It is hard to know what to make of this... There are a lot of references to studies... But (sigh) I'm not much of a 'reading studied' kind of person.

We had this reading group the other week on the genetic basis of schizophrenia. There have been lots of studies on that one and popular headlines proclaim 'scientists discover the genetic basis of schizophrenia'. Except of course, they haven't.

Schizophrenia isn't to do with a single gene. If it was to do with a single gene then that would be obvious from looking at the pattern of affected people within a family. If you look at a family tree and mark the affected people then you should notice patterns that tell you that the condition is single gene dominant or single gene recessive etc etc. Basically... Those patterns aren't there, so we know that schizophrenia isn't to do with a single gene.

Geneticists use the term 'gene' in an ambiguous way to refer to two distinct things. Sometimes a 'gene' means a certain location on the chromosome (basically a string of t's and a's, and g's, and c's that are physically lined up next to one another and that code for - make - a single protein). We know schizophrenia isn't like that because of the family tree data.

The other use of the term 'gene' refers to different locations on the chromosome but the unity is in their united effort in coding for the protein. Sometimes physical genes (that are seperate in space) can act as a functional unit (in that they all contribte to making the protein). Researchers think that schizophrenia will be due to some complicated combination of physically located genes. So the million dollar question is 'which physically located genes?'

There have been many many many many many different candidates. The problem is lack of replicability. One study suggests one physical gene might be important... Another study finds that that physical gene isn't statistically significant in schizophrenia at all. Another study suggests another physical gene might be important... Another study finds that the physical gene isn't statistically significant in schizophrenia at all. And so it goes on...

People have similar 'success' in looking at the genetic basis of 'voting behaviour' or 'religion' or whatever. We read a lit review type paper which looked at all the candidates that had been shown to be implicated in MORE THAN ONE STUDY (because there are lots of papers that are lacking additional support). Still... Two studies isn't a very stringent criteria.

The statistical significance is low...

I wonder... What do we need to show us that schizophrenia ISN'T a genetic disorder? Do we just keep looking and eventually conclude that failure to find a robust genetic basis shows us that there isn't one? That isn't a very good inference... What would we need to find to rule that out? How do we prove that that is NOT the case? Even if we find social causal mechanisms... They aren't mutually exclusive...

I do think people are on a wild goose chase with this...

And with neurology too... I'm concerned about the reasoning processes used in psychiatry... I think... There is reason for concern. I never wanted to be such a sceptic... But I don't see how in good conscious I'm expected not to be...