5

What is the biological origin of the following behavior:

Sometimes, during the early, formative phase of sleep (when you have a gradually dimming consciousness of what all is happening around), you feel some kind of a tremor of sorts - a feeling as if you are about to fall off. The body responds by twitching the limbs, so as to "balance", and you end up waking up for a second or two, where you realize that you aren't about to fall and all is fine.

Since Cerebellum is involved with maintaining equilibrium and balance, I have the intuition that this is related to some transient behavior where the Cerebellum is dimming down. When you are in deep sleep, you do not have a sense of balance, so I guess Cerebellum does dim down the sense of equilibrium.

But the apparent flaw in my reasoning is - if this was indeed the origin, one would have experienced these tremors every time one slept. That doesn't happen, it is only occasional (AFAIK).

So, biologically, what is the true origin of this behavior?

299792458
  • 293
  • 1
  • 5
  • 19

1 Answers1

7

These are called hypnic jerks, also known as hypnagogic jerks or sleep starts. They are normal part of the wake-to-sleep transition. Hypic jerks consist of non-periodic myoclonic movements, generally involving an isolated limb.

The exact physiologic origin of hypic jerks is unknown. We can say that they correlate with a particular waveform on the EEG known as a K-complex:

enter image description here

Picture from http://en.wikipedia.org/wiki/K-complex

K-complexes are a normal part of Stage II sleep and can evoked using auditory stimuli, even if sleep is not interrupted. In the case of hypnic jerks, the K-complex is usually followed by alpha (wake) activity. To my knowledge, there is no evidence that hypnic jerks originate in or are routed through the cerebellum. K-complexes themselves are related to cross-talk between the cortex and the thalamus, a circuit that doesn’t involve the cerebellum.

In addition to the EEG findings, it has also been shown that the jerks are accompanied by autonomic activation — transiently increased heart rate and respiratory rate. There is also an (unexplained) subjective aspect. You described it as feeling "as if you are about to fall off,” which seems to be consistent with the published literature on this. As mentioned in the review cited below, some people report a sensation of "falling into the void.”1

The clinical implications, in case you were wondering, are summarized well by Vertugno et al.:

Sleep starts are common physiological phenomenon affecting up to 70% of the adult population and their course is benign, resolving without any neurological sequel. Therefore, reassurance and counselling are all that is needed to treat the patient. Though an adequate explanation and reassurance may be sufficient, some patients may require a small dose of clonazepam (0.5–1 mg at bedtime) to ameliorate the symptoms on a short-term basis.

1. Vetrugno R, Montagna P. Sleep-to-wake transition movement disorders. Sleep Med. 2011 Dec;12 Suppl 2:S11-6.

You requested in a comment a reference to introductory material on the topic. What we know about the neuro-physiology of sleep is mostly based on the analysis of EEG (brain-wave) tracings. I therefore refer you to this introduction to sleep polysomnography if you would like to get a better handle on how scientists think about sleep.

Susan
  • 3,149
  • 17
  • 26
  • Thanks for your response. I have +1'd it and will go through the linked articles a bit later (a bit short on time right now), and get back to you if I need any help in understanding them. I'm deferring the accept for until then, but I'm sure this is only temporary. Thanks again :) – 299792458 Apr 25 '15 at 13:48
  • @TheDarkSide Unfortunately, the Vetrugno paper requires access through a library (or paywall) for viewing the full text. It was the only review I found that includes significant discussion of the topic in a relatively recent issue of a top-tier sleep journal. If you're not able to access it, let me know and I can see about finding something else. The PDF of the polysomnography chapter should be viewable in full, though. – Susan Apr 25 '15 at 14:00
  • I do not think sleep starts are associated with sleep starts. K-complexes are a characteristic of sleep (Stage 2), although they can be seen is deeper stage 3 and 4 sleep as well. Sleep starts tend to occur during the wake-sleep transition. Sleep starts can also be "Hypnagogic hallucinations" which can vary to sensations of falling (common) to loud noises leading to the infamous "exploding head syndrome". Sound is not the defining characteristic of Hypnagogic hallucinations. ;) See https://www.sleepassociation.org/patients-general-public/exploding-head-syndrome/ – Panther Apr 25 '15 at 14:08
  • Sleep starts can be exacerbated by caffeine and "stress" . See also - http://www.sleepeducation.com/news/2013/08/22/sleep-starts – Panther Apr 25 '15 at 14:09
  • @bodhi.zazen Re. your first comment - you’re confusing a lot of things here. 1) Yes, K-complexes are a defining characteristic of Stage 2 sleep, but they are also frequently associated with brief arousals (alpha frequency), i.e. sleep-wake transitions. 2) The literature about auditory-evoked K-complexes within sleep is extensive. 3) Hypnagogic hallucinations are REM-sleep phenomena, most often seen in narcolepsy, sometimes in sleep deprived young healthy people. Totally different deal. 4) Exploding head is different still, a type of HH but likely different mechanism. – Susan Apr 25 '15 at 14:22
  • @bodhi.zazen Re your second comment - agreed. The review cited above mentioned triggers: fatigue, stress, sleep deprivation, vigorous exercise, and stimulants like caffeine and nicotine. – Susan Apr 25 '15 at 14:23
  • Most K-complexs are not associated with arousals and not all alpha activity is awake. An awake rhythm is low amplitude mixed frequencies and the "alpha rhythm" is a relaxed rhythm with eyes closed. The alpha frequency seen with K-complexes has a completely different distribution as K complexes are over the central head regions and thus completely different from the posterior dominant alpha rhythm. – Panther Apr 25 '15 at 14:32
  • K-complexes can be evoked, but they are not a defining characteristic of sleep starts and it is not clear if the K-complex is the cause or the result of the sleep start, ie, just as they can be evoked by auditory stimuli, they can also be evoked by internal stimuli or the arousal of sleep starts. – Panther Apr 25 '15 at 14:35
  • Hypnagogic hallucinations are not always associated with Narcolepsy and the hypnagogic hallucinations seen in Narcolepsy are not really Hypnagogic hallucinations, they are REM sleep intrusions into either wake or light sleep and are detailed visual "dream" imagery. See http://www.sleepeducation.com/sleep-disorders-by-category/parasomnias/sleep-hallucinations/overview-facts . The main difference with Narcolepsy, other the the type of hallucination, is that the HH occur when Narcoleptic patients are AWAKE where in "normal" people they occur when transitioning into sleep. – Panther Apr 25 '15 at 14:40
  • Hypnagogic hallucinations is thus a non-specific term. In Narcoleptic patients they are intrusions of REM sleep physiology into wake or light sleep . In "normal people" they occur in light sleep, during the transition to sleep, and not associated with REM sleep or REM sleep physiology at all. There no evidence to support or refute that exploding head syndrome is or is not different from sleep starts. And I am not confused, I happen to know quite a bit about sleep ;) – Panther Apr 25 '15 at 15:02
  • Thanks very much for these inputs. I am accepting your answer since you pointed out a nice 'correlation'. However, I feel that the graph is not representing a correlation, but rather a direct response to the (unknown) source. That kick in the graph is the EEG equivalent of the kick we feel physically (rather an EEG representation of the same). That sentence "The exact physiologic origin of hypic jerks is unknown" seems overwhelming, but if that correctly represents the current understanding, there isn't any more I can expect from the answer. Many thanks again :) – 299792458 May 02 '15 at 11:19