The “No Function” Theory of Dreams

It’s entirely possible that there is no adaptive function to dreaming: they may be meaningful, and we may have devised ways of using them, but in terms of conferring an evolutionary adaptation of some kind, maybe they just don’t do anything for us.

This theory contends that dreaming is a functionless by-product of consciousness, and dreams exist because the higher order parts of the brain are still active during sleep when brainstem activity stimulates memories, and the higher parts of the brain smooth this memory-activation into a narrative (Flanagan, 2000). It suggests that sleep has functions, like memory consolidation, but that “there is no reason to believe that these jobs require mentation of any sort” (p.117).

This is an argument that is fairly widely made in oneirology. This argument claims that dreaming is ‘epiphenomenal’ – it is a by-product. This argument says that while dreams don’t confer any kind of evolutionary advantage, neither are they detrimental, so they haven’t been selected out. So while sleep is adaptive and functional, dreams are not – they’re like the froth on a pint of beer, they’re just there because we have sleep but they don’t do anything themselves. Sleep has effects on us, but this is nothing to do with dreams; dreams have no causal effect of their own.

This debate necessarily takes us into the fascinating but complex philosophy of Cartesian dualism and the mind/body problem. Put very simply, the original problem laid out by Descartes was to query how the stuff of the body, or physical matter (res extensa) and the stuff of the mind, or non-physical matter (res cogitans) could interact, given that one is physical and one is non-physical. I’m no expert on this subject, but when discussing whether dreams are isomorphic with sleep processes, I find it helpful to appeal to John Searle’s (1992) “Biological Naturalism” response to the mind-body problem. He suggested that mental processes and physical processes are isomorphic, with the former having first-person, subjective ontology and the latter having third-person, objective ontology. They are two different manifestations of one system. Dreaming is causally reducible to neurobiology, but not ontologically reducible: this means that dreaming is caused by objective features of the brain, but these alone do not describe what dreaming is like. Thus, if the no-function hypothesis of dreaming is predicated on dreams being epiphenomenal, then it may be contested at a philosophical level.

That dreams are causally related to physical phenomena – i.e. they have physical effects on our bodies – is evidenced in experiments showing that what we do in lucid dreams has an effect on our bodies both in the sleep state as the dream is happening, and in the wake state after the dream is over. For example, lucid sex dreams may result in , and clenching one’s fists in a lucid dream leads to muscle twitches in the forearm, showing that the mental activity of lucid dreaming leads to appropriate physiological effects in the body (LaBerge, 2015). Furthermore, practicing a motor activity like playing darts in a lucid dream actually improves one’s dart-playing ability in subsequent waking life (Schädlich, Erlacher, & Schredl, 2017). These kinds of studies show clearly that dreaming has direct physiological effects on our bodies, so the claim that dreaming is epiphenomenal and has no causal relationship with physiological processes in our bodies is in fact demonstrably false.

As well as this, brain imaging research has recently shown us the neural correlates of dreaming (Siclari et al., 2017), i.e. the parts of the brain that are active while we are dreaming. Note that this is not a mere by-product of REM sleep, since REM sleep and dreaming are doubly dissociable. Why would energy be diverted to these parts of the brain 4-5 times per night, every night, for our entire lifetimes, if there was no point to it? What a waste of limited and costly resources that would be, especially in evolutionary terms, when, for our ancestors, acquiring a meal was not as simple as popping out to the local shop!

None of this conclusively demonstrates that dreaming, therefore, definitely is functional; it merely pokes holes in the argument that it’s not. Until we have more well-designed experiments, this matter will remain open for debate.


Flanagan, O. (2000). Dreaming Souls: Sleep, Dreams, and the Evolution of the Conscious Mind. Oxford University Press.

LaBerge, S. (2015). Lucid dreaming: Metaconsciousness during paradoxical sleep. In M. Kramer & M. Glucksman (Eds.), Dream Research: Contributions to Clinical Practice, pp. 198-214. New York: Routledge.

Schädlich, M., Erlacher, D., & Schredl, M. (2017). Improvement of darts performance following lucid dream practice depends on the number of distractions while rehearsing within the dream – a sleep laboratory pilot study. Journal of Sports Sciences, 35(23), 2365–2372.

Searle, J. (1992). The Rediscovery of the Mind. MIT Press.

Siclari, F., Baird, B., Perogamvros, L., Bernardi, G., LaRocque, J. J., Riedner, B., Boly, M., Postle, B. R., & Tononi, G. (2017). The neural correlates of dreaming. Nature Neuroscience, 20(6), 872–878.

%d bloggers like this: