Young infants demonstrate a multitude of spontaneous movements of their bodies including arm flapping, leg kicking, and writhing body wiggles.1 Spontaneous kicking of the legs are present in utero and during infancy with or without external stimuli.2,3 The trajectory of spontaneous kicking in healthy infants is well described in the literature by decreasing kicking rates with age, high variability and flexibility, and better joint coordination as purposeful movements begin to emerge.4–6

Inter-limb patterns of alternating, unilateral, and synchronous kicks are important characteristics that describe the coordination and maturity of kicking. Newborns kick with a primarily alternating pattern and then develop dominant unilateral kicking patterns at one month. Synchronous kicking patterns become more prominent after five months of age.7,8 Importantly, type and variability of inter-limb kicking patterns may be associated with delayed development of motor milestones. Specifically, preterm infants who demonstrated predominant unilateral kicks at two months of age and predominant alternate kicks at four months of age started to walk later than infants with typical development.5,9 Although, it is difficult to use kicking in isolation from other clinical tools to identify developmental delay or disability, motor coordination measured with kicking may be an important component in understanding the development of both typical and atypical motor skills.

Intra-limb coordination of hip-knee, knee-ankle, and hip-ankle joint coupling are also important measurements of kicking coordination. In general, infants demonstrate tight coupling movements of the lower extremity as the ankle, knee, and hip joints flex and extend almost synchronously. Hip-ankle joint movements start to disassociate by the second month of age, followed by disassociations of hip-knee and knee-ankle joints between four and six months of age.10 Infants who are at high-risk of motor impairments may demonstrate delayed dissociation of intra-limb coordination.11–13 Importantly, tighter joint coupling (less mature) at two months of age is associated with a later attainment of walking.5 This may also indicate an important relationship between early joint coordination during kicking and the acquisition of new motor skills.

The mobile paradigm is designed to test learning and memory in infants.14 An infant’s leg is tethered to an overhead mobile, such that when the infant kicks, the mobile moves. As a result, the moving mobile provides conjugate visual and/or auditory reinforcement to the infant. With this reinforcement, infants are able to learn the association between their kicking and movement of the mobile by showing increased kicking frequency. Full-term healthy infants learn this paradigm in one day and remember it for one week.15,16 Interestingly, infants also demonstrated the ability to change their inter- and intra-limb kicking coordination in fairly complex ways,6,16,17 such as kicking more with the tethered leg,16,17 or moving their legs with less hip-knee coupling and specifically choosing a range of motion to induce external feedback more efficiently during the mobile paradigm.6,17,18

The existing evidence indicates that the infant’s ability to adapt kicking coordination may bestow more advanced contingent learning or vice versa. Yet, it is unknown whether there is an association between the developmental status of spontaneous kicking and performance in the mobile paradigm in early infancy. This association may offer an opportunity to investigate whether the development of very early motor skills play a role in supporting the growth in other domains, such as learning and cognitive development. Therefore, the purposes of this study were to assess the relationship between inter-limb and intra-limb coordination during spontaneous kicking and contingent learning assessed by kicking rate changes in the mobile paradigm task in infants at 3–4 months of age.

Ten infants completed both experiments but one of them was excluded due to unusable kinematic data of spontaneous kicking. Available data from nine infants were used for further analyses. The spatiotemporal measures and intra-/inter-limb coordination patterns in all nine infants are described in Table 2. The hip-ankle (71.6 ± 5.6%) and knee-ankle (68.3 ± 6.3%) joint couplings showed significantly higher dissociation movement (p = 0.008), indicating greater percentage of time in out-phase coupling movement, compared to hip-knee joint coupling (49.8 ± 9.3%); the proportion of unilateral kicking pattern (76.1 ± 9.4%) was significantly higher than alternate (5.4 ± 5.6%, p = 0.008) and synchronous (18.5 ± 9.6%, p = 0.008) kicking patterns.

Among the nine infants, six met the criteria of learning in the mobile paradigm task and were defined as leaners (66.7%); the other three infants were defined as non-leaners (33.3%). The spatiotemporal measures and intra-/inter-coordination patterns in both of the groups are also listed in Table 2. There were no statistical analyses for group comparisons between learners and non-leaners due to the small sample size.

The intra- and inter-limb coordination patterns and their relationships with the 3rd acquisition period and extinction phase of the mobile paradigm task are presented in Table 3. There was no significant correlation between intra-limb coordination patterns and normalized kicking rate in the mobile paradigm except a trend for a positive correlation between hip-ankle out-phase coupling movements and normalized kicking rate of 3rd acquisition period (r = 0.58, p < 0.10). There was a significant, moderate, negative correlation between percentage of unilateral kicking pattern and normalized kicking rate of the extinction phase (r = −0.72, p = 0.03).

This work joins others in demonstrating that the spatial and temporal characteristics of spontaneous kicking in young infants are measurable. Fetters et al.12 showed mean values of peak hip angle (106.8°), hip flexion excursion (40.5°), and hip extension excursion (40.8°) in full-term infants at five months of age. These results are similar to our findings in infants between 3–4 months of age. However, the peak hip flexion and extension velocities (204.4 m/s and 289.3 m/s) tended to be higher in the study by Fetters et al.’ compared to our results. In comparison, Heriza20 showed smaller range of hip flexion excursion (62.02°) and extension excursion (59.30°) than results of the present study, but slightly slower peak hip flexion velocity of (148.31°/s) in full-term infants at three days old. Therefore, we speculated that hip flexion/extension velocities may continuously increase from birth to at least five months of age, while the total excursion, or amplitudes of hip movement, may tend to decrease from birth to a few months later.

In our study, hip-ankle and knee-ankle joint coupling showed dissociated movement prior to hip-knee coupling movements. This result is consistent with Thelen’s observation that the hip-ankle joint coupling shows the earliest dissociated movement from tight intra-limb joint coupling during kicking.10 Jeng et al.9 also showed higher hip-knee joint coupling, indicating less dissociated movement, in full-term infants who are at 2 and 4 months of age. Similarly, van der Heide et al.13 using medians of correlation coefficients to investigate intra-limb coordination patterns, found a trend of tighter hip-knee joint and knee-ankle coupling compared to hip-ankle joint coupling in full-term infants at 3 months of age. Based on previous and the current studies, dissociation of intra-limb joint coupling during kicking seems to start from the most distal joint. A better control of dissociated lower extremity joint coupling may enhance the efficiency of kicking, which decreases energy consumption, while interacting with the mobile to show positive impact on the learning process.6 However, the disassociated intra-limb kicking pattern at any pairs of lower extremity joint coupling was not associated with changes of kicking rate in the mobile paradigm in our study. This finding might be due to our definition of a kick, when the hip joint flexed and then extended ≥15° and vice versa without additional requirement for knee joint movement. In this case, we may miss counting the kicks that moved the mobile mainly by dissociated knee movements. However, we found a trend for a positive correlation between the out-phase movements of hip-ankle joint coupling and the normalized kicking rate of 3rd acquisition period, during which infants demonstrated highest kicking rate throughout the mobile paradigm task. Therefore, the demand for more efficient motor control may be important in supporting infants’ explorative behaviors during a challenge. We speculate that the first dissociation of the lower extremity intra-limb joint coupling (i.e. hip-ankle joint coupling) may produce a less constrained and more energy saving option when infants try to increase the number of kicks. Please note that the infants in this study were developing in a typical manner; the natural variation of maturity of intra-limb kicking patterns may not be large enough to impact kicking rate changes in the mobile paradigm task.

Previous studies showed variable results on the development of inter-limb kicking patterns in young infants. In our study, a unilateral kicking pattern was the dominant inter-limb coordination pattern. This is consistent with van der Heide et al.’s 13 findings in full-term infants at 3 months of age. Our result showed a negative moderate correlation between the normalized kicking rate of extinction phase and unilateral inter-limb kicking pattern. This means that infants who showed relatively fewer unilateral kicks had a higher normalized kicking rate in the extinction phase (i.e. learning). During the extinction phase, infants were no longer receiving visual and auditory feedback, and thus these infants may exaggerate and increase their kicking frequency or change kicking patterns to try to move the mobile. We speculate that without sufficient variability of movements, such as in infants with a mono-type of inter-limb kicking pattern (caused by delayed or impaired motor function), infants may demonstrate a reduction in exploratory kicking movements in the mobile paradigm task. We therefore suggest that a smaller percentage of a unilateral kicking pattern (and therefore more variability) at this age may indicate more mature leg control allowing for diverse kicking movements in reaction to the extinction phase. In contrast, infants who “failed” to learn the cause-and-effect relationship during the acquisition phase may be less motived to explore new or less frequently used movement strategies to move the mobile. These speculations support the perception-action perspective that cognitive elements, such as perception, interaction with the environment, learning, and memory, may drive the development of motor domain by eliciting new movement strategies and vice versa.6,17

At this stage, our preliminary findings have two clinical implications. First, we consider that greater capacity of voluntary movement control may offer a better opportunity for explorative behaviors and cognitive learning in infants. For high-risk infants, especially those who may have global delays like preterm infants, adequate motor experience with enriched external feedback may be considered as an essential component in treatment programs. Second, when conducting early assessment in infants with motor and cognitive risk factors, we suggest careful selection of the appropriate task to ensure the result of the targeted domain will not be unduly influenced by the constraints of another domain. If this is not feasible, the assessment results should be interpreted in a more cautious and comprehensive manner.

The inherent limitations of a preliminary study with a small sample prevent strong conclusions. First, the natural variation of maturation of kicking movement may not interfere with the exploratory activities and learning in the mobile paradigm task in infants with typical development. Therefore, we cannot conclude how impairments or delay in one developmental domain affect the other domain. Second, the cross-sectional design of this study may not allow distinguishing the impact of motor development on cognitive outcome, such as learning the cause-and-effect relationship, in this study. Future studies to compare infants with and without motor developmental delays or impairments, including atypical kicking movements, and the use of a longitudinal study design may provide opportunities to further understand the development of high-risk infants.

Our findings provide preliminary evidence to support the importance of motor explorative behaviors in infant development. These results may provide valuable insight to inform future larger scale studies aiming to investigate the influence of motor development on different developmental domains. Moreover, the validity of using mobile paradigm task to reflect learning ability in infants should be assessed in infants who are at risk of or showing motor impairments. In the future, studies to extensively understand how different body systems influence the development of each other with a larger sample and longitudinal design are also warranted to support and enhance growth and development in high-risk infants.

The authors declare no conflicts of interest.