Learning to Crawl Disrupts Infant Sleep (Or, Science Confirms What We’ve Already Observed)
Sometimes, when you’re going through a rough patch with your baby’s sleep, it helps just to know that it’s normal. A recent study published in Monographs of the Society for Research in Child Development helps in this regard, because it shows that when babies learn to crawl, they have a harder time sleeping during the night (1).
The study, conducted in Israel, followed 28 infants from about 5 months to 11 months of age. Every 2-3 weeks, the babies’ motor skills and sleep patterns were assessed at a home visit by a trained researcher. The researchers compared sleep patterns before crawling, during the two weeks around the onset of crawling, and during the time after crawling. Although this study was small, the measurements were detailed, and the fact that the study was longitudinal, following the same infants over time (rather than a snapshot in time of different-aged infants) make the findings pretty robust.
The babies’ motor skills were tracked by parents using an illustrated developmental checklist, and their observations were confirmed by the researcher at each home visit. Each home visit involved a 10-15 minute filmed play session during which the baby played on the floor with toys and parents. Crawling didn’t have to be the traditional hands and knees variety; it was simply defined as at least two deliberate movements forward. Babies find lots of different ways to get mobile. Here’s a video of BabyM scooting around at 8 months in his very own style, pulling forward with his arms with extra propulsion added by his big toes:
Sleep patterns in the study were measured using an actigraph that was worn on the babies’ ankle for three consecutive nights at each assessment. An actigraph measures movement (using an accelerometer, similar to a FitBit) to track when the babies are awake and asleep, allowing researchers to quantify night wakings, which was the primary measure of interest in this study. They only counted night wakings when they lasted for at least 5 minutes. (Shorter wakings were likely to be self-soothed back to sleep without asking for help from parents.)
So did learning to crawl affect the babies’ sleep? Yep. When the researchers looked at individual babies’ night wakings before, during, and after learning to crawl, they saw that the babies did indeed wake more frequently in the night when they started crawling. However, the total amount of sleep per night didn’t change.
The babies in this study started crawling as early as 4 months and 25 days and as late as 10 months and 7 days. Curious if the effects of crawling might depend on the age it began, the researchers split them down the middle into groups of early (average 6 months) and late crawlers (average 8 months). (Note that this was just how the babies in this study split when they were divided into two equal groups. The paper notes that according to World Health Organization norms, 8.3 months is considered the median time for hands and knees crawling.) Indeed, they did see a difference in sleep patterns. The early crawlers had an increase in night waking just around the time when they started crawling, but by their next home visit, they were back to normal. The late crawlers, however, showed a more gradual and sustained increase in night waking after learning to crawl. That is, they woke more often when they learned to crawl, but then they were waking even more often a few weeks later.
These data are difficult to visualize because of the large number of data points for each infant, but here is one way that the researchers show these patterns:
The Z-scores basically measure the distance from the mean number of wakings for each individual infant, so a higher Z-score means more wakings than usual for that baby. The graph shows the average Z-scores in the early and late crawlers across time, where 0 is start of crawling, -1 and -2 are the two time points before crawling, and 1 and 2 are the two time points after crawling. As you would expect, both groups have a drop in waking between -2 and -1, indicating fewer wakings as the babies grew older. Then you can see the disruption at time 0, the onset of crawling. You can see from the figure that the early crawlers show a quick peak in waking when they learned to crawl, but they’re basically back to normal by the next time point. The late crawlers, on the other hand, show a rise in waking with crawling, and waking continues to climb over the subsequent time points.
The authors include a sobering estimate that it took the babies a full 3 months to really return to their pre-crawling trajectories for sleep development. I’m not sure how useful that number is in real life, because there are a host of other factors that can affect sleep at that time in an infant’s life, including sleep training for some babies. But what this does show is that sleep disruption is really common at this age, and in many infants, it lasts for a while before it starts to get better!
The authors of the study speculate on some possible explanations for crawling-related sleep disruption:
“Why should a downward trend in sleep regulation appear in association with the ability to locomote? One explanation for the rise in number of long wake episodes draws on the increased regulatory and emotional reactivity that accompany locomotion onset (Biringen et al., 1995; Campos et al., 2000). The locomotion dependent emotional arousal and heightened excitement, alluded to in Mahler’s theory (Mahler et al., 1975), could compromise the child’s ability to self-regulate back to sleep after a brief nighttime awakening (Anders, 1994). Another possibility is that the developmental reorganization underlying the achievement of crawling (Adolph & Berger, 2006) involves restructuring of sleep–wake states, a premise that fits well with Dahl’s model (e.g., Dahl, 1996).”
No, I’m not familiar with any of these models. That’s why I’m quoting the authors here rather than pretending I know more about them!
“In dynamic systems, downward trends in performance and in behavioral control often mark the emergence of new abilities (Adolph & Robinson, 2008; Thelen & Smith, 1994). This pattern has been identified in diverse domains of infant development including manual reaching (Corbetta & Bojczyk, 2002), vocal production (Hsu, Fogel, & Cooper, 2000), and language acquisition (Verspoor, Lowie,& van Dijk, 2008). The present study adds sleep to this diverse array of behaviors in which a downward trend in performance (herein, regulating sleep–wake states) was linked to new developmental achievements (e.g., Corbetta & Bojczyk, 2002). That the emergence of crawling marks a period of increase in long wake episodes fits well with the dynamic system principle that development is spotted with periods of progress and regress, and that regressive behaviors are precursors of moving toward the next level of developmental organization (Thelen & Smith, 1994).”
Of course, this study made me think about my own baby’s sleep. Part of the reason why it caught my eye is that it reminded me of BabyM’s nap strike around 6 months of age, when he was working hard on rolling and trying out his voice with new babbles. Many parents notice sleep disruptions around developmental milestones, when our babies seem more interested in practicing new skills than sleeping.
I have no idea if learning to crawl disrupted BabyM’s sleep. He started pulling himself across the floor on his belly, as shown in the video above, around 7 months. At the time, we were traveling, and we took another trip a few weeks later. (Travel = sleep disruption, in my experience.) Then he got his first tooth and the worst cold of his life. Not surprisingly, after all of this, he was routinely waking up 2-3 times per night, more often than he was at 3-4 months of age. If I had the time or energy (yawn…) to make a Z-score graph for him, I think it would look a lot like that graph for the late crawlers in this study. And if the infants in the study were anything like my little guy, crawling might cause an initial sleep disturbance, but other factors might add to or help to sustain that waking pattern.
Along these lines, I’m also not surprised that crawling was more disruptive when it developed later in infancy. Studies show, and I’ve observed with my own kids, that the most rapid sleep consolidation occurs in the first few months of life as babies learn night from day and are gradually able to go longer between feedings (2-4). In the second half of the first year, lots of factors can disrupt sleep, and babies that previously slept through the night may begin waking again. Maybe baby wakes and cries because of discomfort from teething or because a little crawling practice in the middle of the night makes it hard to resettle. If you’re like me, you feed the baby, because that’s usually the easiest way to help him resettle. Over time, those wakings and feedings can sometimes become habit. In my experience, more wakings and feedings can develop slowly and unintentionally, but it takes an intentional and concerted effort to get back to a reasonable night of sleep. We’re gradually working on that now with BabyM.
Did you notice a change in sleep when your baby started crawling?
- Scher, A. & Cohen, D. V. Sleep as a Mirror of Developmental Transitions in Infancy: The Case of Crawling. Monographs of the Society for Research in Child Development 80, 70–88 (2015).
- Anders, T. F. & Keener, M. Developmental Course of Nighttime Sleep-Wake Patterns in Full-Term and Premature Infants During the First Year of Life. I. Sleep 8, 173–192 (1985).
- Henderson, J. M. T., France, K. G., Owens, J. L. & Blampied, N. M. Sleeping Through the Night: The Consolidation of Self-regulated Sleep Across the First Year of Life. Pediatrics 126, e1081–e1087 (2010).
- Scher, A., Epstein, R. & Tirosh, E. Stability and changes in sleep regulation: A longitudinal study from 3 months to 3 years. International Journal of Behavioral Development 28, 268–274 (2004).