Ethical issues in neuroscience Thomas Fuchs Purpose of review Introduction The study gives an overview of ethical questions raised by Progress in neuroscience is rapidly increasing our knowl- the progress of neuroscience in identifying and intervening edge of neural correlates of the mind. Researchers are in neural correlates of the mind. beginning to identify brain processes that are related to Recent findings experiences and concepts such as free will, agency, moral Ethical problems resulting from brain research have judgment, self and personality. At the same time, those induced the emergence of a new discipline termed processes become increasingly accessible to specific mod- neuroethics. Critical questions concern issues, such as ifying techniques. This development raises ethical pro- prediction of disease, psychopharmacological blems whose importance is likely to surpass even the enhancement of attention, memory or mood, and implications of modern genetics. What are the social and technologies such as psychosurgery, deep-brain cultural consequences of technologies that enable stimulation or brain implants. Such techniques are capable humans to manipulate their own minds? What impact of affecting the individual’s sense of privacy, autonomy and will neuroscience have on our self-understanding and our identity. Moreover, reductionist interpretations of concept of humans in general? Are we facing an age of a neuroscientific results challenge notions of free will, ‘technology of consciousness’? responsibility, personhood and the self which are essential for western culture and society. They may also gradually In view of these questions, an intensive dialogue between change psychiatric concepts of mental health and illness. neuroscience and the humanities becomes more than These tendencies call for thorough, philosophically ever necessary. Growing awareness of this need has led informed analyses of research findings and critical to the emergence of a new discipline termed ‘neuro- evaluation of their underlying conceptions of humans. ethics’ [1]. Psychiatry has started to join these activities Summary as shown by a recent conference on philosophy and Advances in neuroscience raise ethical, social and legal neuroscience in Leiden [2]. The ethical issues raised issues in relation to the human person and the brain. by neuroscience have proven to be twofold: On the Potential benefits of applying neuroimaging, one hand, new methods and techniques, by laying bare psychopharmacology and neurotechnology to mentally ill neural correlates of personal identity, cause problems of and healthy persons have to be carefully weighed against individual rights on privacy, noninterference and inviol- their potential harm. Questions concerning underlying ability. On the other hand, the results of neuroscience concepts of humans should be actively dealt with by tend to support reductionist concepts of free will, auto- interdisciplinary and public debate. nomy and the self. This interferes with the fundamental intuitions that we have of ourselves and questions notions Keywords such as responsibility and culpability on which central brain enhancement, ethics, free will, mind–body problem, institutions of our society are based. Accordingly, the neuroscience, self present review explores these issues on the levels of (a) practical and (b) conceptual ethics. Curr Opin Psychiatry 19:600–607. ß 2006 Lippincott Williams & Wilkins. Ethical problems of neuroscience-based Psychiatric Department, University of Heidelberg, Heidelberg, Germany diagnosis and interventions Correspondence to Thomas Fuchs MD, PhD, Professor of Psychiatry, Head of the This section concerns problems resulting from neuro- Section Phenomenology, Psychopathology and Psychotherapy, Psychiatric Department, University of Heidelberg, Voss-Str. 4, D-69115 Heidelberg, Germany imaging, pharmacological brain enhancement and new Tel: +49 6221 56 4422; fax: +49 6221 56 8094; technical interventions in the brain. e-mail: [email protected] Current Opinion in Psychiatry 2006, 19:600–607 Neuroimaging: methods and techniques Abbreviation In accordance with the growing sophistication of neuro- ADHD attention deficit hyperactivity disorder imaging techniques, they may be used to infer not only people’s actual mental states but also their unconscious ß 2006 Lippincott Williams & Wilkins attitudes and predispositions to a particular kind of 0951-7367 behaviour. Recent examples include brain structures involved in racial biases, trustworthiness, moral reason- ing, economic cooperation, social rejection and even 600 Ethical issues in neuroscience Fuchs 601 consumer brand attachment [3–5]. This inference opens Reliability and validity up a wide array of options beyond genuine medical use A general problem posed by these techniques is the for diagnosis and treatment. Thus, the brain’s reaction to apparent objectivity of visualizing the ‘brain in action’. certain stimuli could be used to not only detect predis- The complex assumptions required for jumping over the positions to mental illness (‘neuro-screening’), but also gap between subjective experience and electromagnetic to recognize violent offenders (‘brain fingerprinting’), signals are hardly known to the public. Imaging studies examine participants’ responses to advertising (‘neuro- are based on probabilistic covariances and not on causal marketing’) or optimize children’s school carriers by connections. Their interpretation depends on the design measuring specific cortical functions. and theory behind the study and is highly influenced by cultural and anthropological frameworks [15]. More- Unconscious attitudes over, it is unclear in how far complex phenomena such Imaging studies [6,7] found that white participants had as social attitudes and behaviour may be grasped by the greater activity in the amygdalae when viewing pictures of rather simplified scenarios of neuroscience, applying black people as opposed to white people’s faces, pointing video games or faces on a screen instead of real social to unconscious racial attitudes. Similarly, sexual prefer- interactions [3]. Therefore, it seems questionable to rely ences or aggressiveness have neuroimaging correlates on neuroimaging techniques in complex social issues which could be used to detect further potential violence such as solving a crime, assessing a person’s dangerous- in sexual offenders [8]. Furthermore, positron emission ness, or choosing young people’s training options. tomography (PET) studies [9] with committed murderers show poor functioning in their prefrontal cortex to a Neuroimaging is bound to momentary states, and infer- significant degree. Specific functions related to this area ences on personality, propensities and actual dangerous- are the ability to control one’s impulses, to weigh uncertain ness are hardly warranted. For example, latent aggres- risks, to act with regard to potential negative con- siveness need not be harmful as long as it is under sequences, to take another viewpoint and to empathize. control, whereas it is well known that the most dangerous These areas can be damaged in subtle ways, even through offenders act without any or with low emotional involve- childhood abuse or severe neglect, making it difficult to ment [16]. Therefore, the information gained by neuro- draw a line between culpable offenders with normal brain imaging might become a helpful tool, but will at most findings and persons to be exculpated because of obvious supplement, not replace existing criteria of responsibility brain injury [10]. Thus, it seems likely that both healthy in the criminal justice system [17]. The wide-spread and certifiably insane delinquents could in future be misunderstanding of brain scans as direct measures of submitted to certain tests and brain scans in order to assess psychological states or even traits, however, carries the their potential dangerousness. risk that courts, parole boards, immigration services, insurance companies and others will use these techniques Detection of deception prematurely [5]. More sophisticated imaging techniques may enable researchers to distinguish between true and false memory Predictive neuroimaging and to improve the methods of lie detection. Recent Another ethical issue concerns the increasing use of studies [11] found activation in the dorsolateral prefrontal neuroimaging to predict later onset psychiatric disorders. cortex associated with the active suppression of memory. Thus, functional magnetic resonance imaging (fMRI) Other studies [12,13] indicate that intentional deception scans of adolescents considered at high risk of schizo- may be correlated to rather specific neural substrates such phrenia showed structural and functional abnormalities in as the anterior cingulate cortex. Finally, ‘brain finger- certain areas [18,19]. Early pharmacological intervention printing’ attempts to demonstrate the knowledge of facts might prevent or delay the onset of psychosis. The about a crime or other information by measuring the complexity and plasticity of the brain, however, definitely electroencephalogram (EEG) responses (p300 wave) to restrict the reliability of such prognoses [17]. What relevant words or pictures presented to the participant degree of probability would count as sufficient? Which (‘guilty knowledge test’ [14]). This technique was long-term side effects would be acceptable? The possible already used in a court case in Iowa, where a murder benefit of predictive imaging would have to be carefully conviction was reversed