Interesting research topic by the way. I will take a look at your paper.
Dreams don't serve any function at all, they are simply visualisations of brain cells (entire networks) firing in different regions, some of which activate the active processing of touch, smell, sound, taste, vision, and whatever the hell all those 'gut feelings' are, intuition, premonition, even deja vu and the feeling that comes with insights in retrospect.
Your brain winds down, disconnects from the party chat, closes the HUD, turns off most sensors, changes the voltages in brain regions and throughout the entire CNS to do all the shit on the running to do list of the day, week, month, year ( think old injuries or unsolved problems ) that it couldn't do because "it had it's hands full" balancing you on two feet and making sure you move properly through space-time and abide by the law of permanence.
Cells which are wired together, fire together and vice versa.
Why is that surprising? The body is sleeping, your senses are mostly off, reaction times are irrelevant. The brain is not necessarily super relaxed.
I remember that I had to actively react in lucid dreams because my instinctive reactions were slow as fuck.
During REM sleep, surprisingly the amygdala is inactive and norepinephrine is 85% lower than base waking levels (not high anxious levels). So the brain is in a super relaxed state!
I wrote a paper on the implications for dream content and interpretation (I’m a psychotherapist in training).
If you’re interested you can find the paper here:
https://osf.io/preprints/psyarxiv/k6trz
And it was discussed on HN here:
https://news.ycombinator.com/item?id=19143590