Yeah isn't this exactly what one would expect and similar to things like bmi?

I see. Well yes but it might be tricky because inherited variants would tend to have smaller effect sizes than de novos independently of assortment.

I'm not sure why assortment would affect this?

When there is assortment, common and rare variant contributions will become correlated and this needs to be taken into account when analyzing either.

This paper provides the first direct evidence of this by showing a surprisingly large correlation between the common variant education polygenic score and a rare variant burden score.

It's possible that this is because the parents' genetics affect the offspring through the env but it could be assortment: parents with lower education polygenic scores based on common variants are more likely to mate with individuals with rare variants that cause neurodevelopmental conditions.