How do we manage the reality of a different language of the mind and brain? This is more than just semantics as the two languages have been explored in fundamentally different ways and have resulted in entirely different branches of science. For the language of mind we have introspection, the analysis of language and the use of language in assessment tools. The language of brain is one primarily of physiology and as such the branches of science that attend to this are based on the sensory observations regardless of whether they are direct as in the case of neuroanatomy or indirectly as in the case of the many neurophysiological investigations such as electroencephalography of fMRI to name just a few.
Whilst there are many pieces of research which investigate the question of combining these two disciplines these are necessarily more difficult in their methodology and can be more ambiguous than either field alone. There are notable exceptions such as the recent investigation of dream sleep in which the predictive utility of the neurophysiological investigations was found sufficient to enable researchers to predict the actual content of the subjects dreams. This was not of the type seen in the Gallant study which was based on complex modelling although similarly remarkable in its achievements. Both of these studies go some way in showing that there can be a successful combination of these approaches although the Gallant study was not dependent on the subjects reporting of their inner experiences as it was visible and obvious from the moving images that were reverse engineered.
However such studies have not always been so successful and many studies investigating the same question will find opposing answers. So what is the solution? My answer in previous posts has been that we should address this by developing a language of the mind and brain and one which is sufficient to cross the narrative bridges that have arisen in different scientific communities. Such a language has to have a pragmatic utility and to arise from the scientific findings. Such a language would have to be intuitive to the clinician and the scientist alike. Perhaps such a language is not possible although there are considerable benefits if it were. However the language is not the complete answer as there is a much deeper problem.
That problem is what we expect when we undertake investigations into the mind or brain. This is a more fundamental question. The core feature in discussions of the mind is of course consciousness. For Freud consciousness and the unconscious mind were two extremely important aspects of psychoanalysis. Both are intuitively obvious to the general public – so much so that they hardly need any introduction. We know what it is to be consciously aware from moment to moment and we know what it is to be asleep or otherwise unconscious. What though is the ultimate goal of the researcher of the mind? Is it really to make a prediction about what a subject will be thinking in one or two minutes time or to understand the essence of relationships between people for example. I think here we are talking about an investigation of the mind independent of the brain in which neurophysiological correlates are unimportant. In this case the researchers are trying to develop a model which is imperfect but encapsulates a property of the human mind. We may say that this is an aspect of consciousness. Consciousness is not something tangible, something that we can see or hear other than as a distortion through language. Consciousness is an inherent property of another person’s brain and mind which we infer through direct observations (e.g emotional expressions, posture and so on) or hear (e.g through the person’s discussion of their internal experiences).
In so doing there is a certain amount of negotiation that we must make in everyday life in order to understand the mind of another. The ultimate measure of internal experiences is epitomised by the psychometric assessment tool. If we wish to assess a certain characteristic of the person’s mind there are many tools to do so which are effective in doing this. But let us investigate a little further. Suppose we have a tool for happiness. The tool must go through exhaustive trials to validate it. We must be certain that the construct that is being examined is the construct we understand to be happiness. There are various methods to ensure this and I will not go into these here. Let us suppose that this stage has been passed. There are a set of tools to be administered by the rater or self administered. These questions are distilled down to the most essential so that the tool is pragmatic. There are various approaches to refining these questions and we may be certain that the remaining questions are the most effective for the job.
Let us for a moment investigate this question a little further. Let us assume that I describe myself as happy and I ask a question. The question is ‘are you happy?’. When responding to this question I must pause for a moment and consider the response and consider what is being said in the question. Suppose I am experiencing that happiness now. How do I get to responding that I am happy. There are several steps which we may overlook as they are intuitive and immediate. I will pause briefly to consider the question and understand it. Assuming that I have understood it which itself is composed of multiple steps I must then make a choice. This to me, this process of making a choice is qualitatively different from the process of being happy. I must move from the non-reflective state of being happy to the considered state of wanting to make a choice. Assuming that i’m now in a completely different state i must recall how I was a few moments previously before this state was interrupted. If I don’t I may be in a state that is more common to answering questions than a natural state. This may be associated with a slight anxiety as anyone taking an exam will know.
Let us suppose I must give my state of happiness a number from 0 to 10 and that each number has a description which helps me to better match it to my internal state. Now if I answer a 6 or a 7 and come back tommorrow and in a similar state of happiness answer a 4 or 5 we will see that there is a false reduction in the attempt to get to the final state which is a numerical correlate of the experience. Let us suppose on the other hand that it is a good tool and always consistent and that my state is – whatever this means – is always ‘5’. In other words let us assume that there is a perfect correlation between the number and my experience – a perfect numerical compartmentalisation of that experience. If that happiness were to increase by just a small amount it would be correlated with a small increase in the numerical score and that would be absolutely perfect in its relationship. This is a completely hypothetical example as it is such a difficult area and what we are trying to do is so artificial to some extent. When we consider a larger number of people the practical difficulties – the noise, the error – averages out and we get a better relationship.
Let us consider the tool on an individual basis. Suppose that I am answering the question about happiness. The experience of happiness could be subconcious in that i’m not consciously aware of it. I’m not focusing my attention on it from moment to moment but when I need to I can attend to this subconscious experience and create a numerical correlate. I am the instrument for this transition rather than the rater per se. Now I know that from moment to moment my thoughts vary. My thoughts might wander, I might simply be enjoying the scenery, I might be lost in a train of thought, I might be undertaking a complex piece of work but all the time I may be happy. This would be an emotional state. From the neurobiological perspective its entirely possible that while i’m doing all of these things I can be in a state of emotional happiness and its not too far fetched to suppose that the intensity of this emotional experience can vary from none at all through to extremely happy. In this simple example it is not too far fetched to suppose that my conscious experience can be correlated with a number in a meaningful way.
If I am happy this may also influence my thoughts on a continuous basis. I may experience happier thoughts, I may solve problems in a different way and so there is an opportunity for me to examine other qualitative aspects of my conscious experience without needing a numerical correlate. If my rating were zero and it was an accurate representation of my internal state then I might have few thoughts or my thoughts would be of a qualitatively different nature. Indeed if we think hard enough it is entirely possible to turn those qualitative aspects of conscious experience into quantitative components that correlate with my conscious experience.
Where do we go from here? We’ve assumed in the argument up until this point that a certain tool with the right questions will provide us with a simple measure of our internal experience and we have assumed that this is a perfect correlation – that it is a useful measure. What then do we do? I will argue that there are two things we are interested in. The first is providing an explanatory framework and the second is making a prediction. I will further suggest that when we look at physiological correlates we are doing exactly the same. In the clinical arena we are interested in whether people have one or another type of illness. We are interested in seeing if there is a response to treatment or if we should try another approach. So returning to the basic question of whether we have a neurophysiological correlate of internal experience – what do we want it to do. I would say we want it to give us a measure of the severity of an internal state, a measure of the consistency of the internal state and we want it to provide us with an explanation and a prediction.
Returning to our example about happiness. Let us suppose that I have identified the internal subjective state of happiness and assigned it a number 3 and i have obtained the physiological correlate and it is activation in a brain circuit which includes the Ventromedial Prefrontal Cortex. Suppose on the fMRI scan that I therefore have the pattern on a 3T scanner – if I scan people over and again – and given the noise/error rate – I can suspect that there is a correlation between one and the other. I can infer one state from the other. If I know the score on the psychometric tool I can know the physiological correlate. Or if I see the fMRI scan result and assuming a 1:1 correlation between physiological state and conscious experience, I can then infer the conscious experience and the numerical correlate.
Even taking into consideration the problems with methodology which are covered elsewhere which show up the flaws in this idealised argument, the next problem is that the fMRI scan data is fairly limited in its remit. There have been many studies which have pointed to flaws in the methodology ranging from poor blood flow/neural activity correlates to blood flow time delay. So what we’re looking at are changes in blood flow and changes in the BOLD signal. But that doesn’t tell us much about conscious experience. We know that conscious experience is most likely to be due to neural activity. I may say without doubt that conscious is an epiphenomenon of neural activity. The next problem is that if we return to the fMRI study data that we’re looking at the changes in blood flow. We could have a large amount of neural activity with little change in blood flow or the reverse case and this area is still poorly understood. So matching one methodology with another still leaves us with a problem.
The discussion above illustrates a few points. An idealised argument helps us to frame our thoughts. However the deeper we examine questions the more we see how flawed such idealised arguments are. The discussion of language is an important one. Even more important however is the ultimate goal of our research questions and the realistic objectives of interdisciplinary working and model building.
Related Resources on the TAWOP Site
Index: There are indices for the TAWOP site here and here Twitter: You can follow ‘The Amazing World of Psychiatry’ Twitter by clicking on this link. Podcast: You can listen to this post on Odiogo by clicking on this link (there may be a small delay between publishing of the blog article and the availability of the podcast). It is available for a limited period. TAWOP Channel: You can follow the TAWOP Channel on YouTube by clicking on this link. Responses: If you have any comments, you can leave them below or alternatively e-mail firstname.lastname@example.org. Disclaimer: The comments made here represent the opinions of the author and do not represent the profession or any body/organisation. The comments made here are not meant as a source of medical advice and those seeking medical advice are advised to consult with their own doctor. The author is not responsible for the contents of any external sites that are linked to in this blog.