Cultural Diversity in Infant Motor Development: A Comparison of Early Locomotor Experience

Abstract

Most of the research on infant motor development is based on studies with infants and caregivers living in urban and western societies, which is problematic from a scientific and ethical perspective. Both ethnographic descriptions as well as a handful of empirical investigations across diverse cultural contexts indicate there may be significant variability in early motor experience across cultures. To better understand whether differences in early experience are linked to motor development, we observed 52 12-month olds (SD = .66) with their caregivers in three cultural contexts: Tanna (ni-Vanuatu; N = 9), Aka (Central African Republic; N = 10), and rural villages in Tajikistan (N = 33). We coded the videos to determine the duration of time infants spent restricted, the time they engaged in spontaneous movement, and the ways in which they were restriction. When unrestricted, infants spent similar proportions of time exploring their environment through locomotion across the three cultural groups. However, Tajik infants were restricted from movement more than infants in Vanuatu and the Aka community. Restriction duration did not correlate with infants’ movement duration. Notably, Tajik caregivers primarily used devices for restriction, while ni-Vanuatu and Aka caregivers relied on holding. These variations in parental choices and how they shape children’s early environments provide insights into how caregiving practices influence the context of motor development across cultures. While previous studies highlight direct parenting practices’ influence on motor development, our findings suggest indirect practices may shape movement opportunities.

Historically, the development of infants’ motor skills was conceptualized as a maturational process impervious to culture and context (Bayley, 1993; Gesell, 1928). The reason is likely because early scholars focused on meticulously documenting the progression of infant motor skills without much attention to cultural variation in experience, often assuming universality in sequence and timing of motor skills (Adolph et al., 2010). Any potential cultural differences were deemed as noise and therefore received little attention. Assuming universals in the developmental processes and motor outcomes, research focused on the standardization of motor skills by focusing on children from WEIRD societies (Western, educated, industrialized, rich, and democratic; Adolph et al., 2010; Henrich et al., 2010).

When looking to cultures outside of the so-called “WEIRD” societies, experiences and motor outcomes may not generalize (Mendonça et al., 2016). For instance, studies report significant variation in the age of acquisition of motor achievements. Whereas studies from western societies report that 90% of infants achieve independent sitting by 7 months, infants in some countries on the continent of Africa sat independently by 5 months (Karasik et al., 2015). Moreover, 29% of Jamaican infants skipped crawling, and some Jamaican infants began crawling at the same age they began walking (Hopkins & Westra, 1990). In earlier studies from Nigeria, the authors report that the infants achieve most gross-motor milestones such as “crawl,” “stand well alone” and “walk well alone” earlier compared with infants from western societies (Iloeje et al., 1991). However, some transitional milestones such as “roll over” and “stand while holding on” were achieved later by Nigerian infants. These reports show variability in motor skills across diverse cultures but only speculate about the mechanisms for these differences. Cross-cultural differences in motor development are typically assumed to reflect differences in caregiving goals and beliefs about ideal development (Keller et al., 2002). Yet, caregiving goals and beliefs are only one explanation. It could be the case that such contextual differences reflect the living situation (ecological differences), caregiving practices (e.g., restriction due to safety concerns), temperamental differences, or a combination of these, to name a few (Aime et al., 2020).

In the current study, we aimed to document similarities and differences in motor movement and restriction, while also exploring potential explanations for cultural variations. We did this by examining one aspect of motor development in three diverse cultural contexts: Tajikistan (Tajiks), Tanna (ni-Vanuatu), and the Central African Republic (Aka). We examined the duration of time infants were restricted, how they were restricted, and their propensity to move around on their own. We aimed to address two key questions: (a) Do infants experience different types and durations of restriction across cultures? and (b) When unrestricted, do infants differ in their spontaneous exploratory activities? By investigating these questions, we seek to understand how variations in early experiences shape motor development.

Infants’ Opportunities to Move Across Cultures

Infants’ early motor experiences have been linked with the caregiving practices they receive, which, in turn, enhance or limit their opportunities to practice motor skills by structuring infants’ early environment (Adolph & Robinson, 2015). More specifically, differences in the way caregivers structure the environment and interact with their children affect the form of new skills, the ages when they first appear, and the shape of their developmental trajectory. Different child-rearing practices may affect the timing or sequence of motor milestones by providing various kinds of stimulation and opportunities for practicing motor skills in daily routines. For example, in many cultures in Africa and India, the caregiver usually includes formal massage and stretching in infants’ daily bath routine (reviewed in, Adolph et al., 2010). Experiments showed that the timing of motor skills can be accelerated with just a few minutes of stimulation each day over several weeks (Clark et al., 1977; Porter, 1972). Moreover, !Kung mothers of Botswana wear decorative objects around their necks, and infants grasp and manipulate these objects during the day (Konner, 1972). As a consequence of continual access to objects and the practice of grasping, !Kung infants show an acceleration in prehension skills (Konner, 1977). Researchers agree that environmental opportunities for movement influenced by the social interactions or child-rearing practices facilitate improvements in motor behavior and motor competences (Adolph & Franchak, 2017). Thus, the level of motor skill performance may reflect the amount and type of practice encouraged by the culture and be attributed to culture-specific practices.

Different cultural groups also have distinct ways of restricting infants’ locomotion. In some cultures, this restriction is a result of the ecology in which the infant is embedded, while in others, it may be a result of cultural beliefs. For example, among the Navajo, it is common for caregivers to use a “cradleboard” (Donaldson et al., 2020). Chisholm (1978) reports that Navajo culture believes that the use of cradleboards will offer both physical and spiritual benefits to the infant to calm the infant and promote sleep. Infants are often tightly swaddled and wrapped with little ability for active movement or exploration on the cradleboard (Chisholm, 1978). Furthermore, Inuit mothers use an “Amauti” (traditional clothing that is designed with a large hood and pouch) to pack and carry their infants on their backs, and most Inuit children spend their first 3 years nestled in the amauti (Blackduck, 2001). The close and prolonged contact between the Inuit mother and child protects the child from cold and frees the mother’s hands for other activities. Similarly, mothers from a foraging society, the Ache of Eastern Paraguay rarely put their infants down or let them venture more than a meter away because of the dangerous forest environment (Kaplan & Dove, 1987). These descriptions of child-rearing practices across diverse cultural environments suggest that culturally distinct child-rearing can lead to differences in infants’ opportunities to move. We do not know, however, whether infants’ opportunities to move will affect how much they move. Based on previous literature, we expect that increased opportunities for movement may lead to greater exploration of the environment.

The Effects of Restricted Movement Across Cultures

The effects of prolonged restricted movement are not well known because research on infant motor development around the world is limited (Carson et al., 2022). Cradleboards have often been thought to potentially influence motor development in Navajo infants, but there is insufficient evidence to support this assumption (Chisholm, 1983). For instance, in a study of Hopi children on the cradleboard, Chisholm (1983) found that motor development appeared typical and the motor skill milestones such as the age of onset of walking occurred at around the same time in those who had not been kept on the cradleboard. The author also pointed out that the cradleboard did in fact lower infants’ levels of arousal and activity, however, these effects do not constitute a perturbation in their long-term motor development. Some prior research claimed that restricted movement could delay the development of infants’ postural and locomotor skills as well as delay the acquisition of certain motor skills (Cole et al., 2012; Hospodar et al., 2021; Karasik et al., 2018).

However, recent research examining a traditional child-rearing practice referred to as gahvora cradling in Tajikistan and other parts of Central Asia found that restriction differed, and some aspects of motor development may be affected but the effects are not lasting (Karasik et al., 2018, 2023). The gahvora is used for sleep practices as well as to contain infants. Caregivers tightly swaddled and bound infants’ arms, legs, and torsos to the cradle bed, and infants could only move fingers and toes in most cases. The authors found that Tajik infants’ movements were indeed severely restricted: 20% of 12- to 24-month-olds spent more than 15 hr bound in the gahvora. In the following study, the authors found that fewer Tajik infants demonstrated several motor skills compared with same-age infants represented in a report by the World Health Organization, the motor development of infants (Karasik et al., 2023). Notably, Tajik 4 to 5-years-olds showed comparable performance on motor skills compared with the U.S. children. It suggests that although movement restriction affects infants’ motor development and physical growth, Tajik infants likely took advantage of opportunities to practice motor skills outside gahvora. According to the authors of this study, it remains unclear whether infants’ motor exploration when unconstrained is impacted by such restriction.

Cultural Settings

According to Super and Harkness (1986)’s developmental niche, the cultural structuring of child development has three components: the physical and social settings; the customs of child-rearing; and the psychology of the caregivers. Following the theoretical framework of the developmental niche, the differences in infant motor development are shaped by the ecological differences (living situation), or caregiving practices (habits of care and daily practice), or parental beliefs (expectations of onset of motor milestones). For example, specific social contexts in which a child is raised can also place demands on the motor abilities of infants. The kinds of stimuli provided by caregivers and others, including physical spaces, toys, and housing conditions, influence parental practices and ultimately may affect infant development (Ammar et al., 2013; Bombarda et al., 2017). In Oudgenoeg-Paz et al.’s (2020) research, the authors discussed that the settings of motor development, both distal environment (such as different climate and landscape between countries) and proximal settings (such as providing space and objects), could potentially support or hamper motor development by influencing the children’s immediate environment and opportunities for movement. These differences in proximal settings are typically assumed to reflect cultural differences in child-rearing practices, beliefs, and goals, yet they may also reflect differences in the immediate environment that are outside of the parental control. For example, whereas a caregiver may have a goal for their child to explore and achieve motor milestones early, they may be constrained by the safety of the immediate environment. In fact, cross-cultural difference in motor development is typically assumed to reflect differences in caregiving goals and beliefs about ideal development (Keller et al., 2002). Yet, the parental belief is only one explanation, the contextual differences and caregiving practices also provide essential possible explanations for the variations in children’s motor development.

For this project, we examined one aspect of motor development across cultures—motor restriction and motor movement in three settings in which we had long-term, established research relationships examining family life, parenting, and child development. The dataset consists of a collection of videos gathered at different timepoints by three different researchers for different research programs. This research project draws on this natural observation dataset to examine a question about early experience that has not been addressed yet in the literature. There was no a priori theoretical motivation for combining our observations of children in these three culturally diverse communities, rather our goal was to better understand the range of diversity in infant experience in societies that are not represented in the scientific literature (Nielsen et al., 2017). Three of the authors of the current study have been collaborating to address questions relating to caregiving practices in each of these regions. It would be difficulty to conduct an ethical, community-engaged research project across multiple sites without collaborating with experts working in these regions (Amir & McAuliffe, 2020; Broesch et al., 2020, 2023). Therefore, we combined our expertise and our data to produce a novel dataset from which we could examine infants with their caregivers. We examined data from on Tanna Island, Vanuatu, where one of our MacGillivray has been working since 2012 as well as Aka, Central African Republic with observations made by Hewlett and video recordings of caregivers and infants in rural villages in Tajikistan by Karasik. Below we describe notable differences and similarities in caregiving behaviors in each of these societies, based on our own research, experiences, and published work by others in these societies.

Tanna (Vanuatu)

Tanya MacGillivray has been working on Tanna since 2012. Tanna is one of several islands in the archipelago of the Republic of Vanuatu, located in the South Pacific region. The island itself is of volcanic origin with coral reefs and narrow coastal flats surrounding a mostly forested central highland region. From our personal experiences on Tanna, we observed that parents encouraged their children to develop independent subsistence skills very early in life. More specifically, much of adult and child life is spent attending to crops, food gathering, harvesting, and fishing (Aime et al., 2020). In addition, we observed that parents encourage a collective mind-set with a decreased emphasis on the self as a distinct individual (Atkinson, 2023). Caregivers on Tanna expect children to be responsible for assisting adults in labor from a very young age, such as cooking, prepping crops, and helping with the child care of younger siblings. On Tanna, the life of an infant consists of very little exposure to manufactured items, and much of the child’s life appears to be spent outdoors with extended family and mixed-age groups (Broesch et al., 2021). Infants are typically carried by a caregiver for long distances or in the arms of a caregiver with the infant facing outward. This can be contrasted with the developmental science literature which is based largely on infants living in urban households with an emphasis on formalized education (see Lancy, 2014). The data was collected in 2014 and 2016 and although 10 years have passed since the first data collection of natural observations in this reg, much of traditional life remains the same. There remains to be no road and no public utilities in these villages (as of 2024), and therefore, there is very little influence of market economy and western lifestyle and family norms and ideals.

Aka (Central African Republic)

Barry Hewlett has been working among the Aka (or BaYaka) foragers in the tropical of the southwestern Central African Republic in the tropical forests of the Congo Basin for over 30 years. The Aka live in fluid communities of 25 to 50 people who move about four times per year. The Aka economy is based on a mix of hunting and gathering wild foods from the forest, small-scale, low-effort slash-and-burn agriculture, and exchange with neighboring farmers. Community life is open and intimate, and as much as 80% of food procured is shared with other camp members. Aka foundational schemas include respect for autonomy and egalitarianism. Status differences between people are minimized, and all individuals including children do as they please. Aka children typically either accompany their parents on foraging trips or stay close to home with the other children younger than 5-years-old, typically casually attended to by any adult who has not left the camp. There are no fences between houses, and children move freely throughout the neighborhoods (Hewlett et al., 1998). Broesch et al. (2021) described the general features of Congo Basin forager childhood and examined diversity in child care patterns between forager ethnic groups. They emphasize that respect for an individual’s autonomy and a giving or sharing way of thinking represent the foundational schema among forager life. The way of life of most Congo Basin foragers is characterized by high mobility, temporary shelters, and regular foraging. The data was collected in 2010 and although 14 years has passed, traditional lifestyle still plays a strong role in many Aka communities. In general, Aka population still lived in temporary camps established according to the seasons and depend on traditional trade relationships. During infancy and early childhood, foragers show more giving and are more responsive to their infants distress signals than farmer populations in the region. For instance, Hewlett et al. (2011) found that forest foragers breastfed more frequently, had more breast-feeding bouts per hour, and were more likely to be holding infants when nursing than farmer groups.

Tajikistan

Lana Karasik has been working in Tajikistan examining motor development for nearly a decade. Tajikistan is a mountainous landlocked country in Central Asia with a primarily rural population dependent on livestock and small-scale agriculture. The video set we used focused on the warmer, arid, Khatlon district and the colder, mountainous, Rasht district. From Karasik et al. (2023)’s study that documented the “gahvora” cradling use in Tajikistan, the author described that each family lives in a single-room, one-story clay home. Several homes are clustered around a small courtyard or garden with fruits, nuts, and vegetables grown for consumption and trade. Chairs, tables, and beds are rare, instead, indoor surfaces are covered in carpets. Infants are surrounded by many adults and children and cared for by parents, relatives, and neighbors. According to the Karasik et al. (2018), Tajik people historically have cared for newborn children in a small traditional rocking cradle. Infants from birth to 20 months of age are bound on their backs in a tightly wrapped swaddle with arms extended along the sides of the torso and legs straightened and tied together. In this way, the child cannot move its arms and legs, and cannot turn from side to side. The data was collected in 2013 and although 11 years has passed, caregivers in Tajikistan still use the gahvora to cradle their infants nowadays and value it as a traditional way of child-rearing. Much of traditional life remains the same in Tajikistan families as well, such as the involvement of multiple caregivers in childcaring and the multigenerational family structure.

Current Study

Our aims were as follows: First, to determine whether there were differences between the three societies in the duration of time in which caregivers restricted their infants from movement and the ways in which caregivers restricted their infants’ movements, we examined the duration of restriction across the three societies and compared the different restriction practices (restricted by a caregiver or device). Hypothesis 1: As previous work has shown some cross-cultural differences in child-rearing practices in these three societies (Broesch et al., 2021; Hewlett et al., 1998; Karasik et al., 2018), we expected differences between the groups in the duration of time in which infants are restricted. Specifically, we expected Tajik infants to experience longer durations of restriction when compared to Aka and ni-Vanuatu infants. Based on literature describing family daily life in each of these regions (Cebioglu et al., 2022; Hewlett et al., 1998), we also expected that Tajik caregivers would use devices to constrain their infants more than Aka and ni-Vanuatu caregivers. We did not have a priori expectations regarding the comparison between the Aka and ni-Vanuatu caregivers and infants. Second, to determine whether there were differences in infant self-generated locomotion, we examined and compared the proportion of time infants spent exploring their environment through locomotion when unrestricted across the three societies. Hypothesis 2: In line with previous literature indicating that Tajik infants show later walk onset age (20 months) compared with same-age infants represented on WHO standards (16 months; Karasik et al., 2023). Because infant motor milestones have been linked to early experience or practice (Adolph & Hoch, 2019; Lobo & Galloway, 2012), we also expected differences between societies in infant self-generated locomotion, specifically with infants in Tajikistan showing less time engaged in exploration locomotion compared to the other two societies. Third, we examined whether infants’ opportunity to move freely (unrestricted time) was correlated with infants’ self-generated locomotion by analyzing the proportion of time infants engaged in self-generated locomotion (e.g., crawling or walking) and the proportion of time they spent being constrained by a caregiver or device (e.g., being held, being in a stroller) throughout the day. Hypothesis 3: Based on literature linking infant experience and movement (Adolph & Hoch, 2019; Lobo & Galloway, 2012), we expected that infants’ opportunities to move freely will be positively correlated with infants’ unrestricted self-generated locomotion when they are left unconstrained and therefore free to move.

Method

Participants and Procedure

To examine infant motor movement and restriction, we observed infants and their caregivers in three societies. These videos were collected by three research teams working in each of these three societies as part of larger collaborative projects in each region on family life and child development. We examined video recorded interactions of 52 infants (M age = 12 months, SD = .66, 23 boys, 29 girls) and their caregivers in Tanna, Vanuatu (n = 9; M age = 13.11 months; range = 11–15 months); Central African Republic (Aka; n = 10; M age = 12.1 months; range = 10–12 months); and Tajikistan (n = 33; M age = 12.96 months; range = 11–12 months). The total length of the videos was 3964 min, with an average of 74 min per observation (range = 17–154 min). Overall, the final dataset consisted of 220 videos, and there is variability in the number of videos and the duration of observations between each of the sites (F_2,49 = 7.47, p = .001) due to the nature of data collection for this project. See Table 1 for detailed information regarding the dataset.

Table 1.

Descriptive Information of Datasets.

Population	Infant (n)	Age (Months)		Number of video observations
		M	SD
ni-Vanuatu	9	13.11	1.05	177
Aka	10	12.1	.57	10
Tajikistan	33	12.96	.17	33

The Vanuatu data was collected by a team of researchers working in this region since 2012. While conducting research in this region, the research team (one or two graduate students or professors) resides in one village and travels by foot to each of the surrounding villages with local hosts and local research team for data collection and observation. For this data, the videos were collected by a local individual who was also our research assistant. The researcher identified focal individuals by gathering birthdates for children born the previous year. After obtaining approval from the chief and the local as well as national cultural centers (Tafea Kaljoral Senta), the research team spoke to caregivers and explained the nature of the research and obtained verbal consent to capture their child’s daily life during a series of video recordings. All video recordings were collected during field visits and the field researcher had the objective of capturing video recorded observations of infants and children engaging in their day-to-day activities.

The observations made with families in the Central African Republic were part of a larger, long-term research project by Barry Hewlett. The observations were collected in or around a village in the southern part of the Central African Republic. Aka participants were video recoreded in a naturalistic setting (usually in or near the camp), during one field visit in August and September 2010. Selection was based on age, and all infants of this age within three km of the village were included in the study. The video camera was set up in camp for approximately 30 min before filming started to help diminish attention paid to the camera by the infant and camp members. Caregivers and others in camp were asked to maintain normal activities as much as possible.

The Tajikistan video set was collected by Tajik researchers in collaboration with one of the authors, who developed long-term partnerships with government and NGO agencies in the region to observe and document aspects of family life. Data collection for the Tajikistan dataset began in 2013 over 2 years period. The data collection protocols, coding, and analysis were carried out in close collaboration with the Tajik research team. This partnership ensured that the interpretations remained accurate and culturally informed, reflecting the local context and perspectives.

In each setting, the primary investigator and local research team identified infants who were 12 months of age at the time of the study. After the community and caregivers consented to the study, we video recorded the child at specific intervals throughout the field visit. The consent process is a dynamic one while working in a cultural setting in which one is not a member, therefore we continued the discussion of the project as well as our appreciation for the community participation and effort. We worked with communities at each stage of the project, although it is, at times, difficult to communicate each step when not in the field. We plan to, for example, communicate the findings of this paper to each participating community in the ways in which the community deems appropriate and of value to them. The rationale for selecting these three populations is due to the shared research interests of the authors. Each author was working in a specific region (Tanna, Aka, and Tajikistan) to answer questions about child development and human social life. The dataset consists of a collection of videos gathered at different timepoints by three of the co-authors and their team of students and collaborators for different research programs. The collaboration began as part of a Canadian government-funded project to expand our knowledge of child development beyond urban and western settings. The methods for each society varied slightly, yet the data were comparable in that the goal was to capture a representation of infants’ motor behaviors in their daily life. Although the length of the videos in each society is different, we used the proportions of observed behaviors to make comparisons across the societies. We calculated the proportion of time infants engaged in self-generated locomotion over the entire observation session (accumulated time of infant’s self-generated locomotion duration divided by the total time of unconstrained duration). We calculated the proportion of time infants were restricted by their caregiver or a device during the entire observation session.

Data Coding

To examine our three research objectives, we developed a coding ethogram after watching five videos from each society and refining the definitions for coding these behaviors. After the ethogram (Appendix) was complete, we trained one research assistant to watch all of the videos in one pass and identify four mutually exclusive behaviors: child-generated locomotion, containment by caregiver, containment by device, and falling. To do this, we identified the start and stop time of each of the four categories of behavior of interest using Behavioral Observation Research Interactive Software “BORIS” (Friard & Gamba, 2016). Child-generated locomotion occurred when infants began to move forward by either, that is, bum shuffling, scooting, belly crawling, cruising, and walking. We coded instances when infants were constrained by their caregiver or contained in a device (e.g., gahvora, bathtub, swing, box, and stroller) and were unable to engage in self-generated locomotion. The coding process focused on observable movements rather than subjective or culturally nuanced interpretations. The coding was conducted by research assistants who followed the established coding ethogram to ensure objectivity and reliability.

Inter-rater reliability was achieved by two independent research assistants (RA). The two coders were neither familiar with nor possessed a high level of knowledge of the three cultures being coded. Both coders received high-quality training in coding strategies and were supervised by experts of each cultural context. Each RA coded videos (n = 3) separately after they developed the first version of the ethogram. During training, when minimum levels of agreement were not met (70% agreement), disparities were settled through review of the disputed observation and discussion of ambiguity in the definition of coding behavior, followed by a revision of the definition. Then each coder independently re-coded the video(s) until agreement was achieved. Following the final revision of the ethogram, a primary coder (RA) coded all the videos, and a subset (20%) of randomly selected videos from all three societies were independently re-coded by a second coder. Inter-rater reliability on categorical measures ranged from 96% to 99% and κs ranged from .70 to .85 (ps < .001). The correlation between two coders for durations of each variable ranged from .95 to 1 (p < .001). Disagreements between coders were resolved through discussion.

Results

Research Question 1: Do Infants Experience Different Amounts of Freedom to Move Across Cultures? Are Infants Restricted in Different Ways Across Cultures?

A Shapiro–Wilk test was performed to determine the normal distribution of the proportion of infants who were constrained in three cultural groups separately. The result showed that the distribution of the proportion of time Aka and Tajik infants were constrained departed significantly from normality (W_Aka = .79, p_Aka = .01; W_Tjk = .96, p_Tjk = .29). Based on this outcome, we used nonparametric tests in the following analyses. We conducted a Kruskal–Wallis H test with the different cultural groups as the independent variable and the proportion of time infants were constrained as the dependent variable. The results showed that there was a significant difference in the proportion of time in which infants were constrained between the cultural groups, H (2) = 22.54, p < .001, with a mean rank score of 19.67 for ni-Vanuatu infants, 9 for Aka infants, and 33.67 for Tajik infants. The effect size (η² = .42) indicated a large effect. Dunn’s pairwise tests were carried out for the three pairs of cultural groups. The results showed that there was strong evidence of a difference between the Aka group and Tajikistan group (p < .001, adjusted using the Bonferroni correction). There was also evidence of a difference between the Vanuatu group and Tajikistan group (p < .005, adjusted using the Bonferroni correction). There was no evidence of a difference between the Aka group and Vanuatu group. Indicating that the proportion of time in which Tajik infants were constrained was higher than both ni-Vanuatu and Aka infants. Figure 1 shows the average proportion of infants being constrained of total codable time in three cultural groups.

Figure 1.

Mean Proportion of Constraint Time in Three Cultural Groups.

To further investigate whether caregivers in different societies demonstrate different restriction practices (hold the infant themselves or put infants in a device), we conducted a Kruskal–Wallis H test with the proportion of time infants were constrained by device as the dependent variable and different cultural groups as the independent factor. The results showed that there was a difference in the proportion of time infants were constrained by device between the cultural groups, H (2) = 36.12, p < .001. The effect size was η² = .69, indicating a large effect. The results of Dunn’s pairwise tests showed that there was strong evidence of a difference between the Aka group and Tajikistan group (p < .001, adjusted using the Bonferroni correction). There was also strong evidence of a difference between the Vanuatu group and Tajikistan group (p < .001, adjusted using the Bonferroni correction). There was no evidence of a difference between the Aka group and Vanuatu group. Indicating that Tajikistan infants were constrained by devices significantly longer than Aka and ni-Vanuatu infants. Tajik caregivers restrict their infants mostly by using a device (e.g. baby walker, stroller, and gahvora), whereas ni-Vanuatu and Aka caregivers usually restrict infants by holding infants in arms (Figure 2).

Figure 2.

Tajik Caregivers Restrict Their Infants Mostly by Using a Device, Whereas ni-Vanuatu and Aka Caregivers Usually Restrict Infants by a Caregiver.

Research Question 2: Are There Differences in Infant Self-Generated Locomotion Across Cultures?

A Shapiro–Wilk test was performed to determine the normal distribution of the proportion of infant’s locomotion in three cultural groups. The result showed that the distribution of the proportion of Tajik infant’s locomotion departed significantly from normality (W = .199, p < .001). Based on this outcome, we used nonparametric tests in the following analyses. To investigate whether infants engaged in locomotion when they were not constrained across three cultural groups, we conducted a Kruskal–Wallis test with the different cultural groups as the independent variable and the proportion of time infants engaged in self-generated locomotion (out of total unconstrained time) as the dependent variable. The results indicated that there was no significant difference in the proportion of time infants were in self-generated locomotion out of total unconstrained time between the cultural groups, H (2) = 1.484, p = .476. The proportion of time infants engaged in locomotion was similarly brief across cultural groups (Figure 3). On average, ni-Vanuatu infants spent 20.7% of observation engaging in spontaneous locomotion (SD = 11.3%); Aka infants spent 18.6% of observation engaging in spontaneous locomotion (SD = 13.5%); Tajik infants spent 17.11% of observation engaging in spontaneous locomotion (SD = 17.4%).

Figure 3.

The Average Proportion of Time Infants Engage in Self-Generated Locomotion Out of Total Unconstrained Time in Three Cultural Groups.

Since we had a small sample size for the Vanuatu group and Aka group, we combined these two groups together and tested the normality of the new dataset. The Shapiro–Wilk test was performed and did not show evidence of non-normality (W = .96, p = .575). Based on this outcome, we conducted a one-way ANOVA with the proportion of time infants engaged in self-generated locomotion (out of total unconstrained time) as the dependent variable and the different cultural groups as the independent variable. The results also showed that there was no significant difference in the proportion of time infants were in self-generated locomotion out of total unconstrained time between the two cultural groups, F_1,50 = .298, p =.587.

Research Question 3: What Predicts Infants’ Propensity to Move?

We computed a Pearson correlation coefficient to assess the linear relationship between the proportion of time an infant was in self-generated locomotion out of total unconstrained time and the proportion of time that an infant was being constrained. Results showed that there was no correlation between the two variables (r = –.166, p = .24). There was also no correlation between the two variables in each society.

Discussion

The purpose of this research was to explore how differences in child-rearing practices regarding the ways caregivers structure children’s environment. Specifically, we examined the variability in infants’ daily experiences, focusing on the role of opportunities to move freely and parents’ restrictive behaviors, and their associations with infants’ locomotion development. Indeed, we examined three broad research questions related to these concepts: (a) is there variation in caregivers’ restrictive behaviors across the three cultural groups? (b) is there variation in infant’s self-generated locomotion across the three cultural groups? (c) does more opportunity to move freely lead to more self-generated locomotion?

To fully appreciate the variability in children’s experiences, our research offers unique insight within this area of developmental research by observing a snapshot of infants’ early motor experience across three societies are not well represented in the developmental literature. For the first research question, our results showed that Tajik infants are restricted more than infants in Aka and Vanuatu. This is in line with the previous literature which finds that Tajik infants spend about 6 hr to 19 hr in the gahvora per day (Karasik et al., 2018). We also noticed that when the infant was unrestricted, Tajikni-Vanuatu caregivers often preferred to designate a small, defined area for their infants’ movement, such as placing the infants on a rug or between two caregivers. This shows that although infants were not directly restricted by their caregivers, their moving space was still limited. On the contrary, we observed that and Aka caregivers put infants outside the house with their community members, and infants were allowed to move freely throughout the neighborhoods, while being supervised by the caregiver and/or other members of the community. These variations in parental choices for daily settings and the way in which they shape children’s immediate environment provide valuable information about how caregiving practices shape the context of motor development across cultures.

Caregivers in many societies actively discourage exploration to guard children against environmental hazards; but the ways infants are constrained may vary significantly between cultural groups. Our results showed that caregivers in different societies preferred different restrictive practices. In more detail, Tajik caregivers restrict their infants mostly by using a device (traditional cradle name “gahvora”), whereas ni-Vanuatu and Aka caregivers usually restrict infants by holding them. In Tajikistan and presumably other parts of Central Asia, gahvora cradling is a traditional cultural child-rearing practice. From our observations, Tajikistan caregivers used different devices to help them restrict their infants, including a cradle, stroller, and baby walker. However, Aka and ni-Vanuatu caregivers were not observed using tools to restrict their infants. In contrast to Tajik infants, manufactured items are rare in Aka and ni-Vanuatu infants’ daily life. From our observations, most caregivers living in the Tanna restricted infants’ locomotion because of daily tasks, such as changing infants’ clothes, bathing, toileting, or feeding. Aka caregivers rarely actively restrict infants’ movement, and sometimes they hold infants to respond to them. Although we did not directly examine caregivers’ intention or attitude toward why they restrict infants’ locomotion, our work suggests that caregiver restriction practices may serve different roles in each society.

Tajik infants are restricted more than infants living in Aka and Vanuatu, however, the proportion of time infants engaged in locomotion was similarly brief across three cultural groups. Notably, although Tajik infants are restricted more than infants in Aka and Vanuatu, they are exposed to a similar rich visual and social environment and highly involved in the daily interactions with peers and adults. According to Karasik et al. (2023)’s study, the authors also reported that the time in the gahvora does not inform on infants’ opportunities to move when unconstrained, and infants always take advantage of time to move and spend much of their time on the ground.

In addition, our results indicated that there was no relation between the time infants engaged in self-generated locomotion and the time they were constrained. However, regarding our third research question, the current study provides follow-up results concerning whether variation in time being constrained relates to infants’ developing motor skills. From Gibson’s (1988) perspective, human evolves within specific environments, occupying an ecological niche uniquely suited to its adaptations and engaging in reciprocal interactions with its surroundings. In the context of motor development, the early environment offers children resources and opportunities for action. One previous study identified the affordances and stimuli available in the home as the most important predictors of gross-motor skill development (Valadi & Gabbard, 2020). Instead of focusing on the home environment, we observed the availability of infants’ moving opportunities in both indoor and outdoor environments across different societies. There has been considerable discussion of the possible influence of various child-rearing practices, and different societies hold divergent views about the effect of restricting infant’s movement. However, it is essential to consider a particular child-rearing practice from the cultural context in which it is employed. A recent study investigated the implications of restricted movement for motor development in Tajikistan infants (Karasik et al., 2023). Their findings showed that fewer Tajik infants demonstrated motor skills compared with same-age infants represented on the World Health Organization standards, however, the effect size linking cradling and motor development was mostly small to moderate. Current study’s results may provide evidence to help explore the range of infant experiences out of the gahvora. The proportion of time that infants engaged in locomotion when they were free to move similarly across three cultural groups. Infants take advantage of the opportunities available to them during different cultural activities despite vastly different child-rearing practices. In fact, freedom to move is a central tenet of most caregiving practice; however, to what extent and how long infants’ movements are restricted by caregivers can vary widely by culture and context.

The present study explored the variability of child caregivers’ restriction behaviors in three different societies, providing insights into the diversity of early experiences and their associations with motor development. Previous studies have found that restriction hours are related to motor status and proficiency to varying degrees. Findings from this study uniquely contribute to reveal the role of early experience in motor development and highlight how different societies adapt their practices to environmental and cultural demands.