Autism, assortative mating and the wallflower hypothesis

One fascinating phenomenon I have been investigating for a while now is assortative mating among neurodiverse people. Autism and other neurodiverse conditions often run on both sides of parents of children with ASD, they seem to attract each other like magnets. This phenomenon is in particular very strong in places like Silicon Valley and universities like MIT.

What makes neurodiverse people attracted to each other? Simon Baron-Cohen (Pattern Seekers: 2020) sees three possibilities:

  1. People with similar interests end up in similar places/occupations
  2. People with similar minds feel attracted to each other
  3. Neurodiverse people are usually less socially apt, they find each other after neurotypicals have found each other

I think all of these three possibilities contribute to assortative mating among people with ASD. The first two actually go together, it is clear that two people who are interested in, say engineering (an area where people with ASD are particularly strong) end up working for an engineering company and may be attracted to each other because of their shared interests. The third point is a particularly interesting one, and I will call it the “wallflower hypothesis “. As a former wallflower, myself, who has found his wallflower wife only in his 30s, I can easily see how this works, despite me and my wife being somewhat neurotypical.

The wallflower hypothesis would also account for the fact that children on the spectrum frequently have older fathers. There is the idea around that autism happens due to cumulative bad mutation in older fathers, but this is unlikely to be true. The important thing here is that the fathers were already relatively old by the time they had their first child, who may or may not be on the spectrum. So, it’s really the fact that they find their partners so late in life rather than bad genes, that is correlated with autism.

However, it is not only common interests that connect the partners, but also a host of personality traits termed BAPCO:

These traits expand the characterization of the broader autism phenotype to include a constellation of socially valued traits, termed Broader Autism Phenotype Constellations (BAPCO). The frequency of these traits may have increased due to assortative mating opportunities that occurred alongside social changes in education and occupational opportunities over the last 100 years. ( )

These traits include an often more direct style of communication and blunt honesty. What is interesting to note, are the “occupational opportunities” over the past 100 years in the above quote. It is more likely that this kind of assortative mating has been going on for thousands of years. In my model of personality evolution, autistic people are most likely “hunter” types. Here are the evolutionary personality types with their Myers-Briggs temperaments in parenthesis:

What is true, that in the past decades there have probably been increasing hunter-hunter pairings (women can also be hunter types, e.g. Ada Lovelace and Marie Curie were of this type). Places like Silicon Valley will increase these hunter-hunter pairings. However, the rise of autism cannot be entirely explained by cumulative hunter type genes. Assortative mating has been going for thousands of years and is certainly not restricted to autism. Assortative mating has been found for a host of mental problems, all somehow connected to each other. I have developed a model for hunter-gatherer type risks:

The frequent comorbidity and assortative mating among people with these conditions should provide some empirical basis for this model.

I think there are at least two more reasons for the assortative mating patterns found among neurodiverse, that can be added to Simon Baron-Cohen’s three points:

  1. Hunter-gatherer types are less in-group social and more open towards diversity
  2. Hunter-gatherer types may actually have later puberty due to being more K-selected in life-history theory

Children with hypogonadotropic hypogonadism or delayed puberty are more likely to develop neurodevelopment disorders, such as autism spectrum disorder, ADHD and intellectual disabilities, compared with those who undergo normal puberty ()

I will speculate here that the causation in this study goes the other way round: neurodiverse children are more likely to have delayed puberty. Also, neurodiverse people are frequently of the ectomorph (hunter-gatherer) body type. It has been shown that teenagers with ectomorph body types tend to have the latest onset of puberty (), and mesomorph (pastoralist types) the earliest.

Wallflowers of the world, unite, go forth and multiply!

Originally published at on January 15, 2021.