Unlearning is hard.
Humans are programmable, this is why the State tries to control public narratives - it allows them to impress ideas and views on to most people which they use to uphold their legitimacy and maintain control.
Once learned, de-programming these ideas, any ideas, is very difficult. They become entwined with ego, with one’s identity and entire worldview.
This is especially true of midwits. They cannot stomach the idea that everything they have learned might be wrong.
How could that possibly be?! They aren’t stupid, after all they have a slightly above average IQ! They’re not some lowly labourer doing menial repetitive tasks for minimum wage, they went to that prestigious University and read books by smart people!
They are so invested in what they think they know that they cannot separate the self from the thing. It doesn’t even occur to them that these ideas were implanted on them by other actors with their own incentives, as such it doesn’t occur to them to question why they think what they do.
Only the tail ends of the curve can do this.
The low end because they don’t wrap themselves in ego, they get by on practicality - what works matters more than what some expert says is correct. They don’t emphasise learning so they have much less to unlearn in the first place.
The high end because they’ve overemphasised on learning have understood how their ego can be weaponised against their own self-interest so they expand their circles to collect more data to challenge their own beliefs, and then go with the synthesis which best meets reality.
In essence - the tail ends believe the purpose of a system is what it does in reality. The middle of the curve impart their ego on to systems and wish it to match their understanding of reality and will always seek to explain away the variance.
The tail ends can unlearn and shed their programming. Midwits can not. 
