You might have heard the sage advice given to those struggling through a long process before achieving a lofty goal: You must learn to crawl before you can walk. The meaning is simple: Master the fundamentals first, endure the process of development, and you will be rewarded.
Infants—the literal basis for the idiom—are at the foundation of that development cycle. Multiple systems are simultaneously progressing with astonishing rapidity. Basic functions of life are being established, such as processing stimuli, eating, and communicating. Among these is a motor skill that expands the infant’s world—mobility. And crawling is the first skill that provides children with independent mobility.
Babies are born to move. Even in the womb, those kicks and jabs the mother feels foreshadow the wiggly toddler who’s always kicking off her socks. Newborn infants display reciprocal limb movements that suggest the alternating flexion and extension we see in walking. Even before they have the strength needed for upright standing, infants supported on a treadmill will take steps in a pattern very similar to the gait cycle.1 Development specialists have theorized a central pattern generator in lower nerve centers contains the fundamental process for basic walking.2
As a pediatric biomechanist, I have worked for over two decades to improve the mobility of children with limb loss and neuromotor movement challenges. When I look back, though, I would say that 95 percent of this work has been focused on walking, with a dabble here and there in running and one study on infant crawling in children with limb loss. It’s that last one, the crawling study, that made me realize two things: Crawling is largely overlooked in gait analysis, and crawling is very difficult to study.
Our First Crawling Study
Motivated by innovative ideas at Children’s Healthcare of Atlanta on prescription protocols for young children who will need a prosthetic knee, our research team had completed several studies on walking in toddlers with limb loss before we decided to take a step backward, developmentally speaking. About that time, Colleen Coulter, PhD, DPT, was pursuing her doctorate in physical therapy and, following years of experience working with children, had an instinct that the benefits of an early working prosthetic knee happen prior to walking. So we set out to conduct instrumented 3D gait analysis on the crawling of infants with limb loss.
We found that children with limb loss who receive a working knee joint in their first prosthesis use it well while crawling.3 They flex the prosthetic knee appropriately (close to 100 degrees), and we found, as expected, that large bilateral asymmetries and awkward trunk accommodations emerge if a child tries to crawl using a prosthesis without a working knee.
What the publications and presentations that resulted from that study fail to convey, however, are the challenges we encountered in collecting the data. The standard musculoskeletal models and marker sets we use for gait analysis don’t work for crawling, so we had to create a new one. We had to use tiny reflective markers, because babies make more contact with the ground and a big marker might be uncomfortable. We had to position cameras at new angles to track those tiny markers. Diapers created a big challenge, given their movement relative to the skeleton. And as any parent or caregiver can attest, babies can be a challenging population to work with. They don’t follow directions well, so we tried to create environmental cues and lots of motivational stimuli. They’re temperamental, so we had to be very patient and flexible. And they’ll pull a reflective marker off their skin and pop it in their mouth in a second if you let them, so we had to watch them like a hawk.
Despite the challenges, we succeeded in adding important evidence to support early provision of prosthetic knee joints in young children. While I didn’t quite come away vowing never to study crawling again, I did understand why we have so much data on walking in children and so little data on crawling.
What Do We Know?
Researchers have observed infant crawling since the 1930s. In the 90s, foundational work by developmental psychologists like Esther Thelen, PhD, and Karen Adolph, PhD, used crawling to understand complex development of multiple neuromotor systems. For these researchers, the mobility of crawling represents a unique opportunity to observe children’s interaction with an expanded and oft-changing environment. During this process, children’s disorganized neuronal connections are selected and repeated as they explore, and new goal-oriented connections are formed.4
Crawling is also important in understanding developmental continuity, or where new skills grow from. Examination of motor development in infants reveals skills with physical similarity, like crawling, cruising, and walking, that sometimes develop sequentially but also might overlap considerably. Despite this overlap, there are ways in which motor learning must start over in each stage. For example, infants who learn to avoid steep slopes or large gaps while crawling go right back to attempting them after they transition to walking.5,6 Adolph and colleagues further concluded that crawling, cruising, and walking involve fundamental differences in balance control, based most simply on the number of limbs in contact with the ground or other support structure, and the resultant muscle groups required for postural control.7
Studies have indicated that well over 80 percent of infants progress through hands-and-knees crawling during development of locomotion, while others use alternative crawling strategies.8,9 But what about those babies who just skip crawling and go straight to walking? Does that suggest that crawling is less important? Growing evidence suggests that crawling is a very important stage of development. Cazorla-González et al. assessed a battery of physiological assessments in children seven to eight years old stratified into those who crawled as infants and those who did not.10 Their results were noteworthy. Children who crawled before walking showed a higher number of interactions among variables assessing all sorts of markers, including body composition, cardiovascular and lung function, and physical fitness. In addition, crawlers had lower fat mass, higher relative muscle mass, and lower systolic blood pressure. Furthermore, atypical crawling development has been linked to conditions like autism spectrum disorder. As far back as 1998, Teitelbaum et al. used observational analysis to identify differences in crawling in children with autism.11
Despite this evidence, the Centers for Disease Control and Prevention (CDC), rather controversially, removed crawling from its 2022 update to the checklists for developmental surveillance milestones.12 In their special communication for pediatric physical therapy, Kretch et al. agreed with the removal, but noted that many therapists are concerned that “children who do not crawl will no longer be identified as having an atypical development trajectory.”13 The authors also revealed the reasons crawling was removed: lack of normative data, inconsistency in definitions of crawling, variability in the timing of the milestone, and lack of evidence that all typically developing children will crawl.
Toward a Systematic Understanding of Crawling
As the CDC’s decision confirms, we still do not have a definitive, quantitative understanding of the normal development of crawling. Without that, we are unable to characterize differences found in atypical development. For example, cerebral palsy, which is typically acquired at birth, is often not diagnosed until motor differences are spotted as a child learns to walk. Tools like Prechtl’s General Movements Assessment rely on trained observation of movements in infants, and are not as widely utilized as, say, observation of walking in the office of a pediatrician, therapist, or orthopedist.
My colleagues and I at Kennesaw State University and Children’s Healthcare of Atlanta have just started a three-year project that should directly address the current unknowns that resulted in the removal of crawling from the standard milestones. The research, funded by The Gerber Foundation, gathers a wealth of information about crawling and does not require attachment of markers or measurement of specific body segments. We use a pressure-sensing mat made by ProtoKinetics that can sense the amount of force under each limb as the infant pushes. We’ve coupled that with software originally designed for quadruped veterinary research that also tells us how many limbs are in contact with the ground at any time and the timing and distance of each contact. We’re uncovering details about crawling that you can’t find through standard observation and doing so in one of the largest prospective studies ever conducted for this stage of development.
The study will bring 75 typically developing children to our lab from the onset of crawling and then every two weeks until they transition to walking. In addition, we’ll measure the crawling of 15 children with limb loss.
What We’ve Learned
Our study is in its early stages—we’ve only had 28 lab visits so far—but we’re already getting some fascinating information. Here are three preliminary findings, given with the caveat that our conclusions could change as we expand the study to its full population.
Speed gains come from more than growth
We know that children walk faster as they grow because their legs get longer, and they therefore cover more distance with every stride. We can naturally assume the same is true of crawling. And that’s what we’ve found. However, physical growth doesn’t paint a complete picture. We’ve found that speed gains come from both step length and from step frequency, also called cadence. And this progression is fairly linear with age, to a point. Which brings us to our second finding.
Infants sacrifice speed to gain new movement patterns
In my gait analysis classes, I teach that if you could pick only one outcome measure to describe what is most important about walking mobility, it’s speed. For a child with movement challenges, it can often be more important that they are able to keep up with their friends than that their gait looks like that of their friends. But we’ve watched an interesting phenomenon occur in a couple of the children in our study: As they get close to leaving crawling behind and progressing to walking, their crawling slows down. They sacrifice what might be the most important aspect of their mobility to try new crawling patterns that will help them transition to walking. This indicates that a complex optimization pattern is at work in infant mobility.
Children with limb loss show a different loading profile
We hope to eventually test multiple populations with atypical motor development, but for now we’re focusing on young children with limb loss. We want to know if the variables we’re measuring will be sensitive enough to identify differences between populations. Our early results show a significant difference in a measure called the anterior-posterior integrated pressure ratio. In short, it means that infants with limb loss bear more of their body weight on their arms than their legs during crawling, unlike typically developing infants. The clinical implications are unclear—we don’t have literature in this area—but my gut says that this ratio should show a shift to weight bearing with the legs as infants move toward walking. That would imply that kids with limb loss will need to learn to shift their weight back before they can progress.
What’s Next
As mentioned, these are only a few preliminary results, but as I see our data pile up, our team is excited to anticipate the insights we’ll be able to make. With age-based norms, quantitative descriptions of distinct crawling patterns, and new knowledge of typical and atypical development, we are hopeful that crawling will become a well-understood stage of infant motor development, and a window for valuable early intervention when it is needed.
Mark Geil, PhD, is the associate dean for research in the Wellstar College of Health and Human Services at Kennesaw State University. He holds degrees in mechanical engineering and biomedical engineering and has conducted research in pediatric locomotion for over 25 years.
Top photogragh: Oksana Kuzmina/stock.adobe.com
References
- Yang, J. F., M. J. Stephens, and R. Vishram. 1998. Infant stepping: A method to study the sensory control of human walking. Journal of Physiology 507 ( Pt 3):927-37.
- Thelen, E., G. Bradshaw, and J. A.Ward. 1981. Spontaneous kicking in month-old infants: Manifestation of a human central locomotor program. Behavioral and Neural Biology 32(1):45-53.
- Geil, M. D., C. Coulter-O’Berry, M. Schmitz, C. Heriza. 2013. Crawling kinematics in an early knee protocol for pediatric prosthetic prescription. Journal of Prosthetics and Orthotics 25:22-9.
- Sporns, O., and G. M. Edelman. 1993. Solving Bernstein’s problem: A proposal for the development of coordinated movement by selection. Child Development 64(4):960-81.
- Adolph, K. E. 2008. Learning to move. Current Directions in Psychological Science 17(3):213-8.
- Adolph, K. E., C. S.Tamis-LeMonda, S. Ishak, L. B. Karasik, and S. A. Lobo. 2008. Locomotor experience and use of social information are posture specific. Developmental Psychology 44(6):1705-14.
- Adolph, K. E., S. E. Berger, and A. J. Leo AJ. 2011. Developmental continuity? Crawling, cruising, and walking. Developmental Science 14(2):306-18.
- Storvold, G. V., K. Aarethun, and G. H. Bratberg. 2013. Age for onset of walking and prewalking strategies. Early Human Development 89(9):655-9.
- Robson, P. 1984. Prewalking locomotor movements and their use in predicting standing and walking. Child Care Health and Development 10(5):317-30.
- Cazorla-Gonzalez, J., S. Garcia-Retortillo, M. Gacto-Sanchez, G. Munoz-Castro, J. Serrano-Ferrer, and B. Roman-Vinas, et al. 2022. Effects of crawling before walking: Network interactions and longitudinal associations in 7-year-old children. International Journal of Environmental Research and Public Health 19(9).
- Teitelbaum, P., O. Teitelbaum, J. Nye, J. Fryman, and R. G. Maurer. 1998. Movement analysis in infancy may be useful for early diagnosis of autism. Proceedings of the National Academy of Sciences of the United States of America 95(23):13982-7.
- Zubler, J. M., L. D. Wiggins, M. M. Macias, T. M. Whitaker, J. S. Shaw, and J. K. Squires, et al.2022. Evidence-informed milestones for developmental surveillance tools. Pediatrics 149(3).
- Kretch, K. S., S. L. Willett, L. Y. Hsu, B. A. Sargent, R. T. Harbourne, and S. C. Dusing. 2022. “Learn the signs. Act early.”: Updates and implications for physical therapists. Pediatric Physical Therapy 34(4):440-8.
