The temptation to create personality categories to measure in the context of any branch of psychology is virtually overwhelming, because unless you create categories you have nothing to measure and if you don’t have anything to measure then you can’t call yourself a science. This is because the relationship between psychology, as a science of the mind (if such a thing can exist at all), and neuroscience, as a science of the physical brain, remains only dimly understood. Neuroscientists can try measuring the presence or absence of certain sorts of chemicals in any particular brain but it is no easy exercise and it is not obviously or easily correlated with particular types of personality or behaviour. Nonetheless neuroscientists can say that they are practising a science because they are measuring something. Likewise pharmacologists invent medicines or other chemicals that have effects on the neurological system of the body, and the interaction between those pharmaceuticals and neuroscience can be measured in an approximately empirical way.
However what proves to be much more difficult is to measure consistent and robust relationships between psycho-pharmaceutical medications and behavioural or phenomenological changes within the field of psychology as opposed to neuroscience. This interface is particularly delicate in our want of understanding it; but the point is that unless we create personality categories and try measuring them somehow, then there is no science we can do and we cannot even attempt to measure neuroscience and technology each against the other. That is why efforts like the Myers-Briggs test remain so popular: they are attempts to create ways of measuring non-quantitative, qualitative things like personality types that ordinarily defy categorisation.
Without something like Myers-Briggs, it turns out that you can’t really do psychology at all. Nor can you do things like management psychology, educational psychology, recruitment psychology or all the other derivative soft disciplines that hinge upon the presupposition that psychology is a science. Hence although Myers-Briggs has been comprehensively discredited, people keep on using it because at least it is a structure even if one that does not withstand the slightest heavy scrutiny. The main problem with Myers-Briggs is that it is arbitrary and that it tries to fix personalities which in reality are highly variable. What Myers-Briggs describes as a personality “type” is really just one of a set of overlapping moods that we can all lapse into from time to time. We may vary from one day to the next, depending on a particular experience we have had; that’s why when people sit Myers-Briggs tests on consecutive days they get different results. None of this correlates to any necessarily different neurological reactions in the brain; it’s just that human nature is fungible in a way that science is not; or, as Donald Davidson famously observed in his essay Mental Events, “mental events are not susceptible to capture in the nomological net of physical theory”. They work by different kinds of rules.
It follows that it is not just the Myers-Briggs model that is flawed, in choosing the wrong personality types to measure or making it too easy to cheat by guessing the answers that will reveal the desired personality type; but rather that the very idea of any model in the nature of Myers-Briggs is flawed; human personalities cannot be placed upon a grid and measured on axes and with units. Units of what? What is being measured? What is the unit of extroversion or of sensing? There are no such units, and that is because there is nothing that is really being measured, except how people answer to arbitrary questions on a questionnaire. Asking people questions in this way is an extremely unreliable way of doing science and not just because the same person can answer the same question differently on a different day or in a different context, for example depending on what their mind, or because they want to deceive the person asking the question or to have a certain sort of impression of themselves that makes them feel better. It is because the answers to questions do not represent some objective underlying scientific feature or quality of the personality and therefore questionnaires are not accurate measurements of anything.
It is not only management, relationship and recruitment psychologists who fall into these traps. Psychiatrists routinely do it as well, because they are faced with a patient describing symptoms and they have at their disposal for the most part a series of pharmapsychoactive medications that will have neurological consequences of certain kinds but they don’t actually know what the effect of any specific medication will be on the personality or moods of the patient which is what they are trying to improve when the patient says for example “I am depressed”. Depression might be considered as a variant on introversion within the Myers-Briggs personality test; a person who answers questions leaning towards introversion might also just be feeling down and depressed, because depressed people do not like spending time around others much of the time and neither do introverted people, according to Jung. Hence psychiatrists are reduced to asking questions on checklists, just as Myers-Briggs management psychologists are; but one hopes that the questions they ask are somewhat more subtle and based upon greater levels of experience.
In the psychiatrist’s case, a person who answers “depressive” questions in a way that indicates depression is more likely to be prescribed anti-depressive medications of which there are various kinds and you tend to experiment with one and then move onto another one if the first (or second) doesn’t work or has unwanted side effects. Depression is not the only thing that psychiatrists try to measure by asking questions. Autism and schizophrenia are two other personality spectra used by psychiatrists to try to establish the existence of disorders that need treatment, and the method by which psychiatrists proceed is really just a complicated version of Myers-Briggs. Indeed one might attempt to transpose the Myers-Briggs personality types onto a grid of contemporary psychiatric personality disorders that are measured in much the same way (i.e. by questionnaires) in order to see the way in which psychiatrists try to assess mental disorders and appropriate treatments for them.
Like people who have undertaken versions of the Myers-Briggs personality test multiple times, patients who are the subject of routine assessments by psychiatrists (e.g. hospital or prison inmates) get used to the sorts of questions being asked and the sorts of inference the psychiatrist is likely to make from one’s answer, and therefore what sort of pill one is likely to be prescribed as a result. Psychiatry ultimately suffers from the same fatal flaw as does Myers-Briggs which is that personality traits do not easily map into scientific boxes (to simplify Davidson’s quote) yet the things about brains that we can actually measure, and that the pharmaceutical industry works by (each psychiatric medication is based around a specific neurological effect in the brain) are scientific and it is an age-old conundrum, as old as philosophy itself, to measure these two things up. In philosophy we call it the mind-body problem, and the underlying paradox that plagues psychologists and philosophers alike is that ascriptions of personality, mood, behaviour, desires and inclinations portray a different sort of logic to that exhibited in conventional science.
At the heart of this disparity is the mysteries of religion and the question of the soul. Because if our minds and our personalities work by such different and unscientific logic compared to our brains and our bodies, then the argument goes that the two things are not the same and therefore we must have a personality or a spirit or a soul that is separate from the body. That would appear to entail theism of some kind, maybe; Immanuel Kant thought so but he was one of the few who ever did. These problems are ultimately virtually insoluble, but they do leave us with some insights and we will leave the reader with two.
Firstly, the fact that Myers-Briggs continues to be used even though it is so comprehensively discredited stems from the urge on the part of psychologists to prove their discipline a science when it is not. Nobody has really improved Myers-Briggs, made it better, added more categories or questions or classifications; it is nonsense, and nonsense cannot be improved. It is quite complicated enough without being added to and being so confounding as to be implausible. Now it is barely plausible; if we augmented it then it would cease to be so. Secondly, because we do not understand the connection between the mind and the body, and we never have, and for all the neurobiology in the world we probably never will, psychiatrists cannot work to rule books or textbooks and their spectra are as flawed as Myers-Briggs. At best they are just an approximation, the beginnings of a sort of guesswork. Then you try a medication and you experiment with the individual patient because we don’t have theories of how these different medications affect the behaviour, feelings and moods of individual patients; each case is distinct. Therefore psychiatrists just try with different medicines and see what works in each case, blundering around in the most abstract and uncertain of the branches of medicine. There is no certainty in any field of psychology, and any attempt to measure things with certainty is doomed to look foolish.
Those are the things that Myers-Briggs teaches us, and it is the epitome of foolishness although all sorts of gurus and experts will continue to use variants on it and it has proven remarkably durable over almost 100 years of use. Artificial intelligence and the internet will make its influence, accessibility and availability increasingly broad, and we will all find ourselves being assessed ever more by these sorts of capricious methods as human assessment is something we try to hand over to the machines. It’s impossible, of course; the Turing test can never be passed; machines can never be made to act like humans; human reasoning and machine reasoning are two fundamentally different things. Kurt Gödel taught us that.
Comments