Dreams mean nothing? What science says and why write them down anyway

In 1977 two Harvard psychiatrists, Allan Hobson and Robert McCarley, published a paper that should have put the dream dictionaries out of business. They described what happens in the brainstem during REM sleep: groups of neurons fire almost at random, send signals up to the cortex, and the cortex does what it does best, builds a story out of fragments. A dream, in this model, is not a message but the brain's attempt to make sense of its own noise. Water in a dream does not "mean" change. Water in a dream means the cortex received a signal and reached for the nearest image.

The dream dictionaries did not close. But since then the question "what do dreams mean" has split in two inside science: why the brain produces them, and what in them is nevertheless ours. The first has four answers that do not quite agree with each other. The second has one, and it turns out to be useful to anyone who writes dreams down.

Answer one: noise

That is Hobson and McCarley. The model's strength: it explains why dreams are so strange. If the cortex assembles a plot from random signals, the jumps in place and time, the people who are simultaneously someone else, the staircases to nowhere all come for free, with nothing to interpret. Its weakness: it explained REM sleep, and in 2017 Francesca Siclari and colleagues in Giulio Tononi's lab woke people many times a night across sleep stages, hundreds of awakenings in all, and asked whether anything had been dreamed. People dreamed in slow-wave sleep too, and activity in the back of the cortex predicted whether a dream would be remembered, regardless of stage. Dreams turned out to be wider than the mechanism used to explain them.

Answer two: cleanup

In 1983 Francis Crick, the Crick of the double helix, together with Graeme Mitchison proposed that REM sleep exists to "unlearn": over a day a neural network accumulates spurious, parasitic connections, and at night the brain runs them in reverse so they weaken. Dreams are a by-product, and remembering them, by Crick, is if anything harmful: you reinforce what the brain was trying to erase. This was a hypothesis from network theory, not from a lab, and in that form it did not hold up. But the idea that sleep sorts through the day proved durable, in the opposite direction.

Answer three: rehearsal

In 2000 the Finnish philosopher and neuroscientist Antti Revonsuo gathered data on dream content and noticed a disproportionate amount of threat: chases, falls, attacks, being late, losing things. He proposed that dreams are an evolutionary simulator, a safe place where the brain rehearses its response to danger. The hypothesis explains why nightmares exist in every culture and in children who have not yet seen anything frightening. It does not explain why dreams also contain the dull: work, conversations, food. Revonsuo says so himself.

Answer four: the day continued

Here the data is the most down to earth. In 2003 Michael Schredl and Friedrich Hofmann had participants keep parallel diaries for two weeks: what they did by day and what they dreamed by night. Daytime activities did enter dreams, but unevenly: what was emotionally loaded and involved people got in more often; reading, writing and working at a computer almost never did, despite taking hours. A dream does not copy the day. It picks from it what touched you.

And in 2010 Erin Wamsley and colleagues had people learn a virtual maze, then nap, then try again. Those who dreamed about the maze did noticeably better afterwards. Not those who had thought about it more before sleeping. Those who dreamed it.

Participants who dreamed about the maze they had learned improved markedly more after sleep than those who did not; thinking about the task while awake produced no such effect.

A small sample, a single task, and the authors are careful: dreaming of the task may be a sign that the brain is busy consolidating it rather than the cause. For our question that does not matter. What matters is that the dream in which the maze appeared was not random.

So do they mean something or not

Put the four answers together and it comes out like this. The plot of a dream is most likely not a message: the cortex assembles it from whatever is to hand, and there is nothing to lean on when reading water as change. But what it is assembled from is not random. Into it goes what touched you during the day, who you had dealings with, what you were learning, what you fear. A dream does not mean, a dream shows. The difference is who authors the meaning: a dictionary or you.

That answers the question in the title. Dreams are worth writing down not to decode them but because it is the only way to see what repeats. One dream about being late says nothing. Eight dreams about being late in a month when you took on too much say exactly what you already knew but had not written down.

What to do with nightmares

Here science is more careful, and so am I. In 2007 Tore Nielsen and Ross Levin described nightmares as a failure in sleep's ordinary handling of emotion: a dream normally takes a daytime fear and re-places it in a different context so it loses its edge, and in a nightmare that transfer does not happen and the fear arrives raw. A practical thing follows from the model: writing a nightmare down does not reinforce it, Crick notwithstanding. A written nightmare gets the context it was missing at night.

But if nightmares recur for weeks, break your sleep and make you dread going to bed, a journal is not the tool. Sleep specialists work with dreams like that, and your entries will be useful to them: the conversation will be more concrete.

How to write them down so there is something to reread

Briefly, because I have a separate piece on this. Before you get up. Fragments, not plot: "staircase, former colleague, cannot find the door". Tags for images that return. Once a month, reread and look not at meanings but at frequencies: what repeats, after which days, with whom.

And one rule that follows from everything above. If the phrase "this means that" appears in a dream entry, replace it with "that day I". The first leads to a dictionary. The second leads to your journal, where the answer most likely already is.