Life is short, and we spend a lot of our time sleeping, which is why the idea of hypnopedia, or the ability to learn during sleep, has been tantalising scientists for so long.

Now, a new study adds more evidence against the whole idea.

According to the new research, while the brain is capable of hearing different sounds during sleep, the ability to accurately group these sounds into a sequence completely disappears when we doze off.

The report suggests that the human brain does not function in the same way when we are sleeping, potentially putting a biological limit on the sophistication of lessons that we can learn while we snooze.

Ever since the 1950s, when the first studies of sleep-learning were published, hypnopedia has gone in and out of favour in the scientific community.

At first, it seemed like a great way to extend learning time, by teaching through the subconscious brain into the wee hours of the morning.

For a while there, it was popular practice to fall asleep listening to audio tapes that promised to help you quit smoking, memorise Spanish vocabulary, master martial arts, or even boost your material wealth.

But then, evidence debunking the outlandish claims of hypnopedia began to pile up, and scientists soon had to admit that sleepers cannot wake up with brains that are somehow able to recall entirely new facts and figures.

Recently, however, the idea of hypnopedia has begun to resurface, although the claims are a tad less far-fetched now than they were just a few decades ago. 

Several studies in the past few years have suggested hypnopedia isn't quite as baseless as we once thought.

In 2014, a team of Israeli neuroscientists found that they could train sleepers to make subconscious associations between cigarette smoke and foul odours.

Another more recent study, found it is possible to teach acoustic lessons to people while they sleep.

While this research did not support the idea of learning dictionary definitions in our sleep, it did find evidence to suggest that the human brain can undergo pattern learning, even when sleeping.

The new study puts a further damper on these ideas, arguing that it is not clear if sleep allows for more sophisticated forms of learning.

The researchers used magnetoencephalography (MEG) to measure the brain activity of 26 participants during wakefulness and during slow wave sleep (aka non-rapid eye movement sleep, or NREM), which is a part of sleep when brain activity is highly synchronized.

On both occasions, participants were exposed to a series of sounds, either randomly organized or structured in a way that the sounds could be grouped into sets of three.

The findings reveal a key difference between the way a brain functions during sleep and the way it functions during wakefulness.

While the MEG brain responses showed evidence that patients could hear individual sounds during sleep, there was no response observed in the patients that could be attributed to sound grouping.

When awake, however, all participants had MEG responses that reflected the grouping of sounds into three elements.

This is a small study that needs to be replicated. But examining the results, the authors suggest that the reason we are incapable of higher learning during sleep is because the high-order brain structures are deactivated during slow wave sleep.

This would account for why we are still able to detect sound while sleeping, even though we aren't able to analyse that sound into statistical regularities.

So even though some elementary forms of hypnopedia appear to be possible, it remains unclear whether we can learn foreign vocabulary or other complex topics after we hit the sack.

This study has been published in Scientific Reports.