The Degenerative Agenda of Explaining Away Genetic Research in Schizophrenia

Jay Joseph's guest post on Duncan Double's blog – criticizing the notion that schizophrenia has a substantial genetic component – reminds me of the distinction between progressive and degenerative scientific programs from philosophy of science. Progressive scientific programs show a growth of theoretical frameworks, development of new research methods, generation of novel findings, and validation of predictions. Degenerative programs lack these features; instead, they are engaged in the never-ending task of explaining away findings.

I marvel at the vacuousness of the efforts to deny the overwhelming scientific consensus that a substantial portion of the risk of development of schizophrenia comes from aggregate genetic factors, and that these genetic factors don’t confer this risk in isolation but rather through complex interactions with non-genetic and environmental factors.

In order to appreciate the progressive nature of genetic research into the nature of schizophrenia, consider this wonderful review paper from 2017: Genetics of Schizophrenia: Overview of Methods, Findings and Limitations by Henriksen et al.

Consider the progression of genetic research in schizophrenia:

Early 20th century: Family studies demonstrate that rates of schizophrenia are higher in relatives of patients with schizophrenia than in the general population

Mid & Late 20th century: Twin studies report that the concordance rate (i.e., both twins suffering from schizophrenia) was elevated in monozygotic (MZ) twins compared to dizygotic (DZ) twins

Mid & Late 20th century: Adoption studies report that schizophrenia spectrum disorders are more frequent in adopted-away children of mothers with schizophrenia than in their control adoptees

Late 20th century & Early 21st century: Linkage analysis studies suggest several chromosomal regions may contain schizophrenia susceptibility loci.

Early 21st century: Case control studies suggest the involvement of several candidate genes.

Early 21st century: Genome-wide association studies identify a large number of new susceptibility gene loci of very small individual effects; many of these genomic loci are replicated in subsequent GWAS and reach meta-analytic genome-wide significance. These GWAS utilize extremely stringent thresholds of significant p-values (p < 5 × 10⁻⁸) to limit false positives.

Each wave of genetic research has its own set of limitations; some findings don't survive replications; many others do. The theoretical frameworks and research methods generate more and more consistent empirical findings. A century of pre-molecular and molecular genetic research consistently demonstrates that aggregate genetics constitute a strong risk factor for schizophrenia.

Dr Joseph would like us to believe that this entire century’s worth of research is false. That all these researchers from across the world separated by time and space collected similarly flawed data and arrived at the same false conclusions. Dr Joseph primarily focuses on the limitations of the methodologies of twin studies and adoption studies. The limitations are recognized by many psychiatric geneticists (and it is true that these limitations need to be recognized more). The Henriksen et al. review has a whole section on it. Dr Joseph thinks these limitations invalidate the entire body of research. They don't.

Consider the weight of what Dr Joseph is asking us to explain away. The 2003 meta-analysis of schizophrenia twin studies includes 12 studies. 2 from USA, 2 from England, 2 from Norway, 2 from Denmark, 2 from Finland, 1 from Sweden, 1 from Germany. The meta-analytic summary for additive genetic variance in liability to schizophrenia was estimated at 81% (95% CI 73-90%).

Now, it is true that in order to take the higher concordance rates in MZ than in DZ twins as evidence for a genetic component, one of the problematic assumptions is that the environments experienced by MZ and DZ twins do not differ in any way that may be relevant for the development of schizophrenia. Since the environment is not going to be exactly the same, the genetic estimate is likely to be off (it’s an assumption that allows us to get some approximate, ballpark answer), but it also doesn’t mean that a genetic component doesn’t exist at all.

If genetics doesn’t explain that MZ twins have a higher concordance of schizophrenia than DZ twins, then what does? Dr Joseph says

i) MZ twins have more similar environments than DZ twins

ii) MZ twins have higher levels of identity confusion and attachment to each other

iii) MZ twins have greater tendency to experience “folie à deux” (shared psychotic disorder)

On what strength of evidence are these claims being made and where is the evidence which shows that these factors offer a more rigorous explanation than genetic factors? How do these factors explain the elevated rates of schizophrenia in other biologically related relatives, like parents, aunts, first-cousins? Do aunts and nephews also have a greater tendency to experience “folie à deux”? I’m not sure if the explanatory gap is being appreciated here. Family histories are not easy to explain away.  

This is the classic asymmetry of applied skepticism. Genetic explanations are subject to the highest level of scientific scrutiny, while psychological explanations are being offered with the flimsiest of evidence and more commonly mere speculation.

Dr Joseph is also forgetting that environmental factors also include profoundly biological processes such as exposure to infectious agents, nutritional deficiencies, environmental toxins, and complications during pregnancy (intrauterine factors). How are these environmental factors being taken into account? How do they fit into the pretty little picture of ‘identity confusion and attachment’ as the cause of schizophrenia?

Look at the dismissal of the results of GWAS: “The most recent claims of “multiple genes of small effect” are based on associations, not causes, and like previous claims it is extremely unlikely that they will hold up. Although we have seen many reports that various genetic variants are “associated” (correlated) with schizophrenia, decades of molecular genetic studies have failed to produce variants shown to cause it.”

First, many GWAS associations have already been replicated in multiple studies (although definitely, more replications are needed). Second, genetic associations in fact have a much higher likelihood of being causal because they are more proximal in the development of the individual: your psychology cannot influence your genetic code, but your genetic code can (at the very least theoretically) influence your psychology.

If twins studies and GWAS don’t live up to Dr Joseph’s uber-rigorous scientific standards, why doesn’t he suggest how to do better research? What research design does he have to offer us that will more accurately distinguish between contributions of genetic factors, biological environment factors, and psychosocial environmental factors?

As a final point, the consensus position in genetic research of schizophrenia is not a reductionistic one. The conclusion is not that schizophrenia is purely a genetic disorder. In fact, the consensus is that schizophrenia is a complex trait that results from interaction of genetic and environmental influences. This is a pluralistic account that accepts the essential and important role of factors such as psychological trauma and even social inequalities. For all this rhetoric of reductionism directed at a pluralistic position, one has to wonder who is really being reductionistic here?