When I dream it seems that the (subjective) time is slower than the objectively passed time as seen on my clock radio. Is this because neurones fire collectively at a faster rate, so you can "put an hour of dream time in a minute of objective time"? If they fire at a faster rate, then this does not mean that you perceive the events in your dream as going faster, as you adapt to your experience of time in the real world.

  • 2
    $\begingroup$ Possible duplicate of How long do dreams last? $\endgroup$ Commented Feb 8, 2017 at 6:32
  • 1
    $\begingroup$ @RobinKramer I don't fully agree; there's certainly a lot of overlap, but this particular question targets firing rates. $\endgroup$
    – AliceD
    Commented Feb 8, 2017 at 12:40
  • $\begingroup$ I stand corrected. I'll retract my vote :) $\endgroup$ Commented Feb 8, 2017 at 15:05
  • 2
    $\begingroup$ It's not a duplicate, but it does answer the question: Dream sequences may play faster (or slower) than real time, but people nonetheless perceive time accurately (when measured via sleep mentation reports). It's a lot like playing a video in fast-forward - the sequence plays faster, but the actual frame-rate is not changed (frames are just skipped). So there is no reason to suggest that neurons "fire faster". Firing rates are limited by laws of physics - if they could fire faster, then it would be more useful when awake. $\endgroup$
    – Arnon Weinberg
    Commented Feb 8, 2017 at 20:58

2 Answers 2


In REM sleep, the EEG is remarkably similar to that of the awake state (Purves et al., 2001). Although the EEG represents the synchronized activity of many neurons in the cortex, it does give us a clue whether they are firing faster or not.

Wakefulness is mainly dominated by beta and gamma waves (source: Scholarpedia), i.e. 12 - 100 Hz. REM sleep is characterized by low-amplitude mixed-frequency brain waves, quite similar to those experienced during the waking state - theta waves, alpha waves and even the high frequency beta waves more typical of high-level active concentration and thinking, i.e. 4-30 Hz (table 1) (source: Sleep).

Table 1. EEG bands. source: Neurosky

So if anything, I would say REM sleep, being largely devoid of the gamma band, is associated with slightly lower-frequency oscillations and hence lower neural activity overall.

- Purves et al. (eds). Neuroscience. 2nd ed. Sunderland (MA): Sinauer Associates (2001)

  • $\begingroup$ @AliceD-Though very informative, I am not able to figure out from the data in REM sleep (without the gamma band) at which rate individual neurones fire. For example in the visual system, where neurones fire in sync to produce conscious images in your dreams, how can you extract from these data that the neurones there don't fire at a higher rate? It would IMHO explain why dreams appear to have e greater timespan than the time that has passed on my clock radio because in that case, my dream shows me more visual information in the same time. $\endgroup$ Commented Feb 8, 2017 at 14:35
  • $\begingroup$ It's not a definitive answer, as EEG is a compound potential, as written in the first paragraph. $\endgroup$
    – AliceD
    Commented Feb 8, 2017 at 14:53

Is this because neurons fire collectively at a faster rate, so you can "put an hour of dream time in a minute of objective time"?

I believe it has more to do with what information becomes available to the association cortex and ncc (or consciousness) in the waking/sleep brain.

In the waking brain the neocortex is connected to the environment, so things tend to progress smoothly and for the most part we have contextual references to measure the passing of time,in sleep there is a varying degree of a disconnect both internal and external culminating in REM sleep, where thalamocortical connections (a switchboard/router for the senses) are at it's lowest, so cut off from the environment an cues, the thalamus- neocortex interplay stimulates only neocortical areas ( presumably those recently active through interplay with the hippocampus ) which in turn activate related areas through plasticity. So you could be in one scene where the only measurement of time is internal followed by another one without, the passage of time thus becomes hard to gauge. 1

The firing rate specifically as it pertains to cycles I believe is a wrong measure to focus on, for one, even in the waking gamma ( 30 >htz ) we top at something like 24 fps comprehending a scene and it takes about 200ms to bind information2,there is little to no information ( although if I am wrong please point it out), that we increase our comprehension and perception rates as information presents itself faster, rather we drop frames, and so even at a faster rate the information available to consciousness in the sleeping or waking brain seems to be capped.

Notes, reference & sources:

1. An excellent and readable overview of dream and it's data components is : The Secret World of Sleep (Lewis).

2. Rhythms of the brain (buzsaki) deals in depth with EEG measurements as they relate to information exchange in the brain. In search of consciousness (Koch)looks at the binding problem and ncc in detail.

  • $\begingroup$ The minimum number of frames per second to see a fluid motion while awake is 24 (or 18?).This doesn't mean that in dreams the number of frames per second should also be 24. Maybe it's 100, who knows? In that case, we can put 4 seconds of fluid motion in one second. The forming of a dream image isn't connected to the outside world, so you can form a fluid motion that seems to last for four seconds (a carrousel making one round per second can be achieved in 0,25 seconds in the visual system) $\endgroup$ Commented Feb 10, 2017 at 0:50
  • $\begingroup$ I hope we can agree, at least partially ,Vision Science (Palmer) the chapter on motion perception details how movement is perceived. The 40 Hertz conjecture (Crick , Koch) provides a theory for temporal binding.I agree we don't necessarily know all the details, but a five fold increase in the processing speed of motion perception would entail a change in either EEG signals or a change in neural behavior, so I am more likely to side with a network/systems data explanation for which there is more evidence, see Networks of the Brain (Sporns) for instance. $\endgroup$
    – Keno
    Commented Feb 10, 2017 at 2:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.