Application of the COM-B model to the correlates of children’s outdoor playing and the potential role of digital interventions: a systematic literature review

ABSTRACT

Children’s reduced engagement in outdoor playing in recent years has contributed to increasing academic and practice interests in understanding this behavior, as well as investigating the effectiveness of combined social, physical and digital interventions on stimulating children’s outdoor playing. This paper provides a systematic review of recent empirical evidence on the correlates of the outdoor playing behavior of children, ages 4–12. In addition, the potential roles of digital interventions in stimulating children’s outdoor playing are explored. The COM-B behavior change model is used to establish relevant correlates and functions of digital interventions. COM-B model defines behavior as the result of an interaction between three components: capability, opportunity and motivation. This model provides a basis for designing effective behavior change interventions. This paper’s contribution is twofold: it presents the case for adding ‘digital environment’ as a new component of the COM-B model, and it further develops a conceptual framework of different functions of digital interventions aiming at stimulating children’s outdoor playing behavior. The findings contribute to the theory-based behavior change interventions stimulating children’s outdoor playing.

KEYWORDS:

• Outdoor playing behavior
• children
• COM-B model
• behavior change
• digital

1. Introduction

The provision of opportunities for children to be involved in outdoor playing appropriate to their age is increasingly recognized within research and policy communities as a basic right, and is fundamental to a child’s physical, social, emotional and cognitive health development (Villanueva et al. 2016). The outdoor playing behaviors of children are defined as unorganized, fun, voluntary and child-initiated outdoor activities, which are mostly played by children ages 4–12 years old. Children from this age group are often physically active and autonomously beginning to explore their local environment, where they spend large periods of time outside of home and school (Elshater 2018).

Active outdoor playing has received considerable attention in the environment and health literature (Villanueva et al. 2016). However, children’s outdoor playing includes a wide range of activities. For example, young children (ages 5–8) are more interested in exploring, creativity, imaginary role-play, learning and problem-solving (Cumbo et al. 2014; Crawford, Holder, and O’Connor 2017). On the other hand, among older children (ages 8–12), there is more tendency to physical activity and mobility and compete in bigger groups (Piaget 1962).

Children’s outdoor playing is affected in different ways in today’s digital age. Children spend more time indoors using screen media compared to the previous generations who used to play outside with friends, neighbors and siblings. It is shown that pervasive digital technology can negatively impact children’s motor, social and emotional skills (Felix et al. 2020; García-Hermoso et al. 2020). Despite these negative effects, several studies have claimed that technology, especially in terms of modern mobile and sensor technologies integrated with the physical and social environments, can positively influence children’s health-related behavior change.

These new technologies include context sensing and augmented reality and have led to the emergence of pervasive games: a genre in which traditional, real-world games are augmented with computing functionality or purely virtual computer entertainment is brought back to the real world (Soute, Markopoulos, and Magielse 2010). A well-known pervasive game is ‘Pokémon Go’, a location-based augmented reality game that uses mobile devices to locate, capture, train, and battle virtual creatures in the players’ real world.

It is argued that new technologies can provide children with opportunities to explore, create and engage in relationships with both physical and virtual objects and other people (Karjalainen et al. 2016). They can also be used to build communities of local families and children around their local place or assist parents in engaging their children in meaningful interactions with the local environment (Cumbo et al. 2014; Moyse 2019). Based on the emerging literature, we assume that digital interventions have the potential to promote a change in children’s activities from indoor sedentary activities (e.g. gaming) to outdoor playing. Subsequently, the focus of this review is on the potential positive roles of digital interventions in children’s outdoor playing.

To understand and explore the mechanism of this change, first, the children’s outdoor playing activities and the correlates of this behavior need to be elaborated, as health behavior change interventions can be more effective if grounded in an appropriate theoretical understanding of behavior (Davis et al. 2015). Thus, the correlates of children’s outdoor playing need to be linked to a behavior change model in order to understand and investigate the mechanism of behavior change.

To date, a number of literature reviews exist on investigating some of the correlates of children’s outdoor playing. Lee et al. (2015) conducted a meta-study of qualitative research examining the determinants of children's independent active free play. Boxberger and Reimers (2019) synthesized the parental correlates of outdoor play. Lambert et al. (2019) examined the relationship between neighborhood-level built environment attributes and children’s time spent in self-directed outdoor play.

To the best of our knowledge, a comprehensive review, including different individual/socio-physical/environmental correlates of outdoor play (that does not explicitly focus on active play), does not exist yet. In addition, the existing reviews have only focused on investigating the correlates of children’s outdoor playing behavior, and not how an intervention can prompt behavior change by addressing behavior components. Therefore, a behavior change model with an individual level mechanism is needed to explore how to promote children’s outdoor play while considering the behavior components.

To organize the correlates of children’s playing, Lee et al. (2015) and Boxberger and Reimers (2019) used a social-ecological model. When addressing behavior, an ecological approach is needed to understand the multiple levels of influence ranging from the individual and interpersonal to organizational, community and policy levels (Sallis, Owen, and Fisher 2008). The social-ecological model has evolved around the core concept that behavior is influenced by multiple levels and that an exploration of the reciprocal interaction between these levels is the most effective approach to understanding behavior (Mitra, 2013). While the social-ecological model is a useful framework for understanding the multiple levels of influential factors on a behavior (Sallis, Owen, and Fisher 2008), it does not provide a framework for behavior change (Van Kasteren, Lewis, and Maeder 2020).

There are many theories and models of behavior change interventions such as ‘Fogg Behaviour Model’, ‘Transtheoretical (Stages of Change) Model’ or ‘Theory of Planned Behaviour’. These models, however, do not elaborate on mechanisms through which interventions can prompt behavior change. The COM-B model is a behavior change framework with an underlying theory that explains individual-level mechanisms of behavior change (Michie, van Stralen, and West 2011). It posits that interventions change behavior through different ‘functions’ (further explained below). Using this model enables us to apply a theoretical framework to an intervention. Thus, we can explain how correlates of behavior are related to each other and how they are related to the mechanism of actions of an intervention.

The Capability, Opportunity, Motivation, Behavior (COM-B) model of behavior change (Figure 1) provides a comprehensive framework for understanding behavior, as well as designing behavior change interventions. This model was developed with reference to existing theories of behavior and applies to all behaviors (Michie, van Stralen, and West 2011). It has been effectively applied to understand children’s health behaviors, such as exercising, in the domain of medical and education science (Taylor et al. 2016), as well as in the design and measure of the effectiveness of an intervention to improve children’s physical activity in the domain of human–computer interaction (Cibrian, Tentori, and Martínez-García 2016).

Figure 1. The COM-B model (Michie, 2011).

COM-B theorizes behavior as the result of an interaction between three components: capability, opportunity and motivation. In other words, for a particular type of behavior to occur, a person must have the capability and the opportunity to engage in that behavior and be motivated to enact that specific behavior rather than any other behavior (Michie, van Stralen, and West 2011). This model takes into account the physical and social contexts, physical and psychological capabilities and children’s ‘automatic’ and ‘reflective’ motivational processes. Motivation is the core of the model. Thus, it is a component that attaches to any type of behavior that capability and opportunity would permit in a given situation. In other words, the more capable people are in enacting a type of behavior and the more conducive the environment is to enact it, the more motivated they are to do it (West and Michie 2020).

Based on the COM-B model, interventions change behavior through different ‘functions’ (e.g. enablement, environmental restructuring, training, etc.). The definitions of these functions are provided in Table S2 (see Table S2 of supplementary material). With the applicability of the COM-B model outlined above, this article has three objectives: (1) provide a comprehensive review of different individual/parental/social/built environment correlates of children's outdoor playing behavior; (2) apply the COM-B model to map the correlates of children's outdoor playing behavior and the functions of digital interventions that aim to promote this behavior; and (3) develop a new conceptual framework that includes the role of digital interventions and its different functions in the COM-B model.

2. Method

2.1. Design and search strategy

A large number of studies have investigated children's outdoor playing in different fields, such as psychology, education, public health and environmental design. In line with the research goals, a systematic literature review was conducted to obtain the most relevant articles. The systematic review process consisted of performing a search string of specific keywords in Scopus, that is, (child, kid) AND (outdoor/free/unorganized/social/active play) and their variations. Both studies investigating correlates of children’s outdoor playing and digital interventions stimulating this behavior were extracted. The systematic review follows the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) (Moher et al. 2009).

2.2. Eligibility criteria and study selection

Figure S1 of the supplementary material presents the process of inclusion and exclusion of papers in the review. The study selection process was conducted in three steps: (1) title screening, (2) abstract screening and (3) full-text screening. After the abstract screening, two main types of studies were identified. The first involves studies investigating the correlates of children’s outdoor playing, and the second involves studies exploring the functions of digital interventions in stimulating children’s outdoor playing. The eligibility criteria for selecting the studies investigating the correlates of children’s outdoor playing were five-fold. First, the study should include outdoor playing as a variable. Second, the study population should consist of healthy children ages 4–12 or an average age in this range. Third, the article should present evidence-based findings on the relationship of individual/environmental determinants with the quantity and/or quality of outdoor playing in children. Fourth, studies with different research designs (including both cross-sectional and longitudinal) were included. Fifth, only the English peer-reviewed and more recent articles (since January 2010) were included.

For the studies on the potential role of digital interventions in stimulating children’s outdoor playing, the eligibility criteria were more flexible, as this is an emerging field with few studies on the subject. The first criterium required that the study population consisted of healthy children aged 4–12 or with an average age in this range. Second, exploratory studies with different research designs were reviewed. This includes both studies that report the design process of digital interventions and those that test the feasibility of digital interventions in stimulating children to play outside. Third, only studies investigating the role of technology on children's outdoor playing within outdoor environments were chosen. Fourth, both English peer-reviewed journal articles and conference papers since January 2010 were included.

The studies for full-text screening were screened by considering the eligibility criteria. If all mentioned criteria were met, the article was included in the systematic review. Additional articles were identified by forward and backward snowballing, where the relevant articles citing the selected works and the references of the selected works were considered.

2.3. Extraction and classification of the correlates of children’s outdoor playing

The first type of eligible studies was scanned for the correlates of children’s outdoor playing behavior. Correlates are individual and environmental determinants of children’s outdoor playing and their associations with children’s outdoor playing were investigated. All influential correlates, as identified by the literature, are included in order to not overlook any potential determinants. Overall, nine correlates with their sub-categories were identified from a total of 33 studies. The identified correlates were categorized into four groups based on the definition of the Michie, van Stralen, and West (2011) of the COM-B model components. The definition of components and their sub-components were used to establish codes. These codes were then used to distinguish the correlates. The description of the codes is as follows:

1. If the correlate relates to the child’s psychological and physical capacity to engage in outdoor playing, it was mapped into the capability component. This component includes a child’s physical and psychological ability to play freely and understand the environment (e.g. independent playing and perception of the environment); (2) If the correlate relates to the social-cultural factors and the opportunities afforded by the physical environment that lies outside the individual and make a child’s outdoor playing behavior possible or prompt it, it was mapped into the opportunity component. This component includes neighborhood social capital (e.g. the presence of friends, feeling connected to a community) and the parents’ social correlates (e.g. parents’ socio-cultural values and economic status), the availability of playing facilities (e.g. playgrounds and parks and natural playing areas) and safety and walkability attributes (e.g. traffic safety, the quality of pedestrian facilities and walking distance to play areas and schools); and (3) If the correlate relates to attributes that energize and direct a child’s outdoor playing behavior, it was mapped into the motivation component (e.g. parental perception and practices of the family).

In addition to coding the COM-B model components for the analysis, we counted the share of studies investigating a specific variable and the share of studies that find a significant association. Based on this numeric analysis, we concluded which variable was studied more and which had a significant influence. Table S1 shows the share of studies that investigated a specific correlate and those that found it significant, divided into COM-B model (sub) categories. Furthermore, the significance of the association of each correlates with children’s outdoor playing and the direction of the relationship (positive/negative) is shown (see Table S1 of the supplementary material).

2.4. Extraction and analysis of the functions of digital interventions

The second type of eligible study was scanned for exploration of the functions of digital interventions that stimulate children’s outdoor playing. This includes the potential functions identified in the behavior change wheel introduced by Michie, van Stralen, and West (2011). A thematic analysis was performed to identify the functions of digital interventions in relation to children’s outdoor playing. This is a qualitative data analysis technique that uses theory as its point of departure (Pearse 2019). The analysis process includes two general steps, that is, theme development and coding. Themes are generally broader than codes. Here, themes are functions of the behavior change mechanism described by Michie, van Stralen, and West (2011). The deductive codes were derived from the description of each theme and its synonyms. Table S2 of the supplementary material shows the identified themes and codes for the thematic analysis.

These codes were used to distinguish the intervention functions. Functions were then used to identify which COM-B component was addressed by the intervention based on the study by Michie, van Stralen, and West (2011). According to Michie, van Stralen, and West (2011), capability can be achieved by training, education and enablement functions. Physical and social opportunities could be provided by the function of environmental restructuring, enablement and restriction. Motivation can be achieved through education, persuasion, coercion, environmental restructuring, enablement, modeling and incentivization functions. Restriction and coercion were not found as the functions of digital interventions stimulating children’s outdoor playing. Training and education were considered together as ‘training’ function as their descriptions and examples are very similar. Table S3 of the supplementary material shows the relationships between functions and COM-B model components.

3. Results

3.1. The correlates of children’s outdoor playing behavior and the potential functions of digital interventions based on the COM-B model

This literature review includes the 33 studies investigating the correlates of children’s outdoor playing and the ten studies on the potential roles of digital interventions in stimulating children’s outdoor playing. The reviews of these studies were first done separately regarding their different purposes and methodologies. Overall, nine correlates are identified from the 33 studies, and their association with children’s outdoor behavior are discussed under the COM-B model categories of capability, motivation, and opportunity (see Table S1 of the supplementary material). Moreover, six functions of digital interventions are recognized to contribute to stimulating children’s outdoor playing and are presented under the same COM-B model categories (see Table S3 of the supplementary material). The results, including the correlates of children’s outdoor playing and functions of digital interventions, are presented, discussed and outlined by the COM-B model categories of capability, opportunity and motivation.

3.1.1. Capability

Capability is defined as the individual’s psychological and physical knowledge and skills to engage in an activity (Michie, van Stralen, and West 2011). Two correlates were identified as necessary capabilities of children’s outdoor playing behavior.

Children’s independent playing

Children’s independent playing depends on both physical and psychological capability of playing outside. Children’s independent playing is defined as their ability to move around and play by different levels of freedom and includes unsupervised mobility and playing and supervised playing. Studies show that children with higher independent mobility are more likely to report playing outside every day compared to those who played outside less frequently (Page et al. 2010; Moran, Plaut, and Merom 2017). Also, frequent, independent walking activities by children are confirmed to lead to more play activities, because when they meet their friends while walking around, many cases of playing together occur immediately, even if initially there was no plan to play (Lee et al., 2019). On the other hand, Moran, Plaut, and Merom (2017) report the presence of parents and grandparents supervising younger children in the park as a barrier since children are often asked not to play around too fast or hurt younger children. Moreover, Loebach et al. (2021) observed that the duration of children’s outdoor playing was less likely among those who were not allowed to play beyond home without supervision. Therefore, a lower level of children’s independent playing is related to fewer children’s outdoor playing.

Children’s perception of the neighborhood

Children’s perception of the local environment is particularly associated with inspiration for taking part in outdoor playing (Pluhar et al. 2010). Children’s perceptions of the neighborhood are more about the neighborhood’s perceived safety and child-friendliness. Children’s positive perceptions regarding safety is reported to have a direct relationship with the outdoor free play of children (Kemperman and Timmermans 2011; Page et al. 2010). It has also been shown that the outdoor playing of children is positively associated with children’s perceptions of the environment as ‘child friendly’, which is defined by the extent of the relationship between the availability of things to do and ability to see the levels of freedom a child has to explore and enjoy playing (Moran, Plaut, and Merom 2017; Wang et al. 2020). In addition to safety and child-friendliness, child’s higher perception of the benefits of outdoor playing was reported with more time playing outside (Loebach et al. 2021). Children's negative perceptions of the social safety of the neighborhood was also reported as a restriction on the type of play activity. It was observed that children’s free play is restricted when they feel unsafe, as it requires them to play with a companion rather than alone (Lee et al., 2019; Wang et al. 2020).

The potential functions of digital interventions for stimulating children’s capabilities

Physical and psychological capability can be achieved by training and enablement functions (Michie, van Stralen, and West 2011). An example of the enablement function is facilitating parental involvement in intervention through a mediator role. This can be achieved by mobile device-based games. Patten, Iarocci, and Bojin (2017) demonstrate this function in the case of a playground that is connected to mobile games through a marker. Here the device is the playground, and the mediator role involves assisting children with the use of the playground by scanning the marker or holding the mobile with a game open and delivering playful directives that the child can follow.

An example of training to increase children’s capabilities is giving children the ability to control the playing situation through a controller. For example, Begole et al. (2015) introduce a bracelet as a controller device that turns everyday objects into interactive physical gameplay. Another example is the use of IoT prototypes that could be programed, configured and built by children to shape and create play spaces in their neighborhood (Bernhaupt et al. 2020).

3.1.2. Opportunity

Opportunity is defined as all the factors that lie outside the individual and prompt behavior. Physical opportunity is afforded by the environment and social opportunity is afforded by the cultural and social background (Michie, van Stralen, and West 2011). Six correlates were recognized as social and physical opportunity correlates of children’s outdoor playing.

Neighborhood social capital

Neighborhood social capital is defined as providing opportunities for children to socialize through communication with friends and adults and feel connected and attached to a group or community. Several studies have reported that social capital indicators have a positive impact on children’s outdoor playing behavior. These indicators are the presence of more friends during playing in the neighborhood, having siblings to play with, having a greater sense of community consciousness and social cohesion (Aarts et al. 2010; Page et al. 2010; Veitch, Salmon, and Ball 2010; Bringolf-Isler et al., 2010; Aggio et al. 2017; Reimers et al. 2018; Lee et al., 2019; Orr and Caspi 2020; Wang et al. 2020; Loebach et al. 2021).

Socio-cultural norms and values

Social norms and socio-cultural values regarding children’s outdoor playing include parental cultures and restrictions. Many studies have reported that boys spend significantly more time outside playing compared to girls (Bringolf-Isler et al., 2010; Remmers et al. 2014; Faulkner et al. 2015; Moran, Plaut, and Merom 2017; Aggio et al. 2017). This is probably because parents are more cautious with their female children compared to males, as confirmed by Kepper et al. (2020). It is also observed that the presence of boys in the playground suppresses girls’ outdoor playing, indicating that girls could be distracted or bullied by boys (Reimers et al. 2018). However, one study shows that there is no relationship between gender and participation in the various types of recreational activities (Kemperman and Timmermans 2011). The difference in results is likely due to the different cultural settings that dictate the attitudes and beliefs about the types of activities considered appropriate for different sexes, ages, and groups (Yoon and Lee 2019; Bringolf-Isler et al., 2010; Wang et al. 2020).

Parental socioeconomic status

Parental socioeconomic status includes the household’s education and income levels. The higher level of parental education is reported to be negatively associated with children’s outdoor playing (Dodd et al. 2021; Mölenberg et al. 2019). In addition to education level, the higher level of economic status and the number of cars/driver’s licenses are consistently associated with less outdoor playing in children (Aggio et al. 2017). These results might be explained by the fact that higher-educated parents have more financial resources for organized sports activities, which substitutes for the time spent on outdoor playing (Aarts et al. 2012; Kemperman and Timmermans 2011). The other reason is that lower-educated parents might live in smaller houses, which makes it more likely for children to play outdoors (Aarts et al. 2012).

The walkability of the neighborhood

The concerns of the walkability of a neighborhood include traffic safety and the quality of pedestrian facilities, as well as walking distance to play areas and schools. Measures to increase traffic safety, such as the presence of traffic lights, speed bumps and home zones, are reported to have positive relationships with outdoor playing (Aarts et al. 2012). Additionally, the diversity of routes (Aarts et al. 2010) and more pedestrian facilities, such as fewer path obstructions and the availability of sidewalks, are also positively associated with outdoor playing (Lee et al. 2016; Wang et al. 2020).

On the other hand, lower traffic safety, measured by different proxies, is shown to discourage children’s outdoor playing. For instance, high street connectivity, which means more street intersections, is found to be negatively associated with the outdoor playing of children (Aarts et al. 2012; Yoon and Lee 2019). Similarly, increases in the street density in more urbanized areas are inversely associated with the outdoor playing time among younger children (Bringolf-Isler et al., 2010; Wang et al. 2020). Also, the amount of children’s outdoor playing decreases when the playing areas and schools are in farther walkable distance of children’s homes (Page et al. 2010; Kemperman and Timmermans 2011; Faulkner et al. 2015; Bhuyan 2022).

Greenery

Many studies have examined the relationship between the availability of green or natural resources in the neighborhood and the outdoor playing behavior of children. The results show that living in a green neighborhood and the presence of water features are significantly associated with more outdoor playing (Bringolf-Isler et al., 2010; Aarts et al. 2010; Kemperman and Timmermans 2011; Yoon and Lee 2019). Moreover, Luchs and Fikus (2013) report that children in nature-oriented playing spaces play longer within one play episode, indicating that they have more concentration during playing.

Outdoor playing areas

Outdoor playing areas correlate with concerns about the availability of formal and informal play spaces, as well as the quality, size, layout and maintenance of play areas. Regarding formal play areas, studies indicate that accessibility to play facilities (Remmers et al. 2014; Wang et al. 2020) is positively associated with the play time duration and frequency. Moreover, the spatial layout of the playground and high quality and well-maintained parks and playgrounds are identified as having a positive impact on children’s outdoor playing (Moran, Plaut, and Merom 2017; Sumiya and Nonaka 2021). The presence of informal play areas such as sidewalks or cul-de-sacs and areas created in the common areas of residential buildings are also found to positively impact children’s outdoor playing (Veitch, Salmon, and Ball 2010; Aarts et al. 2012; Moran, Plaut, and Merom 2017).

However, Moran, Plaut, and Merom (2017) report that children’s outdoor playing is less common in suburban neighborhoods compared to inner-city neighborhoods because there are fewer formal play areas. In addition to limited access to play areas, low quality, poor maintenance and the lack of playing structures within playgrounds are described as barriers to children’s outdoor playing (Moran, Plaut, and Merom 2017). As for informal playing areas, living in a flat or apartment without a yard or a garden is negatively associated with outdoor playing among younger children (Aarts et al. 2010).

The potential functions of digital interventions for providing social and physical opportunities

Physical and social opportunities could be provided for children’s outdoor playing by the function of environmental restructuring (Michie, van Stralen, and West 2011). Examples of applying changes to the physical context are interventions that add interactive installations to a schoolyard (Wood et al., 2019) or embedding digital objects into trees and underneath flowers in a park (Begole et al. 2015). Examples of environmental restructuring within a social context include creating a communication method between players using a network of devices (Ofer, 2019; Soute, Markopoulos, and Magielse 2010) or considering communication nodes within the playing area (Back et al., 2016). Furthermore, socializing through playing promotes social opportunities for children’s outdoor playing (Rosales, Arroyo, and Blat 2011; Patten, Iarocci, and Bojin 2017; Blikstein et al. 2017; Delprino et al. 2018; Bernhaupt et al. 2020).

3.1.3. Motivation

Motivation includes automatic and reflective processes that energize and direct behavior. Automatic motivation arises from intrinsic preferences (Michie, van Stralen, and West 2011). A number of studies have investigated automatic motivation (Lepper, Greene, and Nisbett 1973; Malone 1981; Cumbo et al. 2014). For instance, Cumbo et al. (2014) show that children’s automatic (intrinsic) motivation is displayed in their preferences for exploring, creativity, imaginary role-play, learning and problem-solving, rule-based and skill-based team games. However, investigate the relationship between automatic motivation and the quantity and/or quality of outdoor playing of children are lacking. Two correlates were found under the category of reflective motivation, which is about the external drivers of children’s outdoor playing. Two correlates were found under the category of reflective motivation, which is about the external drivers of children’s outdoor playing.

Parental perception

Parental perception has been classified as reflective motivation due to the way it influences children’s perception and motivation to playing outside. For instance, when children hear parents speak about the harmful aspects of the environment, this may affect their motivation for taking part in outdoor playing (Pluhar et al. 2010). Parental perception includes their attitudes toward neighborhood attributes and outdoor playing and safety concerns. Several studies show that a parent’s positive perception of some factors is positively associated with more outside playing. These factors include parental attitudes towards outdoor playing, social capital, the attractiveness of the neighborhood, accessibility to playing area and facilities and a neighborhood’s level of safety (Aarts et al. 2010; Veitch, Salmon, and Ball 2010; Kimbro et al., 2011; Remmers et al. 2014; Mcfarland, Zajicek, and Waliczek 2014; Carmo et al. 2020; Roberts et al., 2016; Yoon and Lee 2019; Parent et al., 2020; Dodd et al. 2021; Loebach et al. 2021). On the other hand, several studies report that parental concern about traffic safety, social safety such as crime and strangers, non-availability of proper playing spaces and screen time are associated with less time playing outdoors in primary school children (Bringolf-Isler et al., 2010; Remmers et al. 2014; Faulkner et al. 2015; Carmo et al. 2020; Yoon and Lee 2019).

Practices of the family

Practices of the family regarding children’s outdoor playing include social support and encouragement of family, friends and others. Remmers et al. (2014) and Xu et al. (2016) have found that supportive parents with high responsibility toward the outdoor playing of their children are more likely to encourage playing outside every day. Moreover, parental encouragement for outdoor play in the form of verbal cues is also recognized to be positively associated with the outdoor active playing of children (Ferrao and Janssen 2015). Regular visits accompanied by the family to playing spaces is also observed to have a positive association with children’s outdoor playing (Veitch, Salmon, and Ball 2010; Chen et al., 2019). Furthermore, the presence of pets in the household is suggested to support children’s outdoor playing as practical action. Two studies test this hypothesis and confirm that family dog ownership is positively correlated with outdoor play in children (Christian et al., 2014; Yoon and Lee 2019). The rationale is that children who have a dog have a companion to play with (Yoon and Lee 2019) and are more independently mobile than children without one (Christian et al., 2014).

The potential functions of digital interventions for stimulating children’s motivation

Motivation can be achieved through training, persuasion, modeling and incentivization functions (Michie, van Stralen, and West 2011). Examples of applying training functions are digital interventions that provide children with a coding platform that empowers them to become game designers through an interactive learning environment (Ofer, 2019; Blikstein et al. 2017). Another notable example is an intervention that helps children capture the images of the elements they find interesting in the physical environment and interact with the collected digital materials to obtain new knowledge (Delprino et al. 2018). Incentivization function motivates children by the act of rewarding. Rewarding is one of the main gamification strategies for motivating players who accomplish desired tasks. Several interventions use this technique in different ways, such as point-based rewards in the form of a score, feedback message and light or sound (Ofer, 2019; Wood, Dylan, Durrant, et al., 2019; Begole et al. 2015; Rosales, Arroyo, and Blat 2011).

Persuasion is another function of digital intervention that is found to have a positive effect on children’s outdoor playing. Persuasion can occur through stimulating different senses. Rich sensory experiences, fantasy elements, and reward structures are methods of persuading players to play more (Soute, Markopoulos, and Magielse 2010). An example is adding sound or music and light to the physical environment to make it inviting for children (Begole et al. 2015; Back et al., 2016). Another way to persuade children is to give them opportunities to create and control the environment according to their needs. This function is present in many interventions that provide children with ways and scenarios to invent their games and play activities (Wood, et al., 2019; Rosales, Arroyo, and Blat 2011; Soute, Markopoulos, and Magielse 2010; Begole et al. 2015; Blikstein et al. 2017; Delprino et al. 2018; Bernhaupt et al. 2019; Ofer, 2019). Examples are functional features of IoT resources that invite particular social behaviors in the context of open-ended play (Wood, et al., 2019; Bernhaupt et al. 2019).

Motivation can also be achieved through modeling, which can provide an example for children (Michie, van Stralen, and West 2011). The model can have a virtual or a real character. For instance, one reviewed intervention encourages children to perform an outdoor playing activity according to an imaginary premise (Patten, Iarocci, and Bojin 2017). Here, the modeling function is used to motivate children to engage in pretend play, which nurtures a child’s imagination.

4. Conclusion and discussion

This review of literature, conducted through the lens of the COM-B model, is the first to investigate the correlates of outdoor playing behavior of 4- to 12-year-old children. COM-B model not only provides a comprehensive framework for understanding the multiple correlates of children’s outdoor playing behavior, as demonstrated in this work, it is also useful in identifying the individual-level mechanisms of behavior change prompted by digital interventions. Based on the COM-B model, in order for outdoor playing behaviors to occur, a child must have the capability and motivation as well as opportunities provided by physical and social environments.

The review on the correlates of children’s outdoor playing includes 33 peer-reviewed articles and provides an overview of individual and environmental correlates of outdoor playing behavior mapped onto the COM-B model. Overall, the results indicate that different levels of these correlates either have a positive or negative impact on the outdoor playing behavior of children through increasing children’s capability and motivation and providing them with social and physical environment opportunities.

In addition, findings from the ten studies on the potential roles of digital interventions stimulating children’s outdoor playing show that digital interventions (including smart interactive objects and pervasive game play) can facilitate children’s outdoor playing through a number of functions. These functions could influence children’s capability, motivation and opportunity. In terms of capability, children’s independent playing and children’s perception of the neighborhood are identified as individual correlates of children’s outdoor playing behavior. Digital interventions were found to possibly stimulate children’s capabilities by training and enablement functions.

In terms of opportunity, neighborhood social capital, socio-cultural norms and values and parental socioeconomic status are recognized as social environment correlates, as well as the walkability of the neighborhood, and the presence of greenery and outdoor playing areas as physical environment correlates. Digital interventions can potentially contribute to social and physical opportunities through their environmental restructuring function. In terms of motivation, parental perception and practices of the family were found to be reflective motivational factors. Children’s motivation for outdoor playing can be stimulated by digital interventions through training, persuasion and incentivization.

The findings of this study are consistent with Lambert et al. (2019) and Lee et al. (2015) in indicating that the availability of formal and informal playing areas and natural resources, as well as higher levels of neighborhood social capital are the main facilitators of children’s outdoor playing. Moreover, we found that parental positive attitudes towards neighborhood attributes, outdoor playing, the social support and encouragement of family, friends are other correlates that have been reported by more studies as facilitators of children’s outdoor playing. On the other hand, in line with Lee et al. (2015) and Boxberger and Reimers (2019), it is found that parental safety concerns are reported more as barriers to children’s outdoor playing rather than facilitators. In addition, we found that farther walking distance to playing areas and schools is another main barrier for children’s outdoor playing.

Moreover, reviewing the studies on the roles of digital interventions showed that the majority of the interventions stimulated children’s outdoor playing through promoting their motivation and creating physical and social opportunities. The findings corroborate the claim of Cumbo et al. (2014), Crawford, Holder, and O’Connor (2017) and Karjalainen et al. (2016) on the effectiveness of applying technology as a motivator for children to engage with digital as well as physical and social environments. A minority of interventions aimed to upgrade children’s capabilities to play outdoors more often by enablement and training functions.

This literature review shows that the digital environment can provide new opportunities for children’s outdoor playing, through new pervasive technologies such as context sensing and augmented reality. However, the original COM-B model does not include the digital environment as a component of its behavior change system. Therefore, this study suggests updating the COM-B model to include the digital environment as an opportunity component besides the social and physical environments (Figure 2).

Figure 2. The digital intervention functions related to COM-B model components.

The digital environment provides opportunities such as rich sensory experiences, fantasy and narrative elements and reward structures. These opportunities are enabled by new technologies such as pervasive games and sensors and differ from those provided by the physical and social environments. Thus, the digital environment is a distinct component that can be integrated with the physical and social environments and can stimulate children’s outdoor playing through different functions.

This extended version of COM-B model, including the digital component, contributes to the theory of behavior change and is beneficial for public health and health geography researchers as well as policymakers aiming to promote healthy behavior. Furthermore, mapping the identified correlates of children’s outdoor playing onto the COM-B model and updating it to accommodate the digital environment has two major benefits for the practitioners and in particular interactive designers. First, it helps them to design effective interventions based on the identified correlates of children’s outdoor playing. For instance, a design to encourage independent play should consider children’s physical and psychological capabilities. Second, it inspires them to consider digital interventions, not as a separate feature but a layer in combination with the physical and social environments.

Thus, it is important to find ways to incorporate the digital environment into other existing traditional playgrounds or natural resources. Furthermore, digital interventions should consider contextual cultural and social differences. It is noteworthy that here digital intervention functions are derived from the studies that have illustrated an intervention or assessed children’s attitudes toward that intervention. However, evidence-based studies that investigate the effectiveness of digital interventions on children’s outdoor playing are lacking and much needed. Such studies can help (i) identify the digital intervention correlates of outdoor playing and (ii) assess their effectiveness for promoting children’s outdoor playing. Figure S2 shows an overview of the correlates of children’s outdoor playing and the functions of digital interventions in relation to COM-B model components. It also presents the gaps for future research based on the reviewed studies.

5. Recommendations for future research

The current review provides some suggestions to increase the understanding of children’s outdoor playing and digital interventions promoting this behavior. First of all, although this review provides a comprehensive framework of the correlates of children’s outdoor playing behavior, some potentially essential correlates are missing. For example, place recognition and wayfinding as a psychological capability that can influence children’s independent outdoor playing behavior (Lingwood et al. 2015) or the influence of the weather conditions on the quantity and quality of outdoor playing can be further investigated by evidence-based research (Remmers et al. 2014). The association of forms of a neighborhood, which in terms of accessibility are led to more intersections and crossings or fewer ones (like cul-de-sac form), is less investigated by studies; however, it deserves more attention as the forms of a neighborhood have a key role in defining safe and sufficient outdoor spaces for children’s outdoor playing (Biddulph 2007). The automatic motivation correlates of children’s outdoor playing also are not clear and need to be investigated.

In addition, there is a need to identify the digital environment correlates of outdoor playing behavior and how much the correlates contribute to this behavior. An example of the correlates can be digital technology types in terms of portable or fixed devices, or the extent of the integration of digital intervention with reality (virtual/augmented reality). Moreover, the relationship between different functions of digital interventions with the correlates of children’s outdoor playing calls for further investigation. For example, it is useful to know how the training function of digital intervention relates to a child’s capability to play independently. More exploratory and experimental studies are needed to design and test the impact of various intervention functions on different correlates of children’s outdoor playing behavior. More importantly, children’s outdoor playing behaviors vary due to differences in specific personal, socio-cultural and contextual situations and needs. Therefore, interventions should be developed according to an accurate and comprehensive understanding of the outdoor playing behavior of children, while taking into account the personal, ethical, cultural and geographical characteristics.

Disclosure statement

No potential conflict of interest was reported by the author(s).