Over twenty years ago, when our children were born, my husband Arjun and I had the sleep debate that many American parents still have today: Where should we put the baby to sleep at night?

Arjun grew up in India, and though he'd slept apart from his parents, his ayah (nanny) had always slept close by. I, on the other hand, grew up in small-town Minnesota and had slept alone in a crib at the end of a hall. When it came to our own kids, we each argued that the other's experience must have been better. He believed in the superiority of modern, Western ideas, and I was sure that the ancient practice of sleeping together was the natural state of parenthood.

"We'll roll over on her," he worried.

"We'll sleep better," I countered, thinking breastfeeding in bed sounded pretty good.

Last week, a New York Times blog post reignited the discussion for another generation of new parents. "Sleep Training at Eight Weeks: Do You Have The Guts?" it asked. Sleep training is the process of getting a baby to sleep through the night through a variety of behavioral techniques, and in the extreme by letting a baby "cry it out" in a room without a parent's responsive soothing or feeding. After a couple of days, the logic goes, the baby "gets used to it," and "learns" to sleep alone through the night.

This school of sleep training, based on operant conditioning, runs counter to the current science of infant development. Here are a few examples:

Crying in babies is not a misbehavior to be modified; it is a physiological signal that something is wrong. Babies who are picked up when they cry learn that their needs will be met and they cry less over the long run. On the other hand, if a baby's crying is consistently ignored, she can learn that her signaling system is ineffective, undermining the developing sense of self-efficacy. Her natural demands, then, can escalate into more anxious ones. The general rule of parenting infants is that you cannot spoil a baby.
Though many Americans want their children to learn to be independent as early as possible, forcing a baby to manage herself alone is not the way to foster independence. Rather, independence arises naturally out of a secure relationship that builds up after many episodes of having her needs adequately met. For a summary of studies on the relationship between cosleeping and later child outcomes, see here.
To a helpless baby (and all babies are), crying and being ignored is inherently stressful. Though mild stress can "inoculate" a little one and help her learn to self-regulate her inner states, overwhelming stress--especially in infancy--can be toxic. Toxic stress can interfere with the expression of genes that set a baby's stress regulation levels in the developing brain.
Each baby is different, with a unique temperament, yet sleep training is a one-size-fits-all approach. Just because one baby sleeps through the night doesn't mean that all babies can and should. A vital part of parenting involves learning your baby's unique needs.
And finally, a systematic review of sleep training programs for babies under six months, published recently in the Journal of Developmental and Behavioral Pediatrics concluded that the strategies have "not been shown to decrease infant crying, prevent sleep and behavioral problems in later childhood, or protect against postnatal depression." In fact, sleep training in the first weeks and months of a baby's life, "risk[s] unintended outcomes, including increased amounts of problem crying, premature cessation of breastfeeding, worsened maternal anxiety, and, if the infant is required to sleep either day or night in a room separate from the caregiver, an increased risk of SIDS."

Cosleeping, not sleep training, is what is "biologically appropriate," says James McKenna, director of the Mother-Baby Behavioral Sleep Laboratory at the University of Notre Dame. McKenna has studied infant-parent cosleeping for most of his career.

Technically, cosleeping is defined as any situation where the infant and parent are within sensory range of each other. It has often meant sharing the same bed, but that has some risks as Arjun pointed out. Nowadays, McKenna, and many others in the United States, recommends separate-surface cosleeping, for example, placing the baby in a bassinet within reach or in a small crib next to the bed.

"There are as many ways to cosleep with your baby as there are cultures doing it," McKenna says.

Here's why keeping babies close is important:

Following birth, babies and caregivers remain physiologically connected to each other in complex ways, and when this bond is supported, babies do better. Breastfeeding, for example, is ideal for brain growth and future health. Babies who are breastfed have lower rates of sudden infant death syndrome (SIDS), diabetes, and other serious health conditions, while breastfeeding mothers have lower rates of postpartum depression, cardiovascular disease, diabetes, and hypertension. Breast milk is low in calories (but easy on digestion) so babies feed every hour and a half to two hours. When babies sleep close to their caregivers, they sleep more lightly, and wake two to three times more often than babies who are further away. The close proximity offers easy access with minimal disturbance.

Individual babies vary in how often they wake, from two to 13-15 times a night. But feeding isn't the only thing that happens more in the frequent wakings. When babies rouse, oxygen levels and heart rates rise, which is good for brain growth and development and immune functioning. The light sleep and frequent stirring also interrupt and stop episodes of apnea, or pauses in breathing, that can be deadly when prolonged.

And babies aren't the only ones responsible for the wakings. When McKenna observes mother-baby pairs sleeping in his lab, he finds that mothers wake babies about 40% of the time, and babies wake mothers about 60% of the time. The nighttime cameras show that mothers are often simply reassuring the babies emotionally: They "touch, hug, inspect, whisper"--loving gestures that also in turn raise baby's heart rate and oxygen levels.

"Remarkable to observe," McKenna says. And, not surprisingly, his cameras show that babies spend almost 100% of their sleep facing their mother.

Staying close to the adult's body helps the baby remain at a more stable body temperature. Physical contact, in close cosleeping, helps babies to "breathe more regularly, use energy more efficiently, grow faster, and experience less stress," says McKenna. Babies, too, who are not necessarily breastfed, as in the case of adoption, will also naturally reap the many other benefits of such close contact.

When babies are artificially put into deeper sleep through formula-feeding and the sensory isolation of a separate room, McKenna says, they not only are deprived of this close interaction and its attendant physical and emotional benefits, but the risk of SIDS rises. By contrast, in cultures where co-sleeping is the norm, incidents of SIDS are far lower or even unheard of.

Not all cosleeping arrangements are safe, though. Parental smoking, drinking, and drug use make parents insensitive to their babies and can be dangerous. The presence of other children and/or heavy duvets that can smother, are also are dangerous. So are places where a baby can get trapped, like gaps between beds or in couches or recliners. (A list of recommended guidelines can be found here.)

Despite the benefits of cosleeping, pediatricians still frequently recommend sleep training to exhausted parents of infants. This is unfortunate, especially for young infants under six months old. Rather than working to harmonize the mother and baby's biological systems, sleep training begins an adversarial emotional relationship between parents and their children. As McKenna points out, it sets us early onto the course of trying to make our children who we want them to be rather than respecting who they biologically are. And ironically, parents' sleep efficiency is not related to the number of times they're woken, but to their overall stress; e.g., mothers who exclusively breastfeed wake more often but have better quality and duration of sleep. McKenna recommends that pediatricians provide information on all sides of the issue so that parents can make informed decisions.

In our case, with a little practice and encouragement, Arjun got used to babies in the bed. And he'll be first to admit how addictive a baby's scent is.

Our girls had different timelines for transitioning to separate beds. By the time they were preschoolers, they began the night in their own beds, often ending in ours. But by this time, a family's values and preferences can be safely in play, and closeness happened to be just fine with us.

Time is always on your side, in parenting. Children won't be twenty and still sleeping with you.

Though in the deepest corners of our hearts, we sometimes miss it.