Bryan Caplan  

Ignoring: There Is Such a Thing As Free Sleep

What I'm Reading... Top-Down Health Care Reform...
New parents' number one complaint has got to be sleep deprivation.  When you've got a newborn, some disruption is inevitable.  But parents' sleep often suffers for years.  I'm pleased to report, then, that children's sleeping problems (and therefore parents' sleeping problems) can usually be solved simply by ignoring them. 

The old-school version is known as "standard ignoring" - if your kid refuses to sleep, you just let him cry himself out.  The softer version is known as "graduated ignoring" or the Ferber method - if your kid refuses to sleep, you wait for a predetermined period, briefly comfort him if he keeps crying, then leave.  Over time, you increase the predetermined period.

There's a whole literature on ignoring experiments, but this one is particularly nice because it also measures both effectiveness and side effects.  The researchers recruited parents with kids with sleep problems, and randomly assigned the kids to (a) standard ignoring, (b) graduated ignoring, or (c) wait-list.

Effectiveness: All of the kids initially averaged 1 or fewer "good bedtimes" and "good nighttimes" per week.  After just one week of treatment, kids in the Standard Ignoring treatment rose to 3.83 good bedtimes and 2.29 good nighttimes; kids in the Graduated Ignoring treatment rose to 3.69 good bedtimes and 2.55 good nighttimes; wait-list kids stayed unchanged.  After three weeks, Standard Ignoring kids were up to 5.36 good bedtimes and 3.43 good nighttimes, Graduated Ignoring kids were up to 4.92 good bedtimes and 4.91 good nighttimes, and wait-list kids stayed unchanged.  At the two-month follow-up, the Standard Ignoring kids had 5 good bedtimes and 4.46 good nighttimes per week, and the Graduated Ignoring kids had 6.47 good bedtimes and 6.38 good nighttimes per week.  (There's no further data on the wait-list kids).

What about side effects?
[O]nly positive side effects were associated with both treatments... Relative to WL [wait-list] mothers, StdI [Standard Ignoring] mothers experienced reduced distress about parenting and were less likely to use verbose discipline strategies; they showed significant pre-post improvement in both areas. GrdI [Graduated Ignoring] mothers reported better interactions with their children relative to WL mothers, and showed significant pre-post changes in this area.  No side effects were observed on measures of general child behavior problems, marital functioning, or parental depression.
If ignoring is so great, why doesn't everyone do it?
Parents in this study who were assigned to the standard ignoring treatment were not more likely than graduated ignoring parents to drop out of treatment; in fact, 82% of all families completed treatment. Parents who dropped had younger children (22- vs. 29-month-old children) and expected to have more difficulty complying with treatment. Drop-out parents told us that they could not ignore their children's crying regardless of the treatment to which they had been assigned... For treatment completers, however, graduated ignoring was easier to implement, particularly during the nighttime, and contrary to suggestions from previous research... graduated ignoring did not prolong treatment by training children to cry longer.
Bottom line: A few weeks of ignoring can easily buy parents extra years of good sleep.  And they don't even have to bite the bullet of standard ignoring; graduated ignoring is about equally effective.  Now that's what I call a deal that's too good to pass up!

Comments and Sharing

COMMENTS (18 to date)
Joe Carl White writes:

This happened to be the first thing in my newsfeed after finally getting my 3-week-old daughter to settle down. Am forwarding to the wife for her to read when she is up for the 4am feeding.

agnostic writes:

Another easy fix, and one that's healthy for them too, is to forbid them to eat anything with lots of carbs after dinner -- sweet snacks obviously, but also fruit drinks / juice, potato chips, anything made from flour, etc.

We produce insulin in response to glucose in the blood, and that mostly comes from carbs in the diet. When insulin is high, we can only burn glucose, which burns pretty quickly (like during a sugar rush / crash). We can't burn fat for energy then because insulin locks fat in fat cells.

Bad sleep, especially nightmares, happen because the body is under physiological stress during sleep. The typical way that happens is that we aren't getting enough energy to run our body's necessary systems while asleep. That stresses us out, motivates us to wake up, and try to get some energy. This is why, in the absence of late high-carb snacking, our insulin level naturally falls after dinner and stays very low throughout the night and early morning.

So, when you eat high-carb food after dinner, your insulin level will remain high even when you're asleep. That locks fat in fat cells, so you can't burn it for energy. And by the time you're asleep, you've burned right through the glucose from the high-carb food -- and you obviously can't eat more glucose while you're asleep. That lack of energy stresses you out, gives you bad dreams or bad sleep, and makes you wake up and want more high-carb food to snack on.

When you go on a low-carb diet, you'll never have bad dreams again, except for the few times you go off of it -- then you're certain to have a nightmare or two. So there's another easy solution -- no carb-y food after dinner. Better yet, don't let them eat such junk during the day either, and you won't have to suffer their sugar-induced temper tantrums either.

Honeyoak writes:

Oh Bryan, you realize that you are now officially evil; telling parents to ignore the cries of their children. Its a good thing that I like you so much, other people would be looking for the best way to burn you on the stake!!!!

Brian S. writes:

I've been reading a book called The Happiest Baby on the Block, where the author, an Harvey Karp, MD, describes a method to quickly calm down crying infants. In short, he says the first three months after birth are like the "fourth trimester", and the babies respond well to what simulates the womb: being tightly confined by a blanket ("swaddling"), being on their side or stomach (rather than their back), hearing white noise, mild "jiggling", and sucking on something.

Here's the Wikipedia entry on the technique.

The videos of him calming a baby down quickly are impressive (here on YouTube), and apparently babies sleep for longer durations when swaddled & with other conditions above.

Anyone here have experience with this?

The Cupboard Is Bare writes:

It's just my opinion, but I've found that there are two types of that occurs when a child is truly in distress; the other is attention seeking. You can generally hear a difference in the sound of the cries. Also, babies who cry because they seek attention will magically stop the moment you pick them up, whereas babies who are in true distress generally take a while to compose themselves, if they can compose themselves at all; i.e. the pain of an ear infection.

I recall the time we were invited to the home of a couple who acted like it was a criminal offense to let a child cry. The entire evening revolved around coaxing their toddler to go to sleep. We were not permitted to speak above a whisper, and the volume on the television was kept so low that I could not hear the movie. My requests to raise the volume were denied, and since I lack the ability to read lips, I had to keep asking, "What did they say?"

The mother repeatedly ran into the child's bedroom at the first sound of a cry, and she would stand there begging her son to go to sleep. I told her that she was rewarding the child for his behavior, which was only making it worse, and suggested that she gradually stretch out the time between his initial cry and her response to his cry. You should have seen the dirty look she gave me. You would think that I'd told her to sell him to the gypsies.

Neal W. writes:

At what age is it ok to start the ignoring treatment?

eccdogg writes:

My wife and I tried some of these techniques with our first daughter and it did not work for us.

The crying seemed to amplify over time to the point where my daughter would get so worked up that it would take extended time with her to get her to calm down.

We switched to a program of going up stairs to see our daughter after a very short amount of crying doing a quick settle down and then leaving. We also let her come get in bed with us when she woke up if we could not get her down after a few tries. Sometimes I felt guilty about being too lax with this program.

However that program seemed to work well and now at 3 she is a good sleeper and when she does not go to bed right away she stays up and talks to herself in bed without crying. She also almost always sleeps through the night.

My point with this is that there are far too many paint by numbers approaches to child raising. Some work in some situations for some parents and kids. I think it is better to trust your instincts and do what feels right for you and your child and not feel guilty about it. Kids are pretty hard to screw up without gross negligence.

Stan Greer writes:

Bryan, perhaps the reason most people are reluctant to ignore their crying babies is because they are less impressed by putatively "scientific" studies showing ignoring babies does no harm than you are.

Some of us tend to trust our gut rather than such "scientific" research, especially when long experience indicates the scientific consensus on the results of ignoring crying babies may be very, very different in another 10 years -- when you, of course, will be unable to make amends.

Anyway, there's a much easier way for you, your spouse, and your baby to get a good night's sleep.

Keep the baby in bed with you, and let the baby nurse when he/she's hungry or needs comforting. This may or may not wake up your spouse, but she will surely get more sleep this way than she will by letting the baby "cry it out," at least during the weeks it takes before you achieve your goal of driving your baby to despair.

If you don't get drunk or take narcotics before you go to bed, and you aren't extraordinarily over weight, your baby is probably safer in bed with you than elsewhere.

And you can always put the baby somewhere else for half an hour or so or whatever if you and your spouse need some private time!

Stan Greer
Fairfax, Va.

Matt C writes:

We used the Ferber method and it worked reasonably well for us (our first was a difficult sleeper, period).

If I recall correctly, Ferber said that the purpose of "graduated ignoring" is primarily to give the *parents* an easier way to learn to ignore the crying child.

q writes:

i'd recommend reading ferber's book to anyone reading this comment who is a parent with a child who doesn't sleep well.

it's the only book on kids' sleep that i was able to find that discussed the science of sleep. whether you follow any of his methods (we didn't) it was extremely helpful to have a framework to think about it.

in our case our kid's sleep problems mostly fixed themselves. i'm sure we could point to one or another thing we did but things just sort of snapped into place on their own. that's part of parenting -- you never know what your influence is or isn't.

i have heard that the first edition of his book makes him come across as much more judgemental than the second edition and that his 'mean reputation' comes from the first edition. so, ignore his reputation and read the second edition.

Norman writes:

I have to second Stan Greer. You can go through all kinds of pains to try getting your night life back to the way it was before the baby was born. It might even work. It might not even be cruel (especially if cruelty is only defined by side effects looked for by scientists). Or you could just make mild adaptations to the fact that you have created a new human being and go with it.

The point is that there are plenty of studies on both sides of this issue, as there are plenty of studies on both sides of the breastfeeding issue. They all have their pluses and minuses and influences by ideology and funding. For me, the most intuitive (which I figure you'd appreciate) and sensible approach is that advocated by William Sears, MD ( for an introduction).

Stan Greer, if we ever meet, I'll have to buy you a beer! (just one, though, as co-sleeping requires sobriety)

Patri Friedman writes:

Brian S - we found the "happiest baby" techniques worked great on our newborn.

Jimbino writes:

My dad stopped the grandkids' crying simply by plugging in his electric shaver near the crib.

E Abraham writes:

Brian S - we also found the "happiest baby" techniques to be very effective, up to a point. They take some practice, and it takes some experimentation to find the exact combination of jiggling, rocking, shushing, etc. that works for each kid. I don't know if these techniques are more effective for my second baby than they were for my first because of her temprament or my experience or some other thing.

The only trouble with these techniques is that babies outgrow them. Sometime between three months and twelve months, you need a new strategy. There are lots of options out there.

I think the big flaw in this study is that it doesn't discuss the issue of parental readiness. Parents who don't feel able to emotionally disengage from a crying child will not be able to pull this off.

Jennifer writes:

For those (men, I'd note) who advocate cosleeping with on-demand nursing:

While I have no doubt that this will result in fewer disruptions in sleep for the partner who does not have to nurse the baby (as compared to CIO), I can state for a fact that, in at least one case, it results in less sleep for the partner who is nursing the baby all night long.

I know many women who believe that cosleeping and nursing when the baby wants to nurse is the way to go, and I say, bully for them. That's great. But I don't know a single one who claims that it leads to more or better sleep.

David writes:

We read both Karp and Ferber. I think the use of Ferber is for when they are at least a year old. The five s's worked brilliantly for my daughter and did nothing for my son. He hated the swaddle, but wanted my wife and I used a sling on him [well, a wrap really that mimicked the position of the womb]. My daughter loved the swaddle and used it until she was almost one [and we stopped when started breaking out of it in her sleep].

We ferberized my daughter at 16 months and now she sleeps like a rock. My son, wno is approaching that age, may or may not get the treatment. We did it for my daughter because she woke up constantly in the night. My son has a tough time getting to sleep, but then sleeps like a rock.

It was much harder on my wife than on me.

Sarah writes:

I can't help but wonder if "good bedtimes" and "good nighttimes" in this study are because the parents have cultivated a state of learned helplessness in their child. I can't imagine treating my children with such a complete lack of empathy. I also notice that they claim there are no side effects on general child behavior problems. That may be true for study period, but what was that? A few months? It seems plausible that there may be long term ramifications to ignoring your children that might not be immediately obvious.

Cosleeping and nursing on demand worked for me when my kids were young. They didn't wake up alone and distressed, and I was able to sleep through most of the nursing. I absolutely believe I got more, better sleep that way- but I also recognize it doesn't work that way for every family.

CB writes:

I have no idea why parents allow disrupted sleep to drag on for YEARS. Weeks, sure. Months, maybe.

But I know women who've not had an unbroken night's sleep in years, let alone a normal, spontaneous sex life. Sorry, but life's too short to walk around in a sleep-deprived, sexless fog for five years. Plus, it's dangerous. Death and divorce lie that way, as far as I can tell.

Small wonder plenty of couples view the idea of having a second or third kid with something like horror!

I'm on my fourth baby now. You cannot pick up a fourth baby every time he squeaks. Sometimes he just has to lay in his crib and bawl for fifteen minutes before I have a moment to attend to him. Half the time, he falls back asleep before I get back to him.

Not surprisingly, this is by far the best sleeper I've ever had. (The worst one was, unsurprisingly, the eldest, who was held about 18 hours a day. Back then I couldn't even manage to wash the breakfast dishes while tending the baby! lol.)

Anyway, I suspect "learned helplessness" for babies is probably par for the course for most of human history. Mama had a lot of other kids and a lot of other chores that came before a squalling six-week-old.

Is the baby fed? Is the baby warm and dry? Is the baby safe? Then let it cry for awhile; holding it while it screams isn't going to help you get anything else done. And lord knows, you've got a lot of other important maintenance tasks to perform, like mopping the floor and making some whoopie. ;)

Comments for this entry have been closed
Return to top