Neuroethics 2007


January, 2007

The field of neuroethics took on a more explicit form in 2006 with the establishment of the Neuroethics Society. Founded by eminent scientists, lawyers, and ethicists, the society hosts a Web site, www.neuroethicssociety.org, and two “partner” publications, the American Journal of Bioethics and the Journal of Cognitive Neuroscience.

The year also saw significant advances (accompanied, of course, by much debate) in four main areas of neuroethics: intervening in emotional and behavioral disorders, brain privacy, the impact of emerging technologies, and subtle changes in our understanding of nonconscious brain states such as the persistent vegetative state.

Placebos in Clinical Trials

An area of ethical concern is the use of placebos in clinical trials. A debate was recently spurred by a study by Sumant Khanna in the British Journal of Psychiatry in which some 12 dozen patients with mania received a placebo instead of treatment with the conventional antipsychotic drug risperidone.1 Some doctors have questioned the validity of the informed consent given by the study participants, says Ganapati Mudur, writing in the British Medical Journal.2 These studies raise the issue of the ability of those with mood disorders to give informed consent.

Brain Privacy

The push for ever-more-sophisticated brain imaging techniques continues to challenge old notions about the mind, such as the inviolability of an individual’s unspoken thoughts. Entrepreneurial scientists have developed lie-detecting devices based on functional magnetic resonance imaging, or fMRI, that they claim will offer greater accuracy than the traditional polygraph, which measures the responses of the sympathetic nervous system.

In an fMRI study designed to simulate the investigation of a shooting inside a hospital, Feroze Mohamed and colleagues identified eight brain areas that showed significantly more activity during the act of deception than in a neutral situation, and two areas that showed significantly more activity during truth-telling than in a neutral situation. They published their work in Radiology.3

To date, most neuroscientists are reserving their opinions, but an editorial in Nature urges the neuroscience community to voice its doubts loudly and clearly, as well as to get ready for a long public debate on the ethical implications of this technology and on the nature of privacy itself.4

A new technique for examining neuroimaging data, called pattern classification, is also making it possible for scientists to predict with some accuracy what a subject is viewing, even before the subject is aware of it. Although this ability may conjure up a disturbing image of mind reading, any attempt to use pattern classification to detect lying, for example, is subject to the same limitations as a conventional polygraph, such as noise introduced by emotional responses, according to an editorial in Nature Neuroscience. The impact of pattern classification techniques is more likely to come at the level of basic research, where it will begin to show scientists “not just where in the brain information is processed, but how.”5

Another area of interest is the attempt to identify biological markers, such as brain abnormalities or specific genetic mutations, that suggest a tendency toward violence. A thoughtful review by Nigel Eastman and Colin Campbell in Nature Reviews Neuroscience questions whether causation, in the legal sense of the term, can be established by such markers, and, if so, whether a person with one or more of these biomarkers can properly be held in detention preventively, to protect the public from future harm.6

Emerging Technology and the Human Brain

A unique clinical trial in which brain meets computer is that of BrainGate, a custom-developed prosthetic arm and hand used by research participant Matt Nagle, the cover feature of the July 13 issue of Nature (also discussed in the Nervous System Injuries section). Nagle, tetraplegic from an injury that severed his spinal cord, activates the device purely by thought—that is, by brain signals representing his intention to extend his arm, open and close his hand, and so on. The movement signals that Nagle’s conscious brain sends are picked up by a 96-electrode array implanted in his motor cortex, decoded, and transmitted to drive the movements of the prosthesis.

In contrast to several other types of assistive technology for patients with multiple paralysis, such as one that is driven by electrical activity on the scalp or that relies on eye movements, this neuromotor prosthesis does not call for months of training or require the user’s full attention. For instance, Leigh Hochberg and colleagues report in Nature that Nagle is able to carry on a conversation while opening simulated e-mail or moving the robotic hand or arm.7

An increase in the information-processing speed, and hence the capabilities, of neuromotor prostheses of the future will raise new questions about who may properly benefit from them and in what ways (therapeutic, financial, perhaps even psychosocial).

In fact, Stephen Scott, also in that issue, suggests that through information-feedback circuits that already exist in the brain, the use of these prostheses may subtly modify the way the brain signals themselves are organized so that brain and man-made devices meld together increasingly well—a hopeful prospect for people with paralysis.8

Other technological advances, in the area of brain imaging, are raising concerns about “incidental findings,” unexpected signs of a possible disease that are discovered in the course of research on an unrelated topic. Reports of incidental findings are increasing. Writing in Science, a working group of about 50 experts from the fields of medical imaging, biomedical ethics, and law considered the question of whether researchers have an obligation to disclose such incidental findings to their study participants, and, if so, under what circumstances.9

The answers are far from obvious. Some of the illnesses that might be detected are very serious, yet the rate of false positives in this setting may be high, and in the end a firm result could come only from a second scan, to be read this time by a diagnostic radiologist. If incidental findings show up on a research scan, does the research subject have a right to know about them, a right not to know, or both?

Four brain scans 
Brain activity in a vegetative state: A patient in a persistent vegetative state showed activity in the same brain areas as healthy volunteers in response to spoken commands to visualize herself playing tennis or moving through her house.  (Image courtesy of Adrian Owen) 

The working group urges all researchers who use brain imaging to decide in advance how incidental findings will be handled. This protocol should be stated clearly as a step in the informed-consent process, they say. Future research that focuses on incidental findings may, of course, lead to new recommendations, but ensuring scientific integrity and engendering public trust will remain the guiding principles.

Nuances of Nonconscious States

Research in 2006 brought attention to unusual patients with severe brain injuries. In the Journal of Clinical Investigation, Henning Voss, Nicholas Schiff, and fellow researchers from New York, New Jersey, and New Zealand described the spontaneous rehabilitation of a man who had lain for 19 years in a minimally conscious state, unable to move or speak as the result of a car accident.10 His condition had shown gradual improvement over the years; even so, his recovery of consciousness, fluent speech, cognition, and movement in three of his four limbs was unprecedented.

The researchers used a noninvasive imaging technique, magnetic resonance diffusion tensor imaging, to examine his brain. They found evidence of the regrowth of axons, which would enable new connections to form within the brain.
A second patient described in the paper, who (also as the result of a car accident) had spent more than a year in a persistent vegetative state and another four years in a minimally conscious state, has shown no comparable clinical improvement or axonal regrowth, but may yet do so. The authors call for more imaging studies using diffusion tensor imaging and also positron-emission tomography, beginning shortly after brain injury, in order to reveal more about the hopeful phenomenon of long-term rewiring in the brain.

An encouraging observation emerged as well about brain activity in a patient in a persistent vegetative state (PVS), from research in Science led by Adrian Owen.11 After a car accident and five months in a persistent vegetative state, a young woman who appeared unresponsive underwent an fMRI scan of her brain that demonstrated clear evidence she could carry out several complicated cognitive tasks in a normal fashion.

In response to spoken sentences, even those containing ambiguous-sounding words (“The creak came from a beam in the ceiling”), the language areas of her brain displayed activity equivalent to that of healthy volunteers. In addition, she showed normal brain activity in response to a spoken command to visualize herself playing tennis or walking through the rooms of her home.

These feats are particularly impressive because they indicate, in a very concise manner, brain activity underlying so many aspects of what we think of as consciousness: intention, awareness of one’s body, visual tracking of an (imagined) moving object, and the recollection of (imagined) familiar surroundings. Some neuroscientists had hypothesized previously about “islands” of preserved function in patients in a persistent vegetative state that might be undetectable by standard clinical methods.

The mental agility of this immobile patient seems to confirm that such “islands” can occur as hypothesized and generates hope of uncovering some kind of window into the recesses of nonconscious states.

Because only a minute number of patients have been found to improve, questions arise about whether costly experimental interventions (such as deep brain stimulation or transmagnetic stimulation) to try to stimulate recovery of some communication functions should be used in all patients. Current research aimed at determining whether these interventions will be effective, and under what circumstances, may help to identify patients most likely to benefit.