All of this, Barrett notes, led many researchers to conclude that dreaming was necessary for one or more of these functions: to replenish neurotransmitters, rest a particular brain area, or restore the thermoregulatory system. In general, these brain conditioning theories don’t place much, if any, emphasis on dream content. As psychologist Steven Pinker writes, “For all we know, dreaming might be a kind of screen saver in which it doesn’t really matter what the content is as long as certain parts of the brain are active.”

External Vigilance
University of California at Santa Barbara anthropologist Don Symons wasn’t entirely satisfied by brain conditioning theories of dreaming, in large part because these theories didn’t really crack the question of why dreams have such a specific sensory profile of being so vividly visual and kinesthetic while comparatively impoverished in sound, smell and other sensory domains. Symons points out that sleepers are particularly vulnerable to real-world threats and hazards in their external environment, and so they must unconsciously monitor their environment with specific senses. For example, if our ancestors were busy having olfactory or auditory hallucinations in their dreams that were equally as rich as their visual hallucinations occurring beneath their fluttering eyelids, well then they might not have noticed the ominous smell of smoke creeping up their nostrils or the threatening strangers or predators pattering around outside. Being a “light sleeper” in relation to these other sensory domains had adaptive benefits, and since we’re in the dark anyway and our eyes are closed, there’s less of a risk in hallucinating in our secret visual worlds while our brains are being recharged through the processes described in the previous section.

Threat Simulation Theory
Originally proposed by Finnish neuroscientist Antti Revonsuo, this clever evolutionary theory holds that dreaming serves a biologically adaptive function because it allowed our ancestors to simulate problem-solving strategies for genuine, waking life threats. Antonio Zadra, Sophie Desjardins, and Eric Marcotte of the University of Montreal neatly summarize the central argument of the theory this way: “By giving rise to a full-scale hallucinatory world of subjective experience during sleep, the dream production mechanism provides an ideal and safe environment for such sustained practice by selecting threatening waking events and simulating them repeatedly in various combinations.” What we should see in contemporary dreams, argues Revonsuo, are “threat scripts” depicting primitive themes of danger that would likely have been relevant in the ancestral environment, such as being chased, falling and so on.

Costly Signaling Theory
Boston University neuroscientist Patrick McNamara has an interesting evolutionary theory of dreaming. McNamara’s theory draws from the well-known “handicapping principle” in evolutionary biology, where some organisms have been observed to display behavioral traits or physical characteristics that seem ostensibly to disadvantage them but in fact simply reflect their genetic value. The classic example of this is “stotting” behavior in healthy young gazelles, where these animals jump up and down in front of a predatory leopard rather than—what would seem to be a smarter move—immediately running away. Stotting is a “costly signal,” but it works, because the leopards take this stotting display as evidence that this particular gazelle is so healthy and fit that it can afford to handicap itself and is therefore unlikely to be an easy target. Usually, the leopard moves onto the sick, old, or young non-stotters.