But that's pretty much exactly what the OP mentioned. An academic was intrigued by the idea that some texts mentioned the idea of "second sleep" in passing and devised a theory of everybody's sleep every night being neatly broken into two halves with an hour of wakefulness in the middle because people go to bed up early and get up late in the absence of artificial light.

But all the evidence from cultures which have little or no access to artificial light even today contradicts the idea of that being a natural, universal response to lack of daylight, and it arguably makes even less sense as a universal approach to daylight patterns in the relatively northern UK, where darkness ranges from 6 hours in the summer to 16 in the winter. The textual evidence is congruent with sleeping patterns being not much different to the modern day, where people also often wake in the middle of the night and go back to sleep (sometimes even getting up to make a drink and twiddle their phone in between) but just don't refer to them as numbered "sleeps" very often