Can you define consciousness?
Nature News ran a feature this week about disputed definitions. One of these was a no-brainer…consciousness (ouch, I know). There are very few topics that can engage both scientists and the public alike as determining an appropriate definition for consciousness. And almost nobody can provide a satisfactory answer.
The current protocol used in hospitals to identify whether someone exists in a vegetative state involves a series of behavioral tests in which the subject is supposed to respond to stimuli and cues. However, this is not a reasonable assessment for those suffering from motor problems. Indeed, researchers using fMRI have suggested that “conscious activity” was present in the brain of a patient clinically-defined as being unaware. This demands that science re-define its standards for consciousness.
Or does it?
Once again, we run into the recurring problem of how to interpret human data derived from brain activity (typically collected using fMRI). At face value, changes in blood oxygen levels offer correlatory evidence in response to stimuli, at best. One could argue that when patients in a vegetative state hear speech, causing their language areas to ‘light up’, this is no more a demonstration of consciousness than when a simple organism responds to a chemical gradient. Normal sensitivity and transduction pathways are intact in the patient, leading to the expected increase of activity in areas integrating this passively-acquired sensory information.
However, naked skepticism breaks down when the patient in a vegetative state is confronted with more abstract and complex requests (which was done in both of the studies I reference above). Strikingly, in the Science Brevia article, a non-responsive patient was asked to visualize herself playing tennis. And her supplementary motor areas became active. She was asked to methodically imagine herself walking through her house. And her parahippocampal gyrus, the posterior parietal cortex, and the lateral premotor cortex all became significantly active. These activation patterns did not statistically differ from the patterns measured in “conscious” subjects. So now what?
Although these findings do give me pause, we have to remember that those remarkable responses are only observed in a small subset of patients. The vast majority of subjects tested in this manner do not exhibit activation patterns resembling much of anything. It is very expensive and time-consuming to conduct these extra scanning experiments on every non-responsive patient. And given the low percentage of those subjects even remotely calling their vegetative state diagnosis into question by these means, it doesn’t seem prudent to add such measures to the current standard array of tests.
But more important, we really don’t know what we are looking at when we examine these results. There are still controversies over the interpretation of fMRI data and under what circumstances it can be reliably used. So without having a definitive answer to the question of what do these fMRI activation patterns actually mean?, and only an answer to the question what could these fMRI activation patterns imply?, it doesn’t seem responsible to bring them into the mainstream just yet. Therapists could probably even make an argument for the dangers of giving family members false hopes of recovery. Or worse, causing more despair when one realizes that his/her loved one is completely unable to move or communicate with the outside world, yet lies there in a locked-down state, fully aware of the environment. Although I am sure that for the family and friends of a patient in a vegetative state, the situation is unbearably difficult, I would think that they might draw some comfort from the fact that the person is peacefully “sleeping”, free from any form of physical or mental anguish/suffering.
Broadening the discussion a bit, these fMRI results induce the same excited-skepticism in me as those studies identifying animals with mirror self-recognition. Like the fMRI experiments, these animal behavioral tests are currently the best we can do to shed any light on a very difficult question (in this case, do animals experience consciousness?). But, again, what do we really learn? Especially when only a subset of the subjects within a MSR-exhibiting species provides a positive result?
The fact that both the fMRI results described above or the identification of MSR are so intriguing, yet so unsatisfying, is a testament to the difficulty of even remotely defining what we mean by consciousness.