We’ve all seen the warnings that sleep training causes so much stress to a baby that cortisol floods the brain, killing neurons and altering development.
Even without these alarming stories, most parents considering sleep training naturally worry about how stressful it is to a baby. None of us like to hear our babies cry. It makes us feel stressed, and the baby probably feels the same way. But how stressful is it? And is it damaging to a baby’s brain?
Despite decades of research on sleep training, most studies have focused on outcomes related to sleep and daytime behavior, but few have examined babies’ stress responses to this change. Those that warn that letting babies cry is damaging to their brains cite studies of babies that were subjected to chronic neglect or abuse or raised in orphanages and lacking strong attachment figures. These are examples of chronic, toxic stress. They are deeply saddening to me, but I’m not convinced that they tell me, or any other loving parent, much about the effects of sleep training my little girl.
This has left me digging through reams of research, trying to put sleep training in perspective among other sources of infant stress. A good place to start is the American Academy of Pediatrics’ (AAP) recently released report entitled The Lifelong Effects of Early Childhood Adversity and Toxic Stress  and an accompanying policy statement. The AAP report gives us a framework for looking at stress that I think is very useful. It defines 3 types of stress responses in children:
- A positive stress response is one that is “brief and mild to moderate in magnitude” and in which a caring and responsive adult helps the child cope. Events that can cause positive stress responses include “dealing with frustration, getting an immunization, and the anxiety associated with the first day at a child care center.” The authors go on to say, “When buffered by an environment of stable and supportive relationships, positive stress responses are a growth-promoting element of normal development. As such, they provide important opportunities to observe, learn, and practice healthy, adaptive responses to adverse experiences.”
- A tolerable stress response occurs because of non-everyday events like a death in the family, divorce, or a natural disaster. Again, what makes this stress tolerable is a child’s relationship with a supportive adult, who can help the child adapt and cope with the changes in his life. In the best of circumstances, tolerable stress can even have positive effects .
- A toxic stress response is caused by “strong, frequent, or prolonged activation of the body’s stress response systems in the absence of the buffering protection of a supportive, adult relationship.” Examples include chronic stressors such as child abuse or neglect, parental substance abuse, and maternal depression. In early childhood, toxic stress can affect brain circuitry and disrupt the development of normal physiologic stress regulation. It can also compromise immune function and cause inflammation, both of which have been linked to a number of chronic diseases.
Given this framework, can we predict what type of stress response sleep training produces in a baby? Is sleep training more like starting daycare, coping with a divorce between the two most important people in your life, or being raised by someone abusing drugs?
In the vast majority of families, sleep training is nothing like the sad situations that cause toxic stress. Most parents find that sleep training takes only a few nights, or if done more gradually, no more than a couple of weeks. If it causes weeks of prolonged crying, then something isn’t working, and parents need to find a new strategy.
This is a good time to add that I define sleep training broadly. I would rather call it “sleep learning,” but I’ll stick with the terms we all know. To me, sleep training involves making changes in a baby’s sleep habits so that he learns to self-soothe, and it almost always involves decreasing parental involvement and some crying. I know that many readers balk at the mere mention of sleep training, but this discussion of change and stress may be helpful even to those considering gentle ways of shaping children’s sleep – such as transitioning to a separate bed or night weaning.
Learning to fall asleep in a new way isn’t easy. But just as we can’t start daycare for a baby, we can’t learn to sleep for a baby either. Stress is unavoidable, even for a baby born to the most caring family. My intention in this research and writing is most definitely not to give a stamp of approval to all sleep training methods. Rather, it is to move beyond blanket judgments, put the stress of sleep training in context, and to think critically about what we can do to support and respect babies as they learn to sleep so that it is a positive, not toxic, experience.
How do babies respond to stress?
When we experience stress, several physiological pathways kick into action. One response is the activation of the hypothalamic-pituitary-adrenocortical (HPA) axis, resulting in the release of cortisol. Cortisol shifts the focus of the body towards mobilizing energy and suppressing some functions such as immune responsiveness, which we can survive without on a short-term basis. Cortisol is essential for survival. It allows us to cope with and recover from stress.
Cortisol can be measured in saliva, so it can be sampled in a non-invasive manner without causing additional stress, even while a baby is sleeping. Without unusual levels of stress, cortisol has a natural circadian rhythm, being 300-400% higher in the early morning than at its lowest point around midnight . This pattern does not develop until infants are several months old, and it is what helps them to sleep at night and be awake during the day. A surge in cortisol outside of this normal pattern can indicate that a baby is stressed. However, it is simplistic to assume that any rise in cortisol is dangerous, given its importance to daily function. Cortisol is not a problem unless it is elevated for extended periods of time as in the case of chronic stress . Further complicating matters, children with a history of toxic stress caused by abuse or neglect often have lower cortisol and blunted cortisol rhythms compared to the norm .
Babies have a very reactive HPA axis at birth. A newborn baby will have a strong cortisol response to a heelstick, a bath, or a physical exam . In one study, newborns were given a mock exam two days in a row. On the first day, the babies had increased cortisol and cried during the exam. However, when the exam was repeated on the second day, the babies did not show an increase in cortisol, though they still cried almost as much as they had on the first day . Somewhere in their brains, they recognized that this experience had happened before. This study illustrates a positive stress response. The babies were initially stressed by the exam, but they were able to recover and learn from the experience so that they could cope with it effectively the next day. It also demonstrates the important point that crying is not always accompanied by increased cortisol.
Other studies have tested infants’ responses to acute physical pain, such as a vaccination. Physical pain causes a cortisol response in babies younger than 6 months, but beyond this age they show very little cortisol response, though they still cry just as much . It has also been shown that anger, fear, and novelty stressors (such as the first time in a swimming pool) rarely cause a cortisol response in older babies .
Importantly, a mother or other attachment figure is usually present in the experiments described thus far. During sleep training, we are often trying to reduce parental presence and soothing around bedtime, and this is what makes it hard on babies. What do studies tell us about the stress response to sleep training or other parental separations?
There is one study of cortisol responses in babies undergoing sleep training . 25 mothers and infants (4- to 10-month-olds) spent 5 days at a residential facility to participate in extinction sleep training (CIO without any reassurances). On the first night of sleep training, all babies had at least 2 bouts of 5-10 minutes of crying before falling asleep. On the third day, the babies all fell asleep with no more than a little fussing. Cortisol was measured in the infants before beginning the bedtime routine and at the time that they fell asleep, after crying-it-out. On both the first and third days of sleep training, there was no increase in cortisol between the pre-bedtime sample and the CIO sample.
At first glance, you might think that this study provides evidence that sleep training is not stressful to babies. However, the study had some serious limitations. First, the standard deviations for these data (error bars in the graph) are HUGE. In addition, the babies’ cortisol levels were high all around, even before the bedtime routine, probably because they were in a strange place. Have you ever traveled with a baby and not found that their sleep routine was disrupted? That higher basal stress could hide the stress that the babies may have felt about crying it out. Still, given that these babies were in pretty much the most stressful sleep situation imaginable (poor kiddos), crying it out didn’t exactly cause their bodies to be flooded with additional cortisol.
How stressful might sleep training be to a baby in his own bed, with periodic reassurances from a parent? We can only speculate, but so far, I’m not convinced that it is as devastating as some would have us believe. What we really need is a study that has not yet been done: a longitudinal study of cortisol in babies that are sleep trained in their own homes, with age-matched controls. I think that such a study would ease our fears about sleep training. If I had to guess, I would predict that age-appropriate sleep training in a familiar and supportive environment would cause a small increase in cortisol for a few nights, but not chronically.
Other studies have looked at cortisol responses in babies briefly separated from their mothers. In 9-month-old babies, being left alone for 30 minutes with a stranger in a lab setting causes a 20-40% increase in cortisol, a minor bump considering that cortisol fluctuates by 10x that much on a daily basis. However, by 12-18 months of age, most studies find that maternal separation does not increase cortisol . And again, being at home in a familiar bed rather than with a stranger in a strange environment may make sleep training less stressful.
There is one other type of stressful situation that might help us to understand sleep training: transition to day care. Like sleep training, starting day care involves a big change in routine and prolonged separation from an attachment figure. Starting day care for the first time causes an increase in cortisol, even if mom stays with baby for the first two weeks to help with the transition . And even after many months, toddlers attending full-day daycare show a gradual increase in cortisol throughout the day, whereas children at home show no increase or a gradual decline in cortisol during the day . However, across ages, the greatest daytime cortisol response is found in 2- to 3-year-old toddlers. Infants, including those around the period of peak separation anxiety at 9-12 months, do not have increased cortisol . The authors speculated that the chronic elevation of cortisol in toddlers in daycare is related to the stress of group interactions rather than the separation from parents. The fact that cortisol decreases in young children during nap time at daycare, even if they do not actually sleep, supports this hypothesis .
Despite the stress of entering childcare at a young age, it has yet to have been shown to have any long-term effects on a person’s ability to regulate stress or form healthy relationships. Thankfully, I don’t see many claims that sending your child to daycare will cause brain damage. And despite the stress of daycare, many families in the modern world still make this choice, because for them, the benefits of daycare outweigh the risks. Daycare may allow parents to pursue careers that they love and for the family to be financially secure. It might benefit the child as well, giving him opportunities to learn about social relationships in group play and stimulating cognitive and social development.
We can think about sleep training as a similar risk-benefit analysis. Yes, it is stressful, but the benefits are also substantial. For most families, sleep training will result in a baby that has a new and important skill: the ability to sleep on his own. The whole family will get more sleep, and that comes with numerous benefits. In fact, recognizing that the current sleep situation is not working and making changes to improve it could reduce a child’s long-term exposure to stress and cortisol.
Consider the following examples:
-Babies that sleep more have lower basal cortisol .
-Toddlers with more fragmented sleep have higher cortisol in the morning, and this is correlated with daycare teachers’ ratings of greater internalizing behavior problems. These children were also more likely to become upset and tearful in response to minor stressors and challenges at daycare . These last two data sets are only correlative, but a causal relationship is certainly plausible.
-When 3- to 6-month-old babies play with a sensitive mother for 15 minutes, their cortisol decreases. However, playing with an insensitive mother can actually increase cortisol, or at best, cause no change . If sleep deprivation is affecting a parent’s ability to be sensitive and responsive during the day, then that alone could increase a child’s exposure to cortisol. It is not selfish to sleep train your baby so that you can cope better with the stressors of life and be a more responsive parent.
There will always be those that judge sleep training to be unacceptable and offer the worst possible examples of childhood stress to incite fear in parents, but I believe that we have to put it in context. Stress is a part of life. It is important to recognize that making changes to a baby’s sleep routine is stressful and to minimize that stress as much as possible. However, a baby’s total stress load comes from a multitude of factors, and sleep training may actually alleviate other stressors. Telling parents that they must protect their babies from stress and do everything they can to stop a baby from crying, at all costs, may be counter-productive. Crying is a baby’s way of communicating, but it does not always communicate despair, and it is not always accompanied by a cortisol response. In fact, some studies suggest that crying may release tension and reduce the activity of the HPA axis and the cortisol response . Lovingly allowing a baby to practice coping with stress in the process of learning a new skill can be a healthy thing.
An essential part of a positive stress response is the support of a caring adult. Yet, in order for babies to learn to sleep on their own, parents often need to decrease their presence and soothing. How do we, as parents, balance our important role of being supportive of our babies during this transition while still allowing them to develop their own self-soothing abilities? That’s the question I’ll tackle in my next post.
Have you made changes to your child’s sleep habits that were stressful? How did that stress compare to other stressors in your child’s life? Did you find ways to help your child cope with the stress?
Check out other posts from my infant sleep series:
- The Cry-It-Out Controversy and My Family’s Sleep Story
- Why Sleep Matters to Babies and Parents
- The Importance of Self-Soothing to Infant Sleep (and how to support it!)
- Sleep Solutions for Every Baby
- Infant Sleep Research: Cosleeping, Self-Soothing, and Sleep Training
- 6 Little Secrets of a Sleeping Baby
1. Shonkoff, J.P. and A.S. Garner. The lifelong effects of early childhood adversity and toxic stress. Pediatrics. 129(1): p. e232-46. 2012.
2. National Scientific Council on the Developing Child, Excessive stress disrupts the architecture of the developing brain: Working paper #3. 2005. p. Center on the Developing Child at Harvard University. http://www.developingchild.net.
3. de Weerth, C., R.H. Zijl, and J.K. Buitelaar. Development of cortisol circadian rhythm in infancy. Early Hum Dev. 73(1-2): p. 39-52. 2003.
4. Gunnar, M.R. and B. Donzella. Social regulation of the cortisol levels in early human development. Psychoneuroendocrinology. 27(1-2): p. 199-220. 2002.
5. Jansen, J., R. Beijers, M. Riksen-Walraven, and C. de Weerth. Cortisol reactivity in young infants. Psychoneuroendocrinology. 35(3): p. 329-38. 2010.
6. Gunnar, M.R., J. Connors, and J. Isensee. Lack of stability in neonatal adrenocortical reactivity because of rapid habituation of the adrenocortical response. Dev Psychobiol. 22(3): p. 221-33. 1989.
7. Gunnar, M.R., L. Brodersen, K. Krueger, and J. Rigatuso. Dampening of adrenocortical responses during infancy: normative changes and individual differences. Child Dev. 67(3): p. 877-89. 1996.
8. Middlemiss, W., D.A. Granger, W.A. Goldberg, and L. Nathans. Asynchrony of mother-infant hypothalamic-pituitary-adrenal axis activity following extinction of infant crying responses induced during the transition to sleep. Early Hum Dev. 88(4): p. 227-32. 2012.
9. Gunnar, M.R., N.M. Talge, and A. Herrera. Stressor paradigms in developmental studies: what does and does not work to produce mean increases in salivary cortisol. Psychoneuroendocrinology. 34(7): p. 953-67. 2009.
10. Ahnert, L., M.R. Gunnar, M.E. Lamb, and M. Barthel. Transition to child care: associations with infant-mother attachment, infant negative emotion, and cortisol elevations. Child Dev. 75(3): p. 639-50. 2004.
11. Vermeer, H.J. and M.H. van IJzendoorn. Children’s elevated cortisol levels at daycare: A review and meta-analysis. Early Childhood Research Quarterly. 21: p. 390-401. 2006.
12. Watamura, S.E., B. Donzella, J. Alwin, and M.R. Gunnar. Morning-to-afternoon increases in cortisol concentrations for infants and toddlers at child care: age differences and behavioral correlates. Child Dev. 74(4): p. 1006-20. 2003.
13. Watamura, S.E., A.M. Sebanc, and M.R. Gunnar. Rising cortisol at childcare: relations with nap, rest, and temperament. Dev Psychobiol. 40(1): p. 33-42. 2002.
14. Scher, A., W.A. Hall, A. Zaidman-Zait, and J. Weinberg. Sleep quality, cortisol levels, and behavioral regulation in toddlers. Dev Psychobiol. 52(1): p. 44-53. 2010.
15. Karraker, K.H. and M. Young. Night Waking in 6-Month-Old Infants and Maternal Depressive Symptoms. J Appl Dev Psychol. 28(5-6): p. 493-498. 2007.
16. Essex, M.J., M.H. Klein, E. Cho, and N.H. Kalin. Maternal stress beginning in infancy may sensitize children to later stress exposure: effects on cortisol and behavior. Biol Psychiatry. 52(8): p. 776-84. 2002.
17. Murray, L., S.L. Halligan, I. Goodyer, and J. Herbert. Disturbances in early parenting of depressed mothers and cortisol secretion in offspring: a preliminary study. J Affect Disord. 122(3): p. 218-23. 2010.
18. Spangler, G., M. Schieche, U. Ilg, U. Maier, and C. Ackermann. Maternal sensitivity as an external organizer for biobehavioral regulation in infancy. Dev Psychobiol. 27(7): p. 425-37. 1994.
19. Lewis, M., D.S. Ramsay, and K. Kawakami. Differences between Japanese infants and Caucasian American infants in behavioral and cortisol response to inoculation. Child Dev. 64(6): p. 1722-31. 1993.