Extinction: Why Cry It Out is Extinct

I have argued before that no modern sleep expert recommends pure “cry-it-out” (CIO) sleep training. Here I want to explain in more depth where CIO stands currently. In order to simplify the discussion, I’ll call it by its more technical name, extinction. Some versions of this method are widely recommended today. One kind, unmodified extinction, has all but disappeared, or become extinct. That’s a shame, because it is highly effective, as I’ll explain below.

When it comes to any kind of human or animal behavior, “extinction” refers to the disappearance of a behavior in the presence of a stimulus. Pavlov famously could make his dogs salivate when he rang a bell (the stimulus). The dogs had learned that the bell meant Dr. Pavlov was about to feed them. That was only the first part of the experiment. In the second part, the dogs stopped salivating after a while. In other words, the salivating behavior “extinguished” with time.

Extinction in Sleep Training

When it comes to sleep training, the behavior we are trying to extinguish is crying. In sleep training, there are now three versions: Extinction with parental presence, graduated extinction, and unmodified extinction.

Extinction with parental presence is a version of CIO where the parent stays in the baby’s room, but does not respond to cries. With time, the caregiver moves farther away from the crib. Finally, the baby sleeps alone. This method has been championed by Kim West. She renamed the method “The Sleep Lady Shuffle”.  The terms get even more confusing because West’s followers refer to her method as “bedtime fading”. This is very different from true fading techniques.

The graduated variety is today better known as “Ferberization“. The method involves answering the baby’s cries, but doing so at longer intervals every night until the baby goes to sleep on her own.

Finally, we come to the true dinosaur, the unmodified CIO technique, sometimes called “cold turkey”. The villain of the CIO story is a late-19th century pediatrician named Luther Emmett Holt. The first time we see the words “cry it out”, they appear in Holt’s 1894 catechism “The Care and Feeding of Children.”

Need a SLEEP COACH?

Holt’s book is not a “sleep book”. In fact, there is no section on bedtime routines such as the kind we’ve grown used to. There is, however, a section on the types of infant cries and what to do about them. The section that follows is really about “problem crying”.

What should be done if a baby cries at night?

One should get up and see that the child is comfortable—the clothing smooth under the body, the hands and feet warm, and the napkin [diaper] not wet or soiled. If all these matters are properly adjusted and the child simply crying to be taken up, it should not be further interfered with. If the night cry is habitual some other cause should be sought.

How is an infant to be managed that cries from temper, habit, or to be indulged?

It should simply be allowed to “cry it out.” This often requires an hour, and in extreme cases, two or three hours. A second struggle will seldom last more than ten or fifteen minutes, and a third will rarely be necessary. Such discipline is not to be carried out unless one is sure as to the cause of the habitual crying.

Holt is describing a particular kind of crying here. This is crying born of habit. These are babies who have learned that crying can draw her parents’ attention. I suspect he is referring to babies who’ve acquired “object permanence“. That is, the baby knows that mom or dad is still there when they leave the room, and she can induce them to come back.

Dr. Holt is not describing a way to get a baby to fall asleep. He is describing a solution to a problem. That sleep problem is “bad sleep associations“. For example, the baby won’t sleep unless she has physical contact with a parent, or if she has a binky in her mouth. Likewise, sleep books that discuss “cold turkey” or any other CIO technique are aimed at families that already have developed bad sleep associations and want to reverse them.extinction 3

The Moa is Extinct. So is the Cold Turkey

Of the three types of extinction methods, only unmodified extinction, or cold turkey, has disappeared. It has ceased to be (see video below). And yet, cold turkey has been tested experimentally and found to be extremely effective. Several arguments agains cold turkey have been raised, not least that the method stresses the baby and caregivers. These objections have also been tested and so far have proven to be false. Even the most-cited article arguing for the negative effects of CIO  failed to show that the stress hormone cortisol goes up in crying babies!

The main reason cold turkey has gone extinct is that it is really difficult to listen to a child cry. Most parents can’t handle it. I know I couldn’t (the boys’ mother was the stronger partner). Listening to crying becomes even more difficult because of a thing called the “extinction burst”. This is an increase in crying as the sleep training process proceeds. Sometimes the burst happens after it appears that you’ve succeeded and the baby is sleeping through the night. In either case, the stress for parents becomes too much.

A second reason, perhaps more important, is that many parents believe cold turkey CIO will psychologically damage the baby. There is absolutely no evidence for this. To the contrary, the evidence suggests that babies sleep trained with unmodified extinction get good quality sleep. Parents report better sleep for themselves and their children. Overall there are only positive results for the family.

extinction 2
The reasons for the disappearance of a perfectly good sleep training method appear to be cultural, rather than scientific.

One more cultural factor deserves mention: consistency. Probably the number one reason why any of the extinction methods fails is that parents cannot or will not stick to the plan consistently. Consistency is the number one most important feature of any sleep training method, whether you are doing scheduled awakenings, bedtime fading, or the “Sleep Lady Shuffle” (which is, to repeat, extinction with parental presence).

To Extinguish This Line of Argument

  • The bottom line is that extinction methods are effective.
  • Extinction methods do not harm your baby or your bonding with her
  • Consistency, consistency, consistency. This is the essential piece of solving any sleep problem

Finally, let us explore the true meaning of life extinction with the help of our panel of experts, John Cleese and Michael Palin.

Cry It Out is OK, Say Researchers

So it turns out that Cry It Out is ok, after all!

I believe that no major sleep expert actually recommends “cry-it-out”. Nevertheless, the term has become attached to a method formally called “graduated extinction“. The method is made most famous by Dr. Richard Ferber in his 1986 classic “Solve Your Child’s Sleep Problems”.

Graduated extinction was criticized by proponents of “Attachment Parenting” styles of sleep training, particularly William Sears. In “The Baby Sleep Book”, Sears makes frequent reference to the relationship between baby crying and elevated levels of the stress hormone cortisol. Lots of crying leads to cortisol going up.  Sears says this leads to all kinds of negative outcomes for baby. For example:

“Science (sic) tells us that when babies cry alone and unattended, they experience panic and anxiety. Their bodies and brains are flooded with adrenaline and cortisol stress hormones. Science has also found that when developing brain tissue is exposed to these hormones for prolonged periods, these nerves won’t form connections to other nerves and will degenerate. Is it therefore possible that infants who endure many nights or weeks of crying it out alone are actually suffering harmful neurological effects that may have permanent implications on the development of sections of their brain?”

It was a rhetorical question. Sears goes on to say that cortisol damages babies brains in all kinds of ways. In extreme cases, I will grant that chronic stress is bad for babies. But the “research” Sears cites has nothing to do with crying at bedtime. No one had ever measured levels of stress hormones at bedtime. Furthermore, no one had tested whether different sleep methods stressed out babies more than other sleep methods. No one had asked whether a particular sleep method damaged (or helped) develop a bond between baby and mother.

Until now.

New Research Says Cry It Out is OK

Investigators in Australia asked mothers in pediatric clinics if their babies had a sleep problem. The doctors recruited 43 babies between 6-16 months of age. They divided the babies randomly into three groups. One was assigned to try “graduated extinction” (AKA “cry-it-out”).  Another group would try “bedtime fading“. This is a technique in which you move bedtime later in the evening to match the child’s “internal” bedtime clock, then gradually move bedtime back to your desired time. For example, if you want your baby to go to bed at 8:30, but she usually does not fall asleep until 10:30, choose 10:30 as his temporary bedtime. Then move bedtime gradually earlier.

A third group in the Australia study was handed a list of sleep instructions. This was the “control group” for comparison to the graduated extinction and the bedtime fading group.

Need a SLEEP COACH?

During the study, investigators had the parents swipe a cotton swab on the inside of their babies’ mouths to measure the hormone cortisol. Motion sensors were given to parents in the study to measure baby’s wakefulness (actigraphy). The parents also filled out questionnaires to measure their own stress and anxiety. Finally, the mothers and babies took a standardized behavior test to measure their level of attachment.

Cry It Out is OK with Respect to Sleep
cry it out is ok

Compared with the control group, babies taught with graduated extinction and bedtime fading took less time to fall asleep (sleep latency, panel A). They also woke up at night less often (panels C). If these babies did wake up, they stayed awake fewer minutes than the control group (panel B).

You can see that the control group showed some improvements across the board. This suggests that sleep education helps. It also could mean simply that babies sleep better as they mature. This is why it is so important to include a control group. To study an intervention like a sleep method, you need to know what the “background” improvement is. Otherwise you’ll never really know if your intervention works.

Babies in the graduated extinction group got more hours of sleep in the early stages of the study. However, these gains appeared to disappear by the end (panel D).

Cry It Out is Ok? What About Stress?

cry it out is ok 6
The central issue in the Australian study was stress. Does graduated extinction really raise cortisol? Does the method damage the relationship between mother and baby? The answer was ‘no’.

Other researchers have shown that minor stressors can elevate cortisol levels in newborns; But this goes effect goes away at about 4 months of age. The Australian study only looked at babies older than 6 months. So you might think that this study misses an important point that Dr. Sears and others were trying to make.

However, the investigators did something clever to get around this limitation. Sears and others insinuate that cry-it-out sleep methods cause chronic increases in cortisol. These investigators measured cortisol twice, once in the morning and once in the afternoon. They saw no difference among groups. This suggests that cry it out is ok from the standpoint of stress as well.

So Cry It Out is OK for Baby: What About for Mom?

Mothers in both intervention groups reported less stress than mothers in the control group. You see this in panels C and D in the second figure I show here. The researchers also looked at mother-baby bonding one year after the study was over. They found no differences in the ability of moms and babies to bond with one another. The levels of attachment between the two appeared to be identical in all groups. Finally, the babies in the cry-it-out groups had no more emotional or behavior problems than the control babies.cry it out is ok 7

This last finding was similar to a much larger study published in 2012. That study, also done in Australia, involved 326 children. These investigators found that sleep method had no effect on attachment, bonding, and behavior. This study was five years after the intervention, suggesting that there was nothing was messed up permanently by allowing your baby to cry at bedtime.

Limitations: Maybe Cry It Out is OK for Some, But Not for All?

The investigators acknowledge that their study has some problems. For one thing, it was relatively small. There were only 43 babies involved. With small studies like this, there’s a chance that you cannot generalize the results to all babies. This includes your own baby! That is why the investigators were careful to compare their results to other similar studies. We can believe in your conclusion more strongly if you do your study in a slightly different way, but come up with the same answers.

They also noted that they didn’t measure cortisol continuously, or even at night when the crying was happening. It’s possible that cortisol does indeed go up in a crying baby. It would be surprising if it didn’t! But the study was looking at chronic elevations. That is, cortisol stays up over time. This, after all, is the accusation that Sears and others hurl at the Ferber method. Here, the investigators have a point: sleep method does not change cortisol long-term.

Implications

The investigators conclude from these data that sleep education alone may not be enough to get your baby to sleep. You may need an intervention if your baby has a sleep problem beyond 6 months of age. They propose an interesting combination of methods: Their results suggest that bedtime fading could be done first. Then graduated extinction could be used to reduce nighttime wakings.cry it out is ok 8

Conclusions

  • Graduated extinction (cry it out) and bedtime fading are both effective.
  • Extinction-based methods do not stress out a baby chronically
  • These methods do no have long-term effects on a baby’s emotional life or her attachment to her mother.

My suggestion: find a sleep method that works for you and don’t worry about it. Show your baby a lot of love and attention and you don’t need to worry about bedtime. Above all, stay consistent: find a routine that works for you and stick with it!

If you need help doing that, go ahead and contact me. I can help.