(Original Review, 1980-09-12)
The classic tests on this subject were done
something like 25 years ago, and became sufficiently well known that by the
late 60's they were turning up (in cut-down versions with animal subjects) in
high school science fairs. The test was fairly simple: a sensor was rigged to
detect Rapid Eye Movements (characteristic of the deepest (fourth) level of
sleep and generally simultaneous with dreams) and attached to various devices
to wake up the sleeper whenever REM began. Other than this, the subjects were
allowed to have whatever they considered a normal sleep period. Within a few
days the number of attempted REM periods went from ~4 to >20 and the
subjects were reported as becoming extremely irritable.
John Brunner wrote a story around this idea in
one of the SPECTRUM anthologies (about what happened when someone didn't react
with irritation and what happened when they stopped preventing him from dreaming)
and has brought up the idea intermittently since then, particularly in THE
STONE THAT NEVER CAME DOWN. Unfortunately, no one at the time thought to do the
obvious control of waking people at completely random intervals, beginning with
~4/night and working up to >20 per night; when this was done the behavioural
results (to the extent that behavior can be quantified and compared) were
reported as being basically the same as in the REM experiments. At about this time, however, it had been
found that for many people 2 3-hour periods of sleep separated by a short
period of activity were as effective as one 8-hour period. It seems that
regularity is as important a factor as duration in measuring the value of
sleep. The one person I know who consistently gets by on ~5 hours seems to take
basically the same 5 hours (0300-0800) every day, both at slack periods and
during SF conventions. (He may be helped
by the fact that he's basically a very tranquil type, though there is Shapiro's
example of the frenetic type who survives on 4 hours. There are a number of cases of non-circadian
patterns being adopted for various reasons; I have only apocryphal information
on the person who allegedly converted his week to 6 28-hour days (his sleep
period cycled completely around the day every week) but Frederik Pohl in his
autobiography describes deciding that he just didn't have enough time for
writing and going on a 48-hour "day", thus getting both time to deal
with mundanes during their active periods and quiet periods in which to write
undisturbed. There are also some developments which I've just seen the first glimmerings
of; it seems that researchers have pinned down two nuclei in the hypothalamus
as being potentially responsible for circadian rhythms (specific hormone
release cycles corresponding to activity cycles have been found). I've gone
through so many magazines in the post-Noreascon catchup that I've no certainty
of where I found this or who was involved; has anyone seen any more detailed
information?
[2018 EDIT: This review was written at the time
as I was running my own personal BBS server. Much of the language of this and
other reviews written in 1980 reflect a very particular kind of language: what
I call now in retrospect a “BBS language”.]
