Why Do We Forget Dreams?

April 16, 2004 10:17 AM

I had a vivid dream last night and when I awoke I thought "man, I'm never forgetting that one." It's been a couple hours and I barely remember anything from it. So my question is why do we forget dreams? Is it because there is a lack of pattern (since they're often so oddball and different from our daily lives)? Is it because they exist as ideas rather than experiences (I forget good ideas all the time, but remember things I've actually done)? Or is it because we experience them in a semi-concious state (if we were more fully aware, we might remember more?)? Anyone got any leads on this?

So far, all I can find is this, which says it's the semiconscious that is the problem.

Personally I want to know more info about this because it kind of bugs me that I keep forgetting great stories and ideas in my sleep. I know I should write them down, but it'd be nice if I could recall them later when I wake up.

posted by mathowie to Science & Nature (30 answers total)

I tend to remember dreams somewhat better if I try hard to remember them in full order a couple of times after I wake up. My best guess of why that helps some is because it forces my brain to make new memories of it while I'm awake, so it has a better chance of being moved on to medium term and long term memories.
posted by skynxnex at 10:31 AM on April 16, 2004

Matt: Perhaps you should see if you can get a Bluetooth adaptor for your brain and just partner it with your phone and have it call you later in the day to add them to your weblog.

Just a suggestion ;)

Seriously, I have a similar problem. If the alarm clock wakes me it acts like an instant erase button for dreams. If I wake up on my own and lie in bed contemplating the dream--or better yet, bug my wife with the details--I am much more likely to remember them. So maybe it isn't so much writing them down, but just speaking about the dreams that will help file the dream in a spot in the brain that is easier to retrieve it from.

A hunch.
posted by terrapin at 10:39 AM on April 16, 2004

I remember hearing something about this on NPR and they were saying that the chemicals in action in your brain while you are dreaming suspend conscious thinking and allow for you to process all of your data and store it deep in the recesses of your mind- encoding it as memories and making connections and such. I think I recall them saying the chemical processes were mutially exclusive from what happens during your normal wake state and that's why those memories of dreams fade away. They aren't natural in our awake state. Of course, this is all layman's terms for stuff I half remember.
posted by pissfactory at 10:41 AM on April 16, 2004

Also, on this topic.... I do think writing them down helps to remember, because there is some connection made in your brain when you write things out... I have no idea what the term for this is, but there is a correlation to learning/remembering and actually writing stuff out.
posted by pissfactory at 10:43 AM on April 16, 2004

Though this quote from the 2001 film Waking Life might well be 100% wrong, it sounds informed, and thus some of it might be correct:

See, in the waking world, the neural system inhibits the activation of the vividness of memories. And this makes evolutionary sense. You'd be maladapted for the perceptual image of a predator if you mistook it for the memory of one, and vice-versa. If the memory of a predator conjured up a perceptual image, we would be running off to hide every time we had a scary thought. So you have these serotonic neurons that inhibit hallucinations and they themselves are inhibited during REM sleep. See, this allows dreams to appear real, while preventing competition from other perceptual processes. This is why dreams are mistaken for reality. To the functional system of neural activity that creates our world, there is no difference between dreaming a perception and action, and actually the waking perception and action.

posted by profwhat at 11:03 AM on April 16, 2004

bugs me that I keep forgetting great stories and ideas in my sleep.

Have had the same problem, and wonder if when first waking up it seems like a great idea for our mind. Yet, later our brain sorts it all out as a thought but not the "great" one as you earlier had. Example, having a few beers and thinking the thoughts you are having are awesome. Then later upon sobering up realize, WTF. Though, I've always thought dreams may become real as reality can come from our dreams about our lives.
posted by thomcatspike at 11:05 AM on April 16, 2004

Along pissfactory's lines, I've also heard that our dreams are stored in a kind of liquid in our brain. Consequently, when the alarm clock goes off in the morning and you have to dart for it, the liquid (or chemicals or whatever it is) gets swished around so the dreams are lost.
posted by jmd82 at 11:09 AM on April 16, 2004

I think it depends on how valuable it will be to you, 'spiritually', to remember the dream.

Sometimes it's good to forget a dream because it has served its purpose and you're moving on, moving on ...
Sometimes you remember them because it's valuable that you remember them. For example, I had a dream at about the age of seven, which I can still vividly recall today, and it was a concise description of my family life, and my existential circumstances, which informed me to some good effect.

The best way to remember dreams, should you feel the need, apart from radically altering your sleeping pattern (which can produce vivid and bizarre dreams hard to forget, or so I've found) is to write them down on a pad that you keep by your bedside, as soon as you wake up.
posted by Blue Stone at 11:19 AM on April 16, 2004

I heard a theory once about the purpose of dreams. It's been a while so I might not get this right, and I don't even know how much faith to put in the theory to begin with since I heard it second-hand.

Think of the neural pathways in the brain like roads. Over time our thoughts strengthen some well used roads into thoroughfares, allowing our thinking to become more efficient. For example, when we hear 6 times 9, our brains immediately go to 54 rather than through a time-consuming calculation subroutine.

However, if this process continued unchecked, the off-ramps on the thoroughfares would close down and every time you started at New York you would have to end up in Los Angeles. Six by nine would never equal 42. We'd be locked into the same patterns without the ability to unlearn mistakes.

Dreams, therefore, are the brain breaking down these overworked paths by firing little used or odd routes instead, resulting in irrational, impossible, and unusual thoughts. Once you're awake it's harder to remember them because your conscious brain is once again wandering down the predictable paths, so unless you work at widening those irrational paths you'll lose them.

Yeah, I know. Sounds like a lot of shit. There's been a lot of brain research on dreams, but little definitely concluded. This theory makes some sense, but it too has problems (if they're so random why do we have recurring dreams?). We're too far from understanding a lot of brain function to get a better answer than that anytime soon, however.

If this sort of thing happens to you a lot (great dreams you want to remember but quickly forget), put a pad of paper and pen beside/under your bed. That's what I do, and I've used it to remember key points of dreams I wanted to keep. They're great material for short stories.
posted by GhostintheMachine at 11:32 AM on April 16, 2004

Memory encoding certainly works differently during sleep and immediately after waking, but there is a practical technique that you can use to improve your dream recall. Most people rarely think about (and consequently rarely remember) their dreams. If you develop a habit of writing down your dreams immediately upon waking, as others have suggested above, you train your attention and your mind begins to think of dreams as "important" and "worth remembering." I've written mine down for over 20 years now and can recall hundreds of dreams, and recognize thousands when I read a written cue. In some cultures, family members all share their dreams verbally every morning at the breakfast table, which apparently produces a similar focus of attention. Basically, with dreams as with other types of perception, your brain will edit out what you don't consider important or aren't interested in and show you what you do/are.
posted by rushmc at 12:05 PM on April 16, 2004

At a slight tangent:
Last night I dreamed that GW Bush was shot dead by a gunman in Sri Lanka. The actual part of the dream I remember vividly is looking at my travelling partner & the 'Oh no, this ain't good' look we cast each other as we found out the news.

The strange thing is, as I woke up I knew that there would be a thread about dreams on MeFi today.

Anyhow, I'm wondering if the 'same-state recall' theory of memory comes into play at all with dreams. ie you are more likely to remember something if you are in the same state (drunk, asleep, relaxed, watching TV etc) & situation (at a desk with no noise, in a bar, in bed etc) as you were when you learned or experienced it.

Ironically, my memory of psychology theories is a little hazy some 16 years on from college so I can't really elaborate much more ;-)
posted by i_cola at 12:06 PM on April 16, 2004

Something that works for me, but I haven't heard anyone else try this, so there are no guarantees is this: Before bedtime, if I drink a cup of gingko tea (like Celestial Seasonings Gingko Smart or Yogi Gingko IQ), it helps me fall asleep, and I almost unfailingly have very vivid, entertaining dreams that night that I always remember the next morning. I haven't yet looked up the reasons for this either. Also, I know for some people it actually makes them a little jittery so they can't fall asleep. So YMMV.
posted by greasepig at 12:09 PM on April 16, 2004

More on point- My guess is that dreams are generally a regurgitation of experiences and emotions; they're issues being worked out, but not in a way that makes waking sense*- they're the waste products of a functioning mind. Our brains probably just shut off the memory function for most of it, sort of like flushing the loo when you've finished with those waste products, too.

What I'd like to know is how the "Sure, I will remember this dream!" and then you don't action can work in the same brain that occasionally has dreams so vivid, you can't stop remembering it, and said dream lingers for days (or even years.) Most of the time, I wake up with the sense that I have dreamed, but I don't remember what, yet I've had a few nightmares that were so vivid, that I can still describe everything that happened, and they still give me the willies today.

*(Hence the "Okay, I was in high school, but it wasn't really my high school, it was a strip mall..." kind of setting. Those elements were emotionally important in the dream, but when expressed in a waking sense, they're just kind of odd and disjointed.)
posted by headspace at 12:21 PM on April 16, 2004

Er, I thought I deleted that "More on point" when I previewed; it's a leftover from the editing process and not a comment on other people's comments!
posted by headspace at 12:23 PM on April 16, 2004

i had a dream last night and, at 4 am, was lying in bed wondering whether it could be classified as a nightmare or not, and what it meant, whether it was a reflection on problems i have now, or problems i've solved, and what i should do to resolve any latent conflicts in my life... stupid obsessive behaviour that probably cost me more sleep than the dream itself.

anyway, one way to look at the problem is to ignore my amateur self-psychology and claim that dreams have no advantage, survival-wise. so there's no evolutionary advantage in remembering them, so it's no great surprise that we don't.

i'm no sure i buy that, though, for the same reasons that kept me lying awake last night... (it was, incidentally, a dream in which i was kidnapped - by my previous employers, although that didn't seem important at the time - and, when i escaped, and got to the police, they said i had already been found... the ex-employer stuff i can understand (the bastards), but the "already found" bit suggested some deep-seated insecurity about identity, i recokon. which is worrying, because i thought it was pretty clear about that.)

posted by someone who thinks that this was
posted by andrew cooke at 12:28 PM on April 16, 2004

In some cultures, family members all share their dreams verbally every morning at the breakfast table, which apparently produces a similar focus of attention.

I had a college professor that was doing a lot of dream research and talked at length about such practices. We had to keep dream journals as part of the class, which was very easy for me, as my mother started asking me about my dreams as soon as I could talk, so I was in the habit of remembering them.

You know the falling dream we all have that tends to wake you with a jolt? He told us it's common practice in some cultures to tell kids who report having a falling dream, "it's too bad you woke up, because everybody has that dream and it means you're going to fall into a big, soft pillow," or something like that. The kid either stops having the dream or stops jolting awake from it. The professor told us it was inappropriate to do this kind of "shaping" in our culture, though he never explained why.

Anyway, I thought it was kind of cool, so that's what I'm telling my kids. I'll let you know if they turn out completely malajusted and/or psychotic.
posted by whatnot at 12:30 PM on April 16, 2004 [1 favorite]

GITM: That theory is very similar to a theory of intelligence by Doug Hofstadter in Godel, Escher and Bach.

i_cola: The same-state recall theory is supported by experimental evidence both at the behavioural and cellular/neural networks level. It's believed that memory is stored by strengthening connections between neurones, through a much-researched process called long term potentiation (LTP). If you imagine a network of neurones that receive inputs from your senses, then as you experience the same input again and again, the same connections will be strengthed. To recall that 'input' (be it an image, smell, event) you only need to activate the first neurone in the network, and the rest will follow because of the strengthened connections. That's what the computational neuroscientists think, anyhow.

From my own experience, I generally find it difficult to remember dreams but if I recall the appropriate trigger within a few hours of waking, then the entire dream will come flooding back. Sometimes the trigger will be as little as thinking about a friend or a face. Interestingly, the trigger is never a sensory experience; it's always an internal thought, which perhaps suggests that recall is dependent on the method in which the memory was encoded (i.e. when you're dreaming, you're not sensing anything, it's all internally generated).

That doesn't mean, however, that you can only remember a smell if you smell it again - if that were the case, memory would be a pretty terrible thing. You could recall a smell by recalling something connected to it, like the image of a rose, which would then activate related neural networks.

Personally, I think you get strong memories by one of two methods (very broadly speaking).

1) Repetition, via repeated strengthening of connections
2) Emotional salience and multimodal encoding.

What do I mean by the second one? I mean that if you're in a car crash, you don't forget that quickly, even though it's a one-off event. A car crash will involve all your senses overwhelmingly and also charge you with adrenaline, fear and excitement; as a result, I think that neuronal connections formed during this period are heightened. Of course, by replaying the memory in your head the connections are strengthened again, like in repetition.

Digressing even further, an interesting finding in neuroscience is that there is no such thing as forgetting - or rather, the process of forgetting (called extinction) is served by the same process as that of learning new memories. While this may seem a little peculiar, it does make sense in the neural network point of view; when you forget something, it's not as if you're destroying connections or neurones; instead, you're just resetting them or overwriting them.
posted by adrianhon at 12:34 PM on April 16, 2004 [1 favorite]

I just realised that what I said about triggering the recall of dreams might not make much sense - of course if I see a picture of a friend, and then think of the friend (who featured in my dream), then it'll come flooding back, so sensory perception can trigger recall in that way. It just seems that an internal trigger tends to work more often and more directly.

BTW, I am a neuroscientist, and have studied experimental psychology, but my area is not memory or dreams.
posted by adrianhon at 12:41 PM on April 16, 2004

Hey mathowie, can you recall if your dream was a travelling dream? 'Cuz mine sure was. See i_cola's post.

In the dream, I was anxious about a kidnapping, but my memory of that aspect of the dream was so vague it didn't write it down. I start my day blogsurfing in bed, so blogging a dream is much like using the proverbial notepad.

I suspect my dream came from the coverage of troop rotation and redeployment and kidnapping on Morning Edition today.

On preview: Hey, look, kidnapping appeared in andrew cooke's dream.

Self-link, I know - in context, in discussion!
posted by mwhybark at 12:44 PM on April 16, 2004

the jolting dream is connected with shutting down your the system that lets you move bits of you body, i've heard. if this didn't get shut down, then we'd all be sleep-walking when we dream about walking, etc. also, it can fail to start up - apparently some people can wake up momentarily paralysed (yuck).

don't think the kidnapping in the news had much to do with my dream - i've not been following the news much the last few days, and while i'm aware it's something that's happening, it's not an important thing for me. in that particular case it's obvious, if you know what happened with my previous employers, that it's more about losing control over my life etc...
posted by andrew cooke at 12:52 PM on April 16, 2004

Response by poster: Hey mathowie, can you recall if your dream was a travelling dream?

All I can remember is that I was driving in one of those new mazda RX-8s for most of the dream, with some famous person, and I think we were on a road trip of some sort. The part I wanted to remember was all the adventures we had, but I can't remember a one. I remember the car sucked though, it wasn't nearly as fast as I thought it would be.
posted by mathowie at 1:09 PM on April 16, 2004

ahem. poorly performing car?! and i was worried about describing my dream in public... ;o)
posted by andrew cooke at 1:25 PM on April 16, 2004

Hmmm. From what I recall there are three schools of thought on dreams. 1> A dream is simply the hindbrain coping with stimulus while the cortex is mostly shut down. Apparently brains can be overloaded if info isn't processed in some fashion. 2> Freudian/Jungian/etc. dreams are our sub/unconscious minds telling us stuff. 3> Dreams are a rehearsal for real life situations. There is a lot of empirical evidence for both 1 and 3 and these two are the most commonly accepted explanations (at least in CogSci circles). As for why we can't remember them, individuals generally have different retention levels. I think that given explanation 1 dreams are not retained as they are not real experiences. I have noticed that I never remember my dreams. However, my GF will frequently tell me her dreams in the morning and this allows her to remember them from then on. If she doesn't tell them to me she generally forgets them. Telling a dream to someone may enable the brain to first process it through the language portion(s) which are linked with longterm memory, and shift the categories assigned to the story. Rather than being something worthless it becomes a story, humans are great at remembering stories.

I haven't studied cogsci in years, so what I said above is a little muddled and may not even be accepted now. I would add that if you can get into the habit of writing down your dreams each morning or telling them to someone then once it becomes a routine you will probably find yourself waking up in time to do it before they fade.
posted by Grod at 3:21 PM on April 16, 2004

i have quite a few day dreams that get mixed in with night dreams to the point where i can't tell them apart. Strange. Need more sleep perhaps.

My brother started keeping dream journals when he was about 10 years old. Now he is 30 and is extremely good at recalling dreams....he has put a lot of practice into it. You may want to try that.
posted by th3ph17 at 4:50 PM on April 16, 2004

I kept a dream journal for about a month. That was all I could take. The entries kept getting longer and longer. I was forcing myself to write down every detail, and I either paid more attention while I was dreaming or became better at salvaging details in my waking state or both. The result was three-page entries. I also enjoyed the dreams less.

What's spooky to me is not being able to recall a dream at all, really working at it and then, with the memory of one detail, remembering everything at once. What explains that cognitive feature? (It's a feature of waking life, too. You don't usually remember 10% of what someone looks like; you just suddenly remember it with the onset of a detail.)
posted by argybarg at 5:22 PM on April 16, 2004

While I have the same experience as Matt with generally forgetting dreams a couple of hours after waking, I've also had a few dreams which were particularly vivid in terms of temporarily convincing me of their reality after the event, (i.e. they conflicted with actual reality but were quite throwing). Does this happen to others? What is its basis?
posted by biffa at 5:47 PM on April 16, 2004

Y'all have dream themes? I have a couple of cities mapped out in my dreambrain that are loosely based on places I've lived. I always like having dreams that involve those places, because they're almost always slightly-to-very surreal.

Then there are the dreams where they just go all wrong, so I actually rewind them and play them out again correctly...
posted by five fresh fish at 11:12 PM on April 16, 2004

I have a recurring dream that has something to do with this water tower, that I am going up into it and inside is this business office. The dream is set in a somewhat cartoony setting so it doesnt really feel like a bad dream. The interesting thing is that I have the exact, to the detail dream years later when ive completely forgotten it otherwise.


Sometimes I dream I am in a gunfight, I shoot every bad guy except the last one, but I am out of bullets (oh... shit) or the bullets dont come out fast enough. Or a fight that I dont have the stength to win, or running away from something but I cannot run fast enough.
posted by Keyser Soze at 4:55 AM on April 17, 2004

I have repeating settings in my dreams too...it bothers me.

I've tried the writing down thing the second i wake up, but i think i'll try speaking it into a tape recorder soon--writing seems to dull it down and order it too much.
posted by amberglow at 8:01 AM on April 17, 2004

I'm wondering if the 'same-state recall' theory of memory comes into play at all with dreams.

Based upon subjective experience, I would say definitely so. It often happens that when I lie down to sleep at night, the dreams of the previous night which had been forgotten all day come flooding back.

the jolting dream is connected with shutting down your the system that lets you move bits of you body

The reticular activation formation is the part of the brain involved in dampening movement signals to the body during sleep. And related is when you are almost asleep and suddenly experience a violent jerking of the body, which is known as a "myoclonic jerk."
posted by rushmc at 9:28 AM on April 17, 2004