How to connect the dots on sleep training
*First author is Angela Braden, journalist at Science Mommy
Mainstream parenting media are asserting once again that the cry-it-out sleep paradigm is harmless to babies—this time in the form of a two-paragraph morsel as one of the “sleep myths” Parents magazine “sets straight” in “Rest Assured” (July 2014 issue). The myth is listed as “crying it out is bad for your baby” and goes on to conclude that au contraire, “whatever sleep training method feels most comfortable for you is just fine.” Never mind how the baby feels. “Just fine”? Yikes! Parents typically does an excellent job educating and supporting parents to raise healthy, happy kids. But alarm bells went off for us when we read this lapse.
Fortunately, most parents will feel decidedly uncomfortable leaving their baby to “cry-it-out,” since their natural response is to soothe and keep a baby calm. In fact, methods that leave the baby to cry are less effective in the home than is documented in sleep labs.
Unfortunately, over 2 million Parents readers have just been told that leaving babies to cry to the point of distress and beyond—to the point of potential neurological damage (Lyons, 2000)—has been proven safe and even that it’s proper childrearing (See prior post. For more studies showing the damaging effects of elevated cortisol, see Lyons, 200, Bremner, 1998, and MecEwen, 2003). It does this by ending with the prolific, misconception that has justified this practice for decades: “[Your baby] needs to learn the important lifelong skills of self-soothing and falling asleep on his own.” Nothing could be further from the truth for a baby.
As Parents magazine did in this case, media reports notoriously and misleadingly back up cry-to-sleep (CIO) advice with a single flawed study. In this case, the editor approved conclusions that crying-it-out is safe based on a study of babies who didn’t, in fact, cry-it-out—not in terms of what all the major sleep-training books recommend or the common understanding of the term.
This Parents piece exemplifies the glaring mistakes made regularly among reporters on cry-it-out as well as sleep training generally. Such failures lead parents to make decisions based on misinformation. Worse, these reporting failures lead our society at large, including non-parents, to think it’s “just fine” to leave babies in distress. This unscientific attitude is bad for us all—regularly or intensely distressed babies grow into unhappy and stress-reactive (inflexible, self-focused) adults that we all have to live with (Read: Gerhardt, 2005).
Here, we outline parenting media’s top factual and logical failures when reporting on this developmentally risky practice.
Points in Brief (read details below):
1. Research does not support what sleep training reports assume:
- That sleep training “is fine” for baby —it’s NOT, even when the baby stops protesting.
- That cry-it-out (extinction of crying) works at home—it does NOT.
- Age and development level of the child does not matter for sleep training. WRONG.
- Research shows no harm for babies—NOT: it doesn’t even examine harm to babies!
- Sleep training research is well done—NOT: poor designs, fidelity, analysis show it is UNRELIABLE.
2. Sleep training reports gloss over the trauma and toxic stress that is done to babies during sleep training, when the brain and body are developing rapidly. Reports overlook how ignoring a baby at night is a form of NEGLECT.
3. Most parents are not comfortable with cry-it-out (extinction of crying) and they should not be. Babies are meant to be with caregivers all the time. Their wellbeing is undermined otherwise.
Here is more detail:
1. Parenting media fail to define “cry-it-out”:
To most people, the term “cry-it-out” has come to mean leaving the baby in the crib until crying stops, however long that takes—a “total extinction” (the scientific term) approach.
In the study cited (Price et al., 2012), for example, parents were actually taught different approaches to train their child to sleep:
· “controlled-crying” (intermittently offering comfort)
· “camping out” or “fading” (staying with and comforting the baby, gradually less)
Fading, in particular, is a far cry from the total extinction experience for the baby (see #5 below on infant trauma).
Yet, despite a flawed, misleading study, millions of Parents readers have been told unequivocally, it is safe to let your baby “cry it out” (total extinction). Otherwise, it would be referred to as just “crying” without the “it out” qualifier.
When parenting media outlets gloss over the nuances of sleep training, which understandably, are difficult to address thoroughly in magazine format, they send parents with full confidence to using total extinction (crying-it-out).
A more balanced approach would include citing the dangers of crying-it-out along-side the flawed study conclusions. The dangers of extensive infant distress are multiple and long term.
2. Parenting media fail to actually read the study:
Though they would have been wise to point out that the cited study didn’t look at crying-it-out, or total extinction, we can’t blame Parents magazine entirely for their erroneous reporting. The study authors themselves stated that “behavioral methods” for sleep training could be used with confidence. A journalist would have to actually look at the methodology of the study to see that the researchers were using that term broadly, loosely and without scientific merit.
If they did take a peek at the study design, though, they’d be confronted with an even bigger problem: the intervention and control groups had no discernible differences, making it impossible to really measure anything.
The researchers report that they used “Intent To Treat” (ITT) in their study design. This means they have no idea what the parents in either group ultimately did or didn’t do in terms of sleep training. They assume their intervention group followed their proposed sleep-training “treatment,” at least more often than not. They compared apples to apples.
Here’s how: Their intervention group consisted of parents who were given brief instructions by nurses trained by the researchers in controlled-crying and camping out or fading interventions, the intervention was similar to what happens naturally. Their control group, for comparison, got advice from nurses who the researchers didn’t train in these interventions. (Did you have to reread to catch the difference?) In other words, the control group had normal care, in which the parents were “free to ask for sleep advice” and nurses were free to give the advice they normally would. They were also free to follow the advice of their friends, neighbors, or baby sleep sites on the internet. No wonder the two groups had no significant differences in outcomes, as the study’s findings freely admit: “Behavioral sleep techniques have no marked long-lasting effects (positive or negative)” (Price, 2012).
Amazingly, the researchers draw their “use with confidence” conclusions even though their control group could very well have included total extinction, skewing the results in favor of the intervention group, which was at worst instructed to intermittently soothe their babies. At best, many of the parents in both groups did what their instinct told them to and ignored the nurse’s sleep-training advice.
“Intent To Treat” research design does not make sense when what’s being measured is likely common in the control group as well. It stands to reason, given that all the study participants had reported sleep problems and controlled crying is typically advised in Australia (Gethin & Macgregor, 2009), that the control-group parents would have asked for sleep advice and that the advice they received would be similar to that given to the intervention group. Bottom line: we’ll never know which babies experienced crying-it-out, if any, so how can the researchers possibly justify advising parents accordingly?
In all fairness, journalists generally consider it their job to communicate experts’ voices, so indiscriminate parroting of a study authors’ conclusion is somewhat understandable. However, it would be in the spirit of journalistic accuracy to recognize (again, by looking at the methodology) that the researchers might be overstating their conclusions.
Journalists need to be scientifically literate and educate parents to be also. This would be a real service to their audience.
3. Parenting media fail to account for age and developmental differences:
A journalist who reports on research interventions with children (anyone under age 25), should attend to the age of the children studied because there are different effects of treatment based on age. For example, when doling out sleep training advice to parents, it’s important to note that in the studies that claim sleep training is harmless, babies were 8-10 months old when they may or may not have been left to cry-it-out. But many science reporters and bloggers go ahead and apply the study “results” to “babies” in general, sending parents of far younger, more vulnerable babies running to the sleep training manuals. There, they are instructed to use total extinction. The second author has received many emails from parents who realized later the damage they did when trying cry-it-out with a weeks-old baby.
Journalistic ethics warrants much more careful reading of research reports. This happens in some areas of study but is woefully inadequate for research of risky parenting practices (e.g., baby isolation, no breastfeeding, stranger daycare). More about expert sleep training mistakes here.
If journalists read more studies, they’d see that a better approach would be to educate parents on the natural needs and expectations of babies to be with their caregivers for reasons of growth and comfort.
4. Parenting media fail to recognize the pervasive influence of behavior modification:
Those of us who work directly with real parents in the throes of the first-year and have recently been there ourselves can attest to the prevalence of behavior modification generally and total extinction sleep training specifically. Extinction, either partial (i.e., controlled crying) or total (cry-it-out), is widely recommended by pediatricians and prominent authors. In fact, the American Academy of Sleep Medicine classified behavioral techniques, which includes total extinction, as standard practice for managing “infant sleep problems.”
- All of the infant sleep science upon which behavior modification-based advice is biased toward “effective” intervention. Effectiveness is based on whether the baby stops crying.
- Sleep training studies are not based on what sleep behaviors are common or developmentally normal. The fallacy is that infant sleep variability (which is normal) is defined as abnormal.
Behaviorism in general has been largely thrown out of psychology circles with regard to normal human beings, because it treats humans like machines. Babies are not machines. They are wonderfully complex, sensitive, dynamic beings who rely on parents’ affection and soothing to develop fully. They need their caregivers nearby. Yet many infant sleep scientists still cling to behaviorism. Then, their act is followed by pro-sleep training studies with invalid conclusions and journalists that extrapolate from there to assert crying-it-out as harmless.
At the same time, the media routinely downplay the role of behaviorist thinking, assuming that parents will use good judgment in carrying out sleep training methods. In doing so they miss that an integral part of the behaviorist approach—the extinction part—specifically requires parents not to use their judgment. For total extinction, they must never “give-in” in order to make their efforts effective in conditioning their baby not to cry. This is the approach the majority of sleep-training books on the parenting shelf promote, to which Parents readers may now turn to in droves.
But even in the loving hands of parents with otherwise good judgment and instincts, the popular baby sleep manuals instruct based on the behavioral conditioning component of cry-it-out. Some of the books start out dealing with gentle ways to encourage sleep; some advise partial extinction or controlled crying; but with few exceptions, they land on: just don’t go back into the nursery . . . no matter how desperate the screaming becomes or how long the baby’s panic state lasts: total extinction.
As an alternative to behaviorist thinking, the second author uses developmentally, neurological, and evolutionarily appropriate baselines for child development to inform parenting advice (see below). And the first author consults parents struggling with infant sleep and soothing with these baselines as a guide. There are gentle alternatives to get the sleep parents need. The Happiest Baby on the Block is a solid starting point.
Problem: “Controlled crying” isn’t controlled
We also see many prominent bloggers and parenting journalists with a ‘what’s-the-big-deal?’ attitude toward cry-it-out concerns. They argue along the lines of “so I let my nine-month-old cry for 10 minutes; I doubt it will scar him for life.” The problem with using this logic to support cry-it-out practices and advising other parents to follow suit is this describes the minority experience, leaving most babies at risk for a more traumatic experience (Middlemiss, 2012).
The dangers apply to many of the controlled crying sleep-training methods as well, since the younger or more sensitive temperamentally the baby is, the faster and higher the stress level will rise when suddenly left alone to sleep, and the greater potential damage done to the child’s developing brain (see #1 above and #7 below) and the parent-child relationship. While a mommy blogger’s older, less sensitive baby only fussed for a few minutes and went to sleep, your baby may not fare so well. Many of the parents we’ve consulted had endured hours of crying for more than a week before giving up. And many babies will reach full panic-mode long before 10 minutes (a common interval in which to offer comfort in most controlled crying methods). Either way, the behaviorist-inspired protocol is the same: don’t intervene—let them cry it out.
And so, media that make light of the cry-it-out experience by lumping it in with mild fussing for a few minutes are steering parents toward the dangers of total extinction, a recipe for toxic distress (see #7 below).
A similar dynamic is often set in motion in the pediatrician’s office. While more pediatricians are progressing beyond the behavioral model of child-rearing in general, too many hold onto it for the purpose of sleep-training (see Schore, 2005, who examines how developmental neuroscience should inform the field of pediatrics). This is because they’ve traditionally looked exclusively to the published (limited) findings of the infant sleep scientists who conclude that extinction is effective (at stopping the crying, not the distress). They then assure parents, per the researchers in the Australian study, that “crying-it-out has been proven safe.” Here again, parents are told when it comes to “helping” your baby sleep through the night, whatever you do is “just fine.” Then they go do “whatever” with no regard to intensity, frequency, age or temperament of their baby. The results are more than likely not at all “fine.”
5. Parenting media fail to understand trauma from an infant’s perspective:
In light of developmental neuroscience, the advice parents get from the baby-sleep-training instruction books is:
Risky for babies under six months, whose nervous systems are calibrating set points for life (Caldji, 2000).
Damaging for babies at risk for an attachment disorder, conservatively estimated at 40% of babies, which puts them at risk for mental illness.
Ill-advised for all babies who experience intense levels of panic from natural and healthy instincts that compel them to stay close to parents when sleeping.
Dangerous for all the babies who don’t just fuss for a couple of minutes but go into full hyper-arousal and then dissociative withdrawal. This is a common response for a distressed infant (Perry, 1998).
The cry-it-out advice is risky because in these cases, being left to cry is a trauma.
Being left alone at all is stressful for baby mammals (Levine, 2005). Their biological systems become disorganized when separated from caregivers because they have no sense of safety apart from adults. When their distress calls (cries) are then ignored they instinctually panic—their lifeline is gone. (More on the brain’s Separation distress neuro-circuitry.) Once we understand that babies are operating from a survival instinct-dominant, immature brain with limited ability to rationalize, the plentiful trauma research clearly applies. Infants can experience PTSD, toxic distress, depression and dissociation in response to crying-it-out.
This is how distress (signaled by crying) becomes trauma.
6. Parenting media fail to recognize the effects of neglect on a continuum:
When concerns over total extinction sleep-training are refuted, we almost always see this line of logic:
‘Since much of the research we have on the devastating effects of infant trauma comes from studies of babies who were subjected to extreme neglect, the outcomes of those studies don’t apply to babies in loving, intact homes.’
Essentially, the argument contends that we can’t extrapolate risks from infant trauma research because, apart from sleep times, extinction-sleep-trained babies are given quality nurturing and responsive care. One needs empirical evidence to verify this and to date there is none. Quite the opposite has been shown—the effects of under care (not abuse) are measurable (Bugental, 2003).
It is true that caregiver responsiveness to the child’s needs generally can mitigate risks for psychological disorders—we parents don’t have to be perfectly responsive at all times. We may misread a hunger cue and offer a toy instead, for example; but when the baby continues signaling we typically try something else until we figure out what he needs.
But make no mistake: The total-extinction experience is so far outside the range of what would be a normal break in responsive care that it’s likely to have some of the same detrimental effects research has linked with extremely neglected babies—only to a lesser extreme. Think of it as a neglect continuum. Is that really where we want to set the bar for our most vulnerable–young children?
Trauma alters normal self-development of critical circuits in the right hemisphere that contribute to emotional regulation and social capacities. These circuits are forming rapidly in the first year of life and require caregiver guidance to develop properly (Schore, 1997; Schore 1996). Read about more on effects here and here.
Therefore, however otherwise limited in terms of duration and frequency trauma may be within a healthy, loving family, it’s still trauma, no matter where the baby is. Trauma is trauma. The baby doesn’t know his devoted parents are holding each other in agony outside the nursery door as they follow the advice of a misguided parenting manual that a major magazine article said was “just fine.”
Regardless of where on the neglect continuum any baby subjected to crying-it-out may fall (depending on other environmental and genetic risk factors), the neurological effects of a single high-stress experience are even harder to ignore.
7. Parenting media fail to understand and convey the dangers of toxic stress:
As discussed in previous posts, one of the neurological risks of extreme distress to the developing brain comes from the excess cortisol released. This has been shown to unhook neurons during synaptogenesis—the time when the brain is wiring itself based on experience (McEwen, 2003) (Perry, 1997). Sounds like a legitimate concern, right?
But a common media push-back on this concern counters that we needn’t worry about damaging these sprouting connections, because brain-cell death is a “normal part” of synaptogenesis. Such logic demonstrates a staggering misunderstanding of brain development, yet we see even journalists who are medical doctors using statements like this to dismiss cortisol concerns during CIO.
Specifically, what’s missed here is that the baby’s brain is exquisitely sculpted in a use-dependent fashion. During synaptogenesis, the brain is indeed busy pruning some connections and beefing up others at an extraordinary rate, but this critical process, which will never be repeated for the rest of the child’s life, is anything but random. Based on the baby’s bio-chemistry, which is influenced by high-quality responsive care and sensory input—primarily emotional—neurons compete for connections. Therefore, the infant brain determines from experience—mostly interpersonal experience—which connections are most important and which to prune. This happens billions of times per minute throughout infancy, shaping the brain to precisely fit its unique social environment—for better or worse.
On the other hand, what neuroscientists have termed “over-pruning,” resulting from excess stress or chronic cortisol elevation is random. It wipes out neural connections indiscriminately. In fact, one effect we know empirically from over-pruning is a smaller hippocampus, a major brain structure with myriad functions (Bremner,1998). This is just one example of damage from excess cortisol that can not accurately be asserted as a “just fine” event. It’s undeniably not beneficial to have a shrunken and less connected hippocampus (it’s related to more depression, poor memory, and additional cognitive problems.)
At the same time, extreme distress during infancy—particularly chronic, prolonged fear states—have been associated with an enlarged amygdala, a feature of a brain that suffers from chronic anxiety and a hair-trigger fear response (Mehta, 2009, Schore 2009). In fact, some evidence suggests that an enlarged amygdala accounts for features of Autism Spectrum Disorder(ASD), like emotional dysregulation, failure to orient to faces, and poor eye-contact (Schore, 2013).
Leaving babies alone to cry is not part of “normal” human development from an evolutionary perspective. (For more details on the long term effects see Levine, 2005.) When we do this to other mammals in experiments they become abnormal specimens of their species (Harlow, 1958). And humans are even more sensitive to caregiver treatment than any other animal.
What is the bottom line for parents?
We hope parents reject the idea that crying-it-out is either healthy or necessary. We hope parents become aware of the legitimate risks for sleep training to escalate to a traumatic situation for baby, and make a truly informed decision. We whole-heartedly support a tuned-in parent’s decision to gently assist their children toward sleep arrangements they believe are best for their whole family. But we hope that parents will listen to their babies with their hearts open and their free-thinking minds engaged, rather than force themselves to carry out what feels exquisitely painful and seems obviously wrong because the experts said it was “just fine.”
What is the bottom line for parenting media journalists?
We are advocating for accuracy and balance in reporting so parents can make decisions based on good information. Read the studies, think critically, and remember, the wellbeing of millions of helpless little ones is at stake.
*Angela Braden is a parenting journalist and parent education consultant for ZERO TO THREE* who began researching developmental neuroscience during her daughter’s first year of life. As a result of her studies, while immersed in daily and nightly care of her own sleep-challenged baby, Angela created developmentally-appropriate strategies to help parents thrive without resorting to cry-it-out sleep training. Ask Angela a sleep question on her Facebook page.
*Opinions expressed are authors’ own.
Bremner, J.D., (1998). “The Effects of Stress on Memory and the Hippocampus Throughout the Life Cycle: Implication for Childhood Development and Aging. Developmental Psychology. Fall 10 (4): 871-85.
Bugental, Daphne Blunt; Martorell, Gabriela A.; Barraza, Veronica (2003). “The hormonal costs of subtle forms of infant maltreatment.” Hormones and Behavior, Vol 43(1), Jan 2003, 237-244. doi: Retrievable at 10.1016/S0018-506X(02)00008-9
Caldji, Christian, Diorio, Josie & Meaney, Michael. (2000). “Variations in maternal care in infancy regulate the development of stress reactivity.” Biological psychiatry, Dec 15;48(12):1164-74.
Lyons et all. (2000). “Stress-Level Cortisol Treatment Impairs Inhibitory Control of Behavior in Monkeys.” J. Neuroscience. October 15. 20 (20):7816–7821.
Gerhardt, Sue. (2004). Why Love Matters: How Affection Shapes a Baby’s Brain. United Kingdom: Brunner-Routledge.
Gethin, A & Macgregor, B. (2009). Helping Baby Sleep: The Science and Practice of Gentle Bedtime Parenting. New York: Crown Publishing Group.
Harlow, H. (1958).The nature of love. American Psychologist, 13, 673-685.
Levine S. (2005). “Developmental determinants of sensitivity and resistance to stress.” Psychoneuroendocrinology. Nov;30 (10):939-46.
McEwen, B. S. (2003). “Early life influences on life-long patterns of behavior and health.” Mental Retardation and Developmental Disabilities Research Reviews, 9(3), 149–154.
Mehta, Mitul A, et all. (2009). “Amygdala, hippocampal and corpus callosum size following severe early institutional deprivation: The English and Romanian Adoptees Study Pilot.” Journal of Child Psychology and Psychiarty. APR. V. 50: 943–951.
Middlemiss, Wendy, et All. (2012). “Asynchrony of mother-infant hypothalamic-pituitary-adrenal axis activity following extinction of infant crying responses induced during the transition to sleep.” Early Hum Dev.2012 Apr. 88(4):227-32. doi: 10.1016/j.earlhumdev.2011.08.010. Epub 2011 Sep 23.
M.R. Gunnar. (1998). “Quality of care and buffering of neuroendocrine stress reactions: potential effects on the developing human brain,” Prev Med 27, no. 2 (Mar–Apr): 208–11.
Perry, B.D. (1997) Incubated in Terror: Neurodevelopmental Factors in the ‘Cycle of Violence’ In: Children, Youth and Violence: The Search for Solutions (J Osofsky, Ed.). Guilford Press, New York, pp 124148.
Perry BD & Pollard R. (1998). “Homeostasis, stress, trauma, and adaptation: a neurodevelopmental view of childhood trauma,” Child and Adolescent Psychiatric Clinics of North America 7, 33-51.
Price, Anna M. et all. (2012) “Five-Year Follow-up of Harms and Benefits of Behavioral Infant Sleep Intervention: Randomized Trial.” Pediatrics September online.
Schore, Allan (1996). “The experience-dependent Maturation of a regulatory System in the Orbital Prefrontal Cortex and the Origin of Development Psychopathology.” Development and Psychopathology 8:59-87.
Schore, Allan (1997). “Early Organization of the Nonlinear Right Brain and Development of a Predisposition to Psychiatric Disorders.” Development and Psychopathology 9: 595-631.
Schore, Allan (2005). “Attachment, Affect Regulation and the Right Brain: Linking DevelopmentalNeuroscience to Pediatrics.” Pediatrics in Review 26 (6) June.
Schore, Allan (2009). “Relational Trauma and the Developing Right Brain. An interface of Psychoanalytic Self Psychology and Neuroscience.” Annual of the New York Academy of Sciences, 1159, 189-203.
Schore, Allan (2013). “Regulation Theory and the Early Assessment of Attachment and Autistic Spectrum Disorders: A Response to Voran’s Clinical Case.” Journal of Infant, Child, and Adolescent Psychotherapy, 12:3, 164-189.
NOTE on BASIC ASSUMPTIONS
When I (Narvaez) write about human nature, I use the 99% of human genus history as a baseline. That is the context of small-band hunter-gatherers. These are “immediate-return” societies with few possessions who migrate and forage. They have no hierarchy or coercion and value generosity and sharing. They exhibit both high autonomy and high commitment to the group. They have high social wellbeing. See comparison between dominant Western culture and this evolved heritage in my article (you can download from my website):
Narvaez, D. (2013). The 99 Percent—Development and socialization within an evolutionary context: Growing up to become “A good and useful human being.” In D. Fry (Ed.), War, Peace and Human Nature: The convergence of Evolutionary and Cultural Views (pp. 643-672). New York: Oxford University Press.
When I write about parenting, I assume the importance of the evolved developmental niche (EDN) for raising human infants (which initially arose over 30 million years ago with the emergence of the social mammals and has been slightly altered among human groups based on anthropological research).
The EDN is the baseline I use for determining what fosters optimal human health, wellbeing and compassionate morality. The niche includes at least the following: infant-initiated breastfeeding for several years, nearly constant touch early, responsiveness to needs so the young child does not get distressed, playful companionship with multi-aged playmates, multiple adult caregivers, positive social support, and soothing perinatal experiences.
All these characteristics are linked to health in mammalian and human studies (for reviews, see Narvaez, Panksepp, Schore & Gleason, 2013; Narvaez, Valentino, Fuentes, McKenna & Gray, 2014; Narvaez, 2014) Thus, shifts away from the EDN baseline are risky. My comments and posts stem from these basic assumptions.