Multiple studies point consistently in the same direction: a bedroom temperature between 20 and 25 degrees Celsius and low CO2 levels through good ventilation improve both sleep depth and sleep efficiency. The evidence comes primarily from smaller field experiments and observational studies, not large randomised trials. For older adults, the effects of heat are greatest and the range for good sleep is narrowest. A practical guideline for most people: cool but not cold, and make sure fresh air can enter.
A cool bedroom does indeed help you sleep better, but the relationship is more nuanced than 'colder is always better'. Multiple studies point consistently in the same direction: a temperature between 20 and 25 degrees Celsius is the most favourable range for most people, and certainly for older adults. An observational study in Japan1 showed that people who felt cold in the bedroom actually slept worse, with an average difference of 2.25 points on a validated sleep questionnaire. So too cold is not good either. On the other hand, a study using wearable sleep trackers in real homes2 showed that sleep efficiency measurably declines once the room temperature rises above 25 degrees.
How seriously heat disrupts sleep was demonstrated in a small-scale experiment involving sixteen older adults3: at 30 degrees versus 27 degrees, they slept an average of 26 minutes less, were awake 27 minutes longer, and lost nearly five minutes of REM sleep. Their nervous systems were also measurably overactive. These are objective measurements via EEG, not self-reported data. The sample size is small, but the outcomes are consistent with the other findings. For older adults, the temperature range within which sleep remains good appears to be narrower than for younger people.
Darkness of the bedroom is not explicitly addressed in the studies provided, but ventilation and air quality turn out to be at least as important as temperature. Two field experiments with students4 showed that lower CO2 levels in the bedroom, achieved by opening a window or a ventilation grille, led to measurably better sleep and better concentration and logical thinking the following morning. A blinded experiment in 29 bedrooms5 confirmed that better ventilation leads to more deep sleep and fewer awakenings, although that study found no effect on cognitive performance. In the previously mentioned study of older adults3, the effects of better ventilation and a more comfortable temperature were additive: together they produced 10 minutes more deep sleep and nearly 4 minutes more REM sleep.
An observational study of 62 participants followed over fourteen nights6 measured multiple factors simultaneously. Higher temperature, more fine particulate matter, higher CO2, and more noise were each independently associated with lower sleep efficiency, with effects of between 3 and 5 percentage points per factor. Notably, participants seemed to subjectively adapt to worse conditions, while their objectively measured sleep nonetheless deteriorated. Humidity had no significant effect in this study. A small study of 17 volunteers7 found that an open window or door improved sleep depth and sleep stage, but not all measured sleep parameters improved.
In practical terms, this means the following. Aim for a bedroom temperature of roughly 20 to 25 degrees; for older adults it is especially worthwhile to prevent the room from exceeding 25 degrees. Also ensure ventilation: a window left slightly ajar or a well-functioning ventilation system lowers CO2 concentration and demonstrably improves sleep quality. Temperature and ventilation reinforce each other. There are large individual differences, so someone who feels cold in the bedroom will also sleep worse. The sweet spot is a well-ventilated, slightly cool room in which nobody feels cold.
Nine studies used: multiple field experiments and observational studies, including two experimental studies with EEG measurement in older adults (n=16), a study with 14 nights of environmental monitoring (n=62), and multiple ventilation experiments. No large randomised trials or meta-analyses are available in the sources provided. Darkness of the bedroom is not part of the claims provided.