted to the “hard problem,” as the philosopher David Chalmers has put it, of why human information processing is accompanied by the subjective experience we call consciousness. Solving this problem has real clinical consequences for some patients (see “The Mystery Behind Anesthesia”).

Researchers typically distinguish between the contents of consciousness and its levels. The contents of consciousness are our subjective experience, such as the taste of coffee. It is sometimes said that an experience ranks among the contents of consciousness if there is “something it is like” to have it. If there is something it is like, say, for a bat to have a sonar sense, then that sense is part of bats’ consciousness.

Levels of consciousness, on the other hand, have to do with outward signs of a person’s total or background state of awareness. Our understanding of these levels directly affects patient care. Today, we recognize three distinct “stages” of degraded consciousness based on a person’s physical reactions. A person in a coma cannot be aroused and is considered unconscious. A person in a vegetative state is also unconscious but has signs of a normal sleep-wake cycle and may even appear to waken. Finally, a person is said to be in a minimally conscious state if an outside observer can see intermittent signs that he or she has some understanding of self or environment.

Yet these outer signs don’t capture the core phenomenon of consciousness—subjective experience. To properly care for patients, we arguably need a new classification system more closely related to the contents of consciousness. Recent research is suggesting how we might create one. In one study, Adrian Owen and coworkers in the U.K. asked a patient in a vegetative state to think of certain mental images (“Imagine playing tennis” or “Imagine visiting the rooms in your house”). The resultant brain activation was no different from that observed in healthy control subjects, suggesting that some people in a vegetative state are more conscious than we realize. This technique was used to communicate with four out of 23 vegetative patients who would otherwise have been considered unconscious; they thought of tennis for “yes” and being in their house for “no.”

Over the last few years, brain imaging studies like Owen’s, along with other projects, have improved our tools for studying consciousness. While still indirect, such methods are still better than trying to assess consciousness by measuring something altogether different, like outward signs of contact with the world. A new classification system might help us predict which patients will benefit most from rehabilitation. It will certainly affect how we make the ethical decisions that arise when caring for patients with degraded consciousness.

Morten Overgaard studies the nature of consciousness as leader of the Cognitive Neuroscience Research Unit in Denmark.