Falsification

Hi all

I wanted to discuss mental disorders and Karl Popper’s ‘falsification criterion.’ I have looked at various sources on the internet but I cannot find much information on this subject. This is the reason why I am writing here. From what little I have read mental disorders do not appear to be falsifiable. If this is the case then I am asking if anyone could give me examples of why not, be it from books, journals, personal knowledge, etc. Thanks in advance.

(Note: I read through several of the posts on the 20 pages listed here but I did not see this subject discussed in detail. If it has been discussed then please point me to the post and delete this thread to avoid redundancy).

I do know that diagnoses are made according to interpretations that a "qualified" physician makes regarding behaviors. The established (conventional) approach generally involves the idenfication of these behaviors as belonging to a list prepared by the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders. See the link for some of these:

http://www.psychiatry.org/mental-health ... r-disorder

Falsification seems like an interesting question because it could be interpreted as the potential for misdiagnosis, which is easy to make given the nature of any particular doctor's interpretations of the "bipolar" behaviors" listed on that website.

Hey Jaymes,

Karl Popper's major contribution to the philosophy of science was his attempted solution of the demarcation problem, which is the problem of how we distinguish science from non-science. Popper was concerned that many putative scientific theories like Marxist interpretations of history, psychoanalysis or Adler's individual psychology were not fully testable because the observational consequences of a particular test could always be interpreted to be in favor of the theory. For example, if a Freudian believes that you have an unconscious drive to have sex with your Mother and decided to put this to the test by asking you, you can only respond in two general ways: "yes, I have thoughts of having sex with my Mother" and "no, I don't have thoughts of having sex with my Mother". However, Freudians can count either result as being in favor of their theory because it's an unconscious drive which you're not supposed to be aware of.

So Popper's solution was to say that in order for a scientific theory to count as scientific the theory in question must be falsifiable by some observations. To clarify a bit, Popper uses the terms testable and falsifiable interchangeably because for Popper a test of a theory doesn't count as a real test unless there's the possibility of the theory being shown to be false.

As for your question about whether or not the existence of mental disorders is testable and falsifiable, I think that's a tough question. Psychologists and psychiatrists are heavily influenced by a position in the philosophy of science known as operationalism which defines the concepts used in science (i.e. time, weight, intelligence, etcetera) in terms of the operations in which they are measured. So on the operationalist's account, there's no distinction between mental disorders and the way that we measure them. For example, Borderline Personality is when someone meets at least 5 out of 9 of the critiera put forward in the DSM-TR-4. You can't ask "yeah, but what is it really?" because that's all there is to the operationalist story.

Operationalists pretty much do science this way: they come up with a concept, explain the concepts grammar, develop a metric for measuring the concept, validate the metric by looking at what the scores correlate and predict, then build and test theories about the concept.

So I guess, to answer your question, the existence of mental disorders isn't really falsifiable and wouldn't pass Popper's test, but most psychologists and psychiatrists, who aren't just totally unreflective about theoretical questions, see the the existence of mental disorders as more of question about what we mean by the concepts that we use and how we decide to measure them.

Of course, there are lots of different psychologists and psychiatrists who believe that psychological predicates are floating around in our heads and the main problem is that we just haven't found them yet, but I think this is just too confused to take seriously. The operationalists seem to have a much more coherent philosophy of science.

My psychologist tells me that most "mental disorders" are only classified as such to make insurance companies happy. My psychiatrist tells me that there is a medical basis. The two ideas being sent at me at once make my head hurt.

In all honesty though, I do think that most of these disorders are just poorly understood problems. Personality disorders, for instance, seem to be survival skills developed to survive in an inadequate enviroment. I do think they're onto something, I just tend to disagree with their interpretations.

Thank you for the responses. I appreciate the feedback. From what little I have looked into this subject there at least four reasons why mental disorders do not seem to be falsifiable to me. The first is that since it is not standard practice to use medical tests to determine who does, or does not, have a mental disorder then one of the criteria used to arrive at a diagnosis is the self-report of symptoms. Because of this I find it to be very difficult to falsify another’s personal testimony. (Note: I am not implying here that each and every person is lying but rather I am trying to point out that problems may occur when such specific criteria is used. I also recognize that other independent verification can be used, such as observing one’s behavior and statements from family members.)

The second is when I read or hear that mental disorders “will likely be proven” in the future. An example of this can be found in the 2003 APA’s Statement on Diagnosis and Treatment of Mental Disorders

http://en.wikipedia.org/wiki/Biopsychia ... ite_note-2

It seems apparent that such claims can neither be confirmed nor falsified at the present.

The third and fourth is in regards to the following statements from Popper listed on Stephen J. Gould’s website.

http://www.stephenjaygould.org/ctrl/pop ... ation.html

“A theory which is not refutable by any conceivable event is non-scientific…”

I can find of no documented ‘conceivable event’ that would refute the existence of these diagnoses. (Admitting though, perhaps I have not come across such information).

“Every genuine test of a theory is an attempt to falsify it, or to refute it…” (emphasis in the original).

Psychiatry seems to do the opposite here. Instead of seeking evidence to refute the existence of mental disorders they only appear to gather evidence to support it.

What are your thoughts on the examples I have listed? Also, are there other examples that can be included? Thanks in advance.

Hey Jaymes,

Jaymes wrote:[ . . . ] since it is not standard practice to use medical tests to determine who does, or does not, have a mental disorder then one of the criteria used to arrive at a diagnosis is the self-report of symptoms. Because of this I find it to be very difficult to falsify another’s personal testimony. (Note: I am not implying here that each and every person is lying but rather I am trying to point out that problems may occur when such specific criteria is used. I also recognize that other independent verification can be used, such as observing one’s behavior and statements from family members.)

You're right that there are some meta-theoretical assumptions in psychiatry, psychology and behavioural neuroscience about behavioural responses like self report: that they're genuine and accurate. However, once you start questioning the genuineness and accuracy of responses like self report you end up giving up a large part of the subject matter of the behavioural sciences. Of course, this is exactly what behaviourists did, limiting themselves to only studying perceptual inputs and behavioural output, which was more epistemically secure, but it really limits what you can study.

So to answer your question, psychiatrists, psychologists, and behavioural neuroscientists make certain meta-theoretical assumptions when doing research, which are not falsifiable, but they only do so because if they didn't they would end up being stuck in basements somewhere with a bunch of rats documenting stimulus-response relationships (not a whole lot of fun).

Jaymes wrote:The second is when I read or hear that mental disorders “will likely be proven” in the future. An example of this can be found in the 2003 APA’s Statement on Diagnosis and Treatment of Mental Disorders

This is true, that is a promissory claim and to use it to support the validity of the constructs you're saying exist is not accepted by the scientific community as support for their existence.

Jaymes wrote:I can find of no documented ‘conceivable event’ that would refute the existence of these diagnoses. (Admitting though, perhaps I have not come across such information).

Well, I don't know if Popper would apply the falsification criteria to theoretical constructs or medical concepts. If we have a theoretical construct like mesons, quarks or black holes, that can't be observed only be inferred from certain observations, does that mean the theory that makes reference to these constructs is unscientific? I don't think so. I think Popper would say that as long as the theory is testable or falsifiable then it's okay to make reference to theoretical constructs. As for medical concepts like "depression", "anxiety" or "psychosis", I'm not sure how we could even apply the falsification criteria to these. How could we put it to the test to see if "depression" exists?

Jaymes wrote:Psychiatry seems to do the opposite here. Instead of seeking evidence to refute the existence of mental disorders they only appear to gather evidence to support it.

There's confirmation bias in all the empirical sciences and this is why Popper advocated that scientists ought to try and falsify their theories instead of trying to confirm them. The more their theories stand up to falsification the more confident we are in their truth value.

Does that make sense?

Take care.

Falsifiability, Popper's Criterion, is defined at Wikipedia. But psych is baloney to its very core and so defies the applying of meaningful attributes. What behaviorist can say you have a disorder if your conduct has never even resulted in any convictions, and certainly none for the offense "disorderly conduct". Yeah I guess I'm just not one of "The Beautiful People"; my shrink is one so he can distinguish those who are from those who are not.

If you honeymooned at Niagara Falls you're not alone! Later in your marriage you might wish your love life were that hot all over again and you'd beseech, "NIAGARA O ROAR AGAIN", a wish that is spelled the same, backwards.

The initial stage of the diagnosis of mental illness involves a discussion with one or more psychiatrists. Based on this discussion you may be judged to have a chemical imbalance in your brain which they, presumably, believe needs to be treated with antipsychotic medication.

If you are hospitalised the idea is that you are observed on the ward. At the end of each day the nurses write a report on you in your notes. In practise though, my experience is that there are only two qualified nurses on a non ICU ward, spending most of the day in the office doing administration or taking care of emergencies on the ward. This means they have little time to do actual observation, meaning that the accuracy of their reports is unlikely to be of a particularly high quality.

Given that there are no "scientific" tests involved in the diagnosis, such as a blood test or brain scan, I don't know how you would falsify the existence of the illness, as all you would have is your word against their word.

motionissue wrote:My psychologist tells me that most "mental disorders" are only classified as such to make insurance companies happy. My psychiatrist tells me that there is a medical basis. The two ideas being sent at me at once make my head hurt.

Why so? They don't contradict. There definitely is a medical (physical) basis for mental disorders else medications wouldn't empirically prove useful. The questions is just how mental disorders can be described on a neuronal level and that is a difficult and complex question and open subject to more research. Still doesn't mean it's "wrong" to describe mental disorders on more "soft" criteria that are perfectly observable and thus falsifiable. If these definitions are always spot-on accurate is another question to be discussed and as you can see on the changes on the new DSM V there is a lot of discussion going on about such specifics.

Furthermore you don't need to know HOW exactly something works (if possible to know at all) if you can prove THAT it works. Popper was concerned about theories that are not falsifiable because they aren't empirically measurable and thus can't be object of scientific research (see metaphysics, esoteric, religion).

Still Popper was and is quite irrelevant to science (as is almost all theory of science). At a certain point you have to stop trying to falsify theories and assume the validity until proven otherwise to make any progress. On a pragmatic basis (which everyone has to act on) Popper's other demands are absurd.

motionissue wrote:My psychologist tells me that most "mental disorders" are only classified as such to make insurance companies happy. My psychiatrist tells me that there is a medical basis. The two ideas being sent at me at once make my head hurt.

I can see where you might be confused - so would I. It would seem odd that a doctor would suggest that the insurance company would be a factor when classifying medical diagnoses. I was under the impression that insurnce companies would not be necessary to motivate the medical community to classify disorders. Wouldn't doctors want to classify regardless of insurance companies?

motionissue wrote:In all honesty though, I do think that most of these disorders are just poorly understood problems.

I would think the same because there is so much learned by researchers every year. If researchers knew what is going on, then it wouldn't be called research.