The Philosophical Problems at the Heart of Neurology

 All Things Neuro-..., Featured, Neurology, Philosophy  Comments Off
Jul 142011
 

I was fortunate enough to experience what neurology is like at the cutting edge recently, when I completed an away rotation in the United States. As should be obvious to anyone, this represented a markedly different practice to the one I usually partake in as a medical student in Africa. But whilst the incomparable economies and educational systems, together with the growing HIV epidemic in sub-Saharan Africa, gives a different flavor to a neurologist’s practice, there are questions at the core of the discipline that are the same everywhere. The most interesting of these relate to the ‘mind-body problem’ in philosophy of mind, and creep into the neurologist’s practice daily.

These questions follow naturally from the fact that most of a neurologist’s time is spent diagnosing, treating, and discussing the plethora of conditions that cause and are caused by damage to the brain. Whether resulting from ischemic, infectious, inflammatory or degenerative processes, the resulting clinical syndromes are both fascinating and depressing. And whereas the distribution of etiologies will differ from country to country, our common anatomy ensures that a similar scope of phenomenology will emerge from different pathologies. Despite the obvious differences between a mother in sub-Sharan Africa whose son is rendered aphasic from tuberculous meningitis, and a father in North America whose daughter’s personality is eroded by a brain tumor, their questions, fears and difficulties can be remarkably similar.

With regards to the resulting philosophical questions, it would seem that we innately1 assume a dualistic position of sorts, taking a viewpoint that’s temporally and spatially unbound to that of our bodies. Haggard2 notes, with regards to our actions, that a “ dualistic view of endogenous causation is engrained in our normal language”. This dualistic language is common amongst patients affected by neurologic illness, as well as their family members. Consider the difference between a statement like ‘my knees aren’t as reliable as they used to be’, and the comparatively absurd ‘my brain isn’t working as well as it used to’. The latter, when considered with anything other than momentary attention, should strike us as bizarre. After all, it has been an endeavor of neuroscience (not to mention other fields like philosophy, cognitive psychology and artificial intelligence) to explain how matter, or the brain in this case, results in the internal environment and our representation of self. The mind, being entirely dependent on the physics constituting the brain, can hardly refer to its constituting parts as ‘not functioning’ without at the same time saying that it is not functioning as it should.

This ‘benign dualism’ does have its advantages. The suspension of the belief that a loved one is nothing but an appropriately arranged constellation of atoms immediately creates a certain sanctity. Problems with the sanctity of human life notwithstanding, it does contribute to a worldview that we all prefer. As Dennett has argued3, the act of treating someone not only as a body, but as a ‘self’ that has a body helps preserve a ‘belief environment’ around things like death and birth, and protects individuals close to those stages. Whilst we might admit that this ‘center of narrative gravity’ is a fiction of sorts, it is a useful construct that informs our day-to-day behavior. Furthermore, viewing people ‘as brains’, or subscribing to a ‘brainhood’ that views people as ‘cerebral subjects’4, opens several philosophical wormholes. Perhaps justifiably, people fear that this view will result in fatalism, where concepts like free will and responsibility will be defunct, resulting in an unavoidable and irrevocable slide towards anarchy.

The issue is most dramatically played out in the legal system5. Here claims of criminality being the result of ‘organic’ pathology bring the issues of free will and responsibility into focus. If sufficient evidence tying someone to a murder is provided, the court has little choice but to find someone guilty and duly sentence them. But what if it is shown that the defendant has a brain tumor disrupting the pathways involved in moral reasoning and inhibition? Suddenly, it’s as if this ‘physical’ reason for the behavior removes the responsibility from the defendant. But surely there are physical processes underlying all of our actions? As neuroimaging improves, deciding where to draw the line will become more complicated.

More commonly, our unwillingness to accept that we are our brains, or at least a subset of thereof, is the reason behind the remarkable interest in cognitive neuroscience, particularly when it involves functional neuroimaging. Rarely a week goes by without a journalist reporting on a specific human quality or ability that has been found to correlate with certain brain regions. We hear of empathy being localized to the anterior cingulate cortex6, of romantic attraction activating the limbic system7 and of deception involving the prefrontal cortex8. But were our position on these things at all altered by these findings? Surely a difference in, for instance, emotional disposition between people would always have been dependent on a difference in brain function? The fact that we now have tools sophisticated enough to show the difference is still impressive, and the research has many interesting implications, but we didn’t discover a new fact about the mind-body problem.

This double-think we all partake in, where we view others and ourselves as being more than matter and physics, yet at the same time acknowledging that we can explain all observed and experienced phenomena in physical terms, lies at the heart of what makes neurology such an interesting and difficult field. In stark contrast to a patient whose heart is failing, the patient suffering from a ‘failing’ brain is suddenly confronted by the abovementioned philosophical conundrums. They can’t help but acknowledge that the illness isn’t just affecting ‘their body’, but that it is affecting ‘them’. As Alcauskas9 notes, neurological illnesses “dehumanize in a way that heart disease and renal failure do not.”

Despite the facetious safety our dualistic positions provide, whereby ‘we’ exist in a dimension unrelated to the mere physics of everything we observe, the neurological patient draws attention to the fragility of our being. A relatively simple physical process, like a clot becoming lodged in a vessel, can remove much of your brain and hence much of you. Your wife can be rendered a stranger, your taste in music may change, you may lose the concept of a ‘left side’. For those unfortunate enough to see their husband slowly change from the loving and caring companion to a paranoid and violent stranger, as may happen with frontotemporal dementia, the innocence and functionality of a dualistic view is unavailable, and insofar as we can provide physiological answers, these are wholly inadequate10. For us as physicians, in the first world just like in the third, these patients provide a constant reminder of the fragile relationship between the deterministic physical processes happening behind our eyes and everything we value.

 

References

1. Carruthers P. Cartesian epistemology: is the theory of the self-transparent mind innate? Journal of Consciousness Studies. 2008;15:28–53.

2. Haggard P. Human volition: towards a neuroscience of will. Nat Rev Neurosci. 2008;9:934-946.

3. Dennett DC. Consciousness Explained, 1st ed. London: Penguin Books; 1993

4. Frazzetto G, Anker S. Neuroculture. Nat Rev Neurosci. 2009;10(11):815-821.

5. Mobbs D, Lau HC, Jones OD, Frith CD. Law, responsibility, and the brain. PLoS Biology. 2007;5:e103.

6. Amodio DM, Frith CD. Meeting of minds: the medial frontal cortex and social cognition. Nat Rev Neurosci. 2006;7:268-277.

7. Younger J, Aron A, Parke S, Chatterjee N, Mackey S. Viewing Pictures of a Romantic Partner Reduces Experimental Pain: Involvement of Neural Reward Systems. PLoS One. 2010;5:e13309.

8. Abe N, Fujii T, Hirayama K, et al. Do parkinsonian patients have trouble telling lies? The neurobiological basis of deceptive behaviour. Brain. 2009;132:1386-1395.

9. Alcauskas M, Charon R. Right brain: reading, writing, and reflecting: making a case for narrative medicine in neurology. Neurology. 2008;70:891-894.

10. Adamo A. Of Minds and Maps. Neurology. 2009;72:1364-1365.

Biological Psychiatry: Endless fascination; indeterminate value

 Biological Psychiatry, Featured, Psychiatry  No Responses »
May 022011
 
phrenology

There is no doubt that our growing knowledge regarding the biological origin of psychiatric illnesses is a source of endless excitement. Since the time of the Spanish neuroscientist Santiago Cajal, discoveries in neuropathology have helped establish the authenticity of a diagnoses ranging from epilepsy to ADHD, and so given science another valuable victory over centuries of ridicule and stigma. More recently, advances in functional magnetic resonance imaging (fMRI) have provided us with a new, essentially non-invasive total for studying psychiatric illness on a biological level. It is hoped by many that these techniques will allow us to eventually fully understand the pathophysiology of a group of illnesses that have, until recently, been outside the realm of true scientific study.

As tends to be the case with any new technology, there are many who have their doubts. This ranges from the extremes of the anti-psychiatry movement, which continues to favour a “dualistic approach”, arguing that mental illness is a property of the ineffable soul, rather than an illness of the brain. Of this we shall speak no further, as it falls outside what this blog will generally consider to be valuable enquiry. Of considerably more value are the legitimate concerns of those who, whilst acknowledging that all psychiatric diseases possess an organic basis, question our ability to accurately ascertain these origins using present technology and methodology. William Uttal, in his book “The New Phrenology: The Limits of Localizing Cognitive Processes in the Brain” argues that we need to more clearly examine the limitations of modern techniques, furthermore implying that the current obsession with localisation is a form of “neophrenology” – or, other wise put, an echo of a once popular pseudoscientific practise, were conclusions about the state of development of specific brain areas were made based on the size of the overlying cranial bone!

Whilst I think that this claim (which is perhaps something of a caricature of Uttal’s actual thesis) is somewhat harsh, it is true that fMRI does have several limitations, which have been well described in the literature. What I wish to focus on in this piece is not the technical difficulties, but rather what I consider to be an inherent methodological problem in using fMRI as a diagnostic tool, which is perhaps the endpoint that is most hoped for by psychiatrists in cognitive neuroscience.

The first problem (which, incidentally, did not escape Uttal’s attention either) relates to weakness within the psychiatric taxonomy. Anyone who has closely followed the development of the fifth diagnostic and statistical manual of mental disorders will be aware of how much classification has changed (and will probably continue to change) within psychiatry. Until now, such classifications have been almost entirely based on phenomenological data, which has allowed us to group syndromes according to commonly occurring symptom clusters. These syndromes initially began as simple descriptions of mentally ill patients, and patients that behaved in similar ways were classified as having the same condition.  This lead to several errors, owing to the fact that several conditions, when viewed for a brief time, look largely the same – even experienced psychiatrists struggle to differentiate between the mania of a patient with bipolar mood disorder and the psychosis of the schizophrenic. Subsequently, syndromes were refined based on further assessment of age of onset, manner of progression and other longitudinal factors. Modern classification systems, such as the DSM-IV (diagnostic and statistical manual of mental disorders, fourth edition), use essentially the same criteria, albeit by employing enormous databases, pooled from studies conducted all over the world. Despite this, significant controversy remains, and most psychiatrists agree that we are some way from achieving anything like a “definitive” classification.

It is worth remarking how different this process is from what has happened in other fields. Take lesions of the cardiac valves, for example. Initially, they were probably also only described according to symptoms, which would have been largely indistinguishable from those of several other causes of heart failure. However, with the advent of instruments as simple as the first, rudimentary “stethoscopes”, one could directly observe the presence of heart murmurs, which could help determine not only the presence of a heart lesion, but, with some skill, the precise valve that was affected, and even the degree to which it was affected!

It is difficult to resist the temptation to suggest that fMRI will do for psychiatry what the stethoscope did for cardiology, but clearly things are not that simple. Going back to classification – psychiatry has a unique problem in that the vast majority of its conditions have been defined based on symptoms, rather than objective pathology. Again, compare cardiology, which has some syndromes (like heart failure, for example), but several diseases, which have been defined based on their anatomical origin and pathophysiology. This is crucial to successful treatment – two different valve lesions may present in the same way, but treating them in the same way is seldom appropriate. Few patients will appreciate an aortic valve replacement when the their disease is clearly confined to the mitral…

Is this the case in psychiatry? I would argue that we don’t know, but that we should probably try and find out before we decide how much diagnostic faith we put into neuro-imaging.  Evidence suggests that patients who present with symptoms of depressed mood, most of the day, every day, along with a prescribed number of other classic symptoms, can be classified as having “major depressive disorder”. This, in turn, can be regarded as a condition that is in most cases responsive to treatment with antidepressants and psychotherapy, either separately or in combination.

In recent years we have successfully described a range of findings, both on neuroimaging and neuropathological study, that appear to be common in patients with depression. Would it not then be reasonable to test patients for these changes, in order to determine whether or not they really have “depression”, rather than some other condition masquerading in its guise? How then, would we respond to such a patient? Would it be appropriate to deny them treatment? Would we label them as being malingerers? These could all be true, but another potential answer needs to be addressed.

When we begin to search for the neurological underpinnings of a disease, we begin with a syndrome. Assume we have a hundred patients with clinical depression on whom we conduct fMRIs, and so determine which features appear to be most common. Lets further assume the unlikely case that all of these patients have at least one or more finding in common, which will then form our new “definition” of what it means to have depression.

All we will really have done in such a study is to define one possible cause of a syndrome. As the case of heart failure and valve lesions have taught us – there could be several others. Perhaps even several of the patients in this study have developed depression due to a different mechanism, and only incidentally display some of the same findings as the rest of the group. And what of our patient? Who is to say that she herself is not the victim of some hitherto undescribed condition, causing her to develop the symptoms of depression on the basis of a unique neuropathological mechanism?

Things in psychiatry are not as clear as in cardiology (or gastroenterology or pulmonology or any other field in medicine for that matter). The functions of the brain are not intuitively elucidated from its anatomy, and although we must recognise that it is the starting point for all mental illness, it is not necessary to assume that it should form the starting point for diagnosis. Some would argue that a syndromic classification has its own weaknesses, a claim to which we must acquiesce – but is it not ultimately the method that would best benefit our patients, at least given our current knowledge?

If a patient comes to a psychiatrist feeling depressed, anhedonic and suicidal, she needs help, whatever might really be going on. A psychiatrist cannot claim that he only has a role to play when a “true” psychiatric illness can be detected – after all, if we are to fully reject dualism, we must acknowledge that even the most minor aberration of psychological function finds its origin in the brain, and since this is the organ we claim to treat, we cannot ignore it. It may sound like I completely reject the role of fMRI, and perhaps even biological psychiatry as a whole. But anyone who has bothered to explore my blog would see this is unlikely – indeed, I have dedicated an entire section to this subject. And next week, I hope to address the issue of treatment response. Here, we will perhaps be exploring an area where fMRI may truly come into its own, with potentially revolutionary consequences.