This scoping review aims to identify and map the available scientific knowledge about CIM during dark hours and further identify knowledge gaps for future studies. We investigate differences of CIM between natural light and darkness. Furthermore, we consider how artificial outdoor lighting may support CIM on foot or by bicycle within neighborhoods during dark hours. The review is based on three overarching research questions:

RQ 1: How and under which circumstances has CIM during dark hours been studied up until now?

RQ 2: What are the effects of light conditions on CIM, and for whom and where are the effects reported?

RQ 3: How are the light conditions defined and operationalized in relation to CIM in previous studies?

The overall goal with the review is to provide knowledge that could make cities more accessible for children throughout the day and seasons.

The design of the review procedure was based on literature on scoping reviews (23, 24) and on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (25). The procedure was initially defined by a review protocol expressing the purpose, search terms and eligibility criteria. The purpose of the review was expressed as follows: to identify, analyze, and describe the available research on children's perspectives, experiences, behaviors and responses to light condition. Search terms were defined in four groups to facilitate the development of a Boolean search string (1) related to the individual (e.g., child, pupil), (2) the activity (e.g., mobility, travel), (3) light conditions (e.g., street lighting, dark^*) and (4) the setting (e.g., outdoor^*, neighborhood). To avoid the scope being too narrow, both variations in natural light and artificial outdoor lighting were considered. After initial searches that combined the search terms in different ways, a final Boolean search string was created. The protocol stated that to be included in the review, papers had to (a) be published in a peer-reviewed scientific journal, (b) be retrievable from internationally available electronic databases, (c) be written in English, (d) include an empirical study regarding children within school age (from the age of six up to 18 years old), (e) focus on urban outdoor settings and (f) include light conditions in the result section. No exclusion criteria were applied regarding date of publication or geographical origin.

In order to cover a wide range of disciplines, the search was conducted within several electronic databases. To get a second opinion on the search strategies and choice of databases before initiating the searches, Lund University library staff was consulted. The searches were then conducted in the following databases: Scopus, Web of Science (ISI), PsycInfo, Eric and Engineering Village up until April 2021. The search string functioned as a template and was adapted to the different databases (See Supplementary Table 1). Relevant papers were identified by the title and, if needed, by reading the abstract. The search resulted in 304 relevant hits out of a total of 3,348. Eighty-nine duplicates were eliminated, and 121 papers were excluded because they did not include all four groups of the search terms. The remaining 94 papers were screened in detail by reading the abstract and, if necessary, screening the full-text to assure that the search terms were fulfilled. Fourteen of the papers were excluded because the search hits referred to something other than what was intended (e.g., light physical activity instead of artificial light). The 80 papers were read by four researchers who independently screened whether the abstracts in addition to the inclusion criteria contained all four aspects of the search terms: individual, activity, light conditions and outdoor environment (setting). The researchers agreed that 55 of the 80 papers were relevant for the present study. The remaining papers' full-text versions were retrieved and read. Twelve additional papers were identified in the reference lists of the selected papers. These were retrieved and assessed for eligibility by full-text reading. The final number of papers eligible for further analysis were 67 (See Figure 1).

To capture the intended scope of RQ 1, a descriptive analysis of the included papers' origin and methodological standpoints was made. The analysis focused on year of publication, type of journal, country, and continent of origin, how CIM was defined and studied, whose perspective the findings were based on, the age of the studied children, in which setting the studies has been carried out, how the light conditions were defined, and methods and theories used.

To answer RQ 2 and RQ 3, the papers were then analyzed with an inductive, thematic approach inspired by Braun and Clarke (26). First, the papers were read, and initial ideas were noted down (step 1). Then, initial codes based on the issues of CIM they explored in relation to light conditions were generated (step 2). Thereafter, the coded papers were organized into potential themes (step 3). The themes were reviewed by checking if they worked in relation to the codes generated in step 2 and if they were representative of the entire set of papers collated in step 3. Categories defining the studied light conditions were then identified within each theme. Lastly, themes and categories were defined and named.

The 67 identified papers, dated from 1999 to 2021, were published in 45 different journals, primarily within the fields of public health, medicine, and transportation research. Most of the papers were based on research in Europe (N = 26) and North America (N = 21). The remaining papers were based on research in Oceania (N = 12), Asia (N = 3), South America (N = 2) and Africa (N = 1). Two papers included data from several continents. In the following sections, it is described how CIM was defined and studied in the identified papers, whose perspective the findings were based on, the age of the studied children, in which setting the studies were carried out, how the light conditions were defined, and the methods and theories used (see Figure 2 for a summary and Supplementary Table 2 for further details).

CIM has been defined and studied in different ways in the papers. About half of them (N = 31) studied children's physical activity (PA) and sedentary time in relation to environmental features, e.g., season, day length or dark streets, or in relation to the perception of an area, e.g., perceived safety. Twenty-one of the papers studied children's or parents' perspectives on public spaces and transportation modes to different destinations and explicitly expressed a focus on children's everyday IM. Other papers are more implicit in their definition of CIM but studied the access and visits to different places (N = 4), or the interaction and perception of places from a broader perspective (e.g., in relation to transitioning toward adulthood or place connectivity) (N = 6). A small number of the papers (N = 4) focused on child pedestrian-vehicle crashes during different environmental conditions (e.g., after sunset). One paper studied the associations between environmental neighborhood features and adolescent homicide.

Most of the papers based their findings on self-reports by children, e.g., through questionnaires, interviews or focus group discussions (N = 28), or on measurements on children (e.g., daily step counts (DSC), heartrate (HR), global positioning systems (GPS) monitoring) (N = 10). Six papers based their findings on data from databases (e.g., data on pedestrian-vehicle crashes, adolescent-homicides, physical activity). Few were based on parents' self-reports, e.g., through questionnaires or interviews (N = 2). Other papers based their findings on both parents' and children's self-reports (N = 13), children's self-reports and measurements (DSC, HR, GPS monitoring or body mass index (BMI) (N = 6) or parents' self-reports and measurements on children (BMI) (N = 2).

The children in the papers were 0–20 years old, where the most common age group was children aged 13 (N = 38). Overarchingly, the children could be divided into two age-groups considering this review's focus on school-aged children: middle childhood (age 6–12, N = 48) and late childhood/adolescence (age 13–18, N = 51). Thirty-five papers included ages from both groups or children younger than six and/or older than eighteen. This means that some of the children's ages fall outside the specified age in the eligibility criteria, but also that some of the children were older than what is defined as a child according to the United Nations Convention on the Rights of the Child (i.e., every human being below the age of eighteen years) (4).

The studied settings ranged from children's local neighborhoods to a focus on e.g., park features or road environment design. Most common were built environments situated in urban areas, e.g., public open spaces, road environments, recreational facilities, parks, the way to school or activity spaces for children (N = 48). Some papers focused on a broader setting (e.g., made comparisons between countries or seasons) (N = 7).

The light conditions considered in the papers were described in a variety of terms (see Supplementary Table 2 for details), but can be divided into two major groups: (1) natural light (N = 33), including the shifting of daylight outdoors over the day or seasons, e.g., evening and night-time hours, day length differences over the year, after dark or darkness, and (2) artificial outdoor lighting (N = 34), comprising studies of outdoor lighting and the lit environment, e.g., the presence of outdoor lighting in an environment or the perception of outdoor lighting or the lit environment. Hereafter we will use the terminology natural light and outdoor lighting to distinguish between the two groups and use light conditions as an overarching concept to describe both.

Considering how the relationships between children and the environments were framed, most papers were unclear in their theoretical standpoint when studying associations between CIM and characteristics in the physical environment (N = 51). A small number of the papers explicitly applied a socio-ecological framework considering the interplay between children and their physical and/or sociocultural environments (N = 9) [e.g., Sallis et al. (27), Lang and Rayner (28)]. Some papers relied on cultural geographical frameworks [e.g., Holloway and Valentine (29), Kato (30), Vanderstede (31)], looking into the cultural dimensions of space and place of adolescents' everyday life (N = 4), while three papers applied social-psychological theories (32–35) to explore parents' attitudes and beliefs toward active school travel, outdoor play or to understand associations between environmental features and adolescent homicide. Quantitative methods and analyses (N = 50) were most common for studying associations between child behavior and environmental factors. Ten papers used a qualitative methodology and seven used mixed-method approaches (N = 7).

The thematic analysis generated four overarching themes related to the effects of light conditions on CIM: (1) physical activity and active travel, (2) outdoor activities and place use, (3) safety perception and (4) outdoor risks. The first three themes were further divided into two categories, related to the identification of effects of (1) variation in natural light or (2) presence of artificial outdoor lighting, focusing on e.g., quality. The fourth theme only included results regarding artificial lighting. For some categories, influencing factors were identified. These factors are indicated in italics in the text and constitute key aspects in relation to physical or social environmental factors, influencing the effects of light conditions on CIM. The themes, categories and influencing factors are outlined in Figure 3.

In response to RQ 1: How and under which circumstances has CIM during dark hours been studied up until now? and RQ 2: What are the effects of light conditions on CIM, and for whom and where are the effects reported?, we highlight the findings in relation to light conditions within the two major categories: natural light and artificial outdoor lighting. In the identified papers, much focus is put on children's PA, active travel, route choices, transportation mode, fear, safety perception and possibilities to be outdoors, whereas some aspects of CIM, in relation to darkness and outdoor lighting (such as exploration, play and socializing), are only mentioned to a limited extent. Hence, such essential aspects of children's lives are overlooked (103–105) with regard to darkness and lighting outdoors, making it difficult to meet the needs of children's right to rest, leisure and play in accordance with the Convention on the Rights of the Child (4). Despite this, the findings illustrated how natural light and outdoor lighting can influence CIM in different ways.

Natural light has an influence on children's PA and active travel. Darkness, with regard to day length and seasonality, comes through as a barrier to children's PA. Children of all ages seem to be more physically active at times when the day length is longer, i.e., during summer months. However, local circumstances in the built environment seem to be a deciding factor when it comes to how light and darkness influence children's PA and active travel. Environmental factors, such as degree of urbanization, the design of the road environment, and factors such as the latitude of the country could play a role. It is widely acknowledged that significant changes in the built environment and urbanization patterns have served to limit children's opportunities for IM within their neighborhoods (16, 106) and that cities are primarily designed with adults and cars in mind, not children (107). PA and active travel in rural settings, sometimes visioned as utopian environments for children to grow up in (108, 109) might also have great impact of winter darkness, where outdoor lighting is sparse or less extensive (39). Colder temperatures, wind and precipitation are presumably also contributing barriers to children's PA and active travel during winter months (38, 40). The geographical differences of PA and active travel due to daylight hours point to the need to also consider light conditions in relation to urban and rural design to understand the barriers of darkness.

Fear of darkness is common among children and influences their safety perception outdoors. This, in turn, leads to darkness constituting a barrier to CIM, both in terms of the consideration of whether to go out or not after dark as well as which places in the built environment should be avoided or not. Therefore, children have developed different coping strategies e.g., not traveling alone, avoiding certain places, or acting confident. There are apparent gender differences in the perception of safety after dark. Girls felt less safe in public spaces than boys, were more afraid of the dark and had denser walking activity spaces after dark compared to that of boys. These findings are in line with research on adults, suggesting that women tend to report more fear than men (86, 110, 111) and that insecurity and fear are factors that can limit IM (112). Also, other factors, such as the type of setting and familiarity with places, could play a role in safety perception and CIM. Hence, it seems that perceptions of safety/fear in the local environment can limit IM, but by familiarizing oneself with the area (at daytime), it could be perceived as safer (88, 113, 114). Places perceived as safe, but also offer children thing to do, can positively support CIM during dark hours (71). This is in line with research showing positive associations between place attachment (i.e., the emotional bonds between people and places) and adults' walking behaviors (115).

For older children (above 14), CIM during night-time can play a role in their identity development and transition into young adulthood, e.g., by offering avoidance of parental control, new possibilities and several “firsts” (74, 75, 82). The night can be viewed as a “second city,” with its own geography and group of people (5). Hence, the dark spaces at night might function as a refuge for older children and offer an opening “for trying to be someone the daytime may not let you be, a time for meeting people you should not, for doing things your parents told you not to do” (116). The findings highlight that promoting CIM during dark hours might not only be a contributor to the accumulation of PA, confidence, and skills, but also for their mental health. This behavior, while often negatively interpreted by adults, might fill an important function for children's self-regulation (117), pointing to the need to develop neighborhood environments in which any child can engage in play and recreational activities independently throughout the year.

The presence of outdoor lighting is related to increased PA and active travel. The perception of outdoor lighting or the lit environment can also influence activity levels, and outdoor lighting is viewed as a physical-activity-facilitator among children. However, findings are inconsistent on how and if outdoor lighting influence children's PA. The findings imply that the relationship is complex, and that child attributes, such as age, gender, and personal interests, as well as parental licenses can be mediating factors when it comes to the relations between outdoor lighting, PA and active travel.

Outdoor lighting seems to also influence children's place use, interaction with the environment and identity development. The presence and quality of outdoor lighting can influence whether children use a place, whether they perceive an area as an activity-friendly environment and affect their interactions with their surroundings and play a role in their navigation and negotiations of public space (74–78, 82). When it comes to children's route choices, however, outdoor lighting did not seem to influence children's cycling routes to an equal extent (79, 80). The reason for this is unclear and should be investigated further.

There are both structural and experienced inequalities in the spaces and places for the children's everyday outdoor life in the identified papers. Some described that there are different possibilities for children to be outdoors after dark depending on where they live (81, 83, 84) and that children experienced dimmed, broken, or absent lighting as a material manifestation of neglect from powerful institutions and class devaluation (81). In previous studies, it has been found that people living in socially vulnerable areas make fewer trips per person and per week than the national average (118, 119). It has been proposed that the way urban spaces are lighted can reinforce inequality, but also challenge them (120). Inequalities in lighting design in relation to CIM and mobility justice should further be addressed in research.

The presence of outdoor lighting and lighting quality may also play a role in children's safety perception, which in turn can influence their IM. Contrary to the literature on adult's responses to outdoor lighting these findings are still limited. Among adults, the relationship between outdoor lighting and perceived safety was by far the most frequently researched topic (121). Moreover, light sources with whiter light seem to have a positive effect on adults' safety perception (121). In our review, one paper suggested that outdoor lighting with greater green or blue hues was related to children feeling safer (96). Findings also showed that areas perceived as well-lit could be perceived as safer and that children requested brighter outdoor lighting as a way to feel safer (81, 90, 96). However, Côté-Lussier et al. (96) relied on lighting data extracted from images taken from an International Space Station rather than photometric on-site measures (e.g., illuminance, luminance, spectral power distribution) or observer-based assessments commonly used to study human perception, evaluation or behavior of the lit environment (8). The assessment of the raw image brightness values using photos from space gives a rough measure of the lighting in an area, but it is uncertain if it can be compared to the actual experience on site. Overall, the restricted number of identified papers and the limited understanding of the lit environment within these papers means that more research is needed to understand the relationship between children's safety perception and the presence and quality of outdoor lighting.

Presence of outdoor lighting might influence outdoor risks for children, but also children's perception of risks. The highest number of accidents occurred when most children were outdoors, e.g., during days with good weather, summer months, and afternoon and evening hours. Overall, the papers (e.g., 97–99) say little about the relative risk of child injuries after dark or in the presence of outdoor lighting. Only two of the papers (100, 102) showed a relationship between greater risk for child injuries and death after sunset, and that parents' perceptions of outdoor risks after dark can limit CIM. Traffic safety issues have been highlighted as a major constraint for children's active travel [e.g., (17, 122)], which emphasizes the importance of untangling the actual and perceived risks of accidents when children are independently mobile during dark hours and the potential of outdoor lighting to influence such risks.

The presence of outdoor lighting is associated with decreased odds of child homicide (101). It has been claimed that outdoor lighting can reduce crime and increase public safety (123), but it has also been disputed (124) and criticized as an approach that elude the deeper socio-economical causes of crime (125–127). The connection between outdoor lighting and crime appears to be sensitive to local conditions and should be context appropriate (128, 129). Since the findings in this review is only based on one paper, no conclusions can be made between the relationship of outdoor lighting and children's risks of being victims of crime.

It should be noted that there are legitimate differences regarding the age when children are allowed by their parents to be independently mobile (e.g., in their neighborhood or commuting), but also that children of different ages have different needs of independence. The focus in this review has been on children within school age (6–18 years old), which brings big differences in the degree of their IM. Despite this, the findings point to how darkness can be a barrier for IM for children of all ages, while also illustrating how this barrier is realized in various ways. For younger children, parental constraints regarding darkness can limit their possibilities to be outdoors and entail e.g., being driven to activities [e.g., Forsberg et al. (50), Nakanishi and Black (102)]. Older children could also be subject to parental constraints, but here it appears that darkness also entails spatial barriers, leading to parental restrictions in the form of “no-go” areas outdoors [e.g., James and Embrey (51), Pooley et al. (70)]. Overall, the findings indicate that outdoor lighting, when perceived as adequate, has an impact on CIM for all children within school age.

In order to cover a variety of perspectives of CIM during dark hours and the role of outdoor lighting, both natural light and outdoor lighting were considered in this scoping review. In RQ 3 we asked: How are the light conditions defined and operationalized in relation to CIM in previous studies? with the objective to further the understanding of the light conditions per se.

A drawback that has emerged in the papers is the lack of details about the studied outdoor lighting. Several of the papers discuss the existence or non-existence of lighting and mention “poor lighting,” “well-lit,” “sufficient light,” “better lighting” or “good lighting” in relation to children's use and experiences of the outdoor environment, but few of them focus in detail on the qualities of outdoor lighting to meet children's needs for IM. What is meant with “well-lit” or “good lighting” from the perspective of children? What type of light source or lighting design could support CIM? Research that includes both observer-based and technical-based environmental assessments of the studied lighting could provide better insights into these matters [cf., (10, 130, 131)]. Côté-Lussier et al. (96) is the only paper that has used technical assessments of the outdoor lighting, but the assessments are arguably too simplified and rough. To better understand the associations between CIM and outdoor lighting, more in-depth research that takes both technical aspects of lighting and the lit environment, together with observer-based assessments from children's perspectives into account, is needed. A good starting point would be to base these assessments on well-established and proven methods within lighting research and child studies.

There are also limitations in the chosen methods and included perspectives in the identified papers with regard to understanding how outdoor lighting could support CIM. Since parental factors, such as parents' perceptions of the neighborhood, parental license and parents previous IM experiences, influence CIM (132), it is of relevance to understand parents' perspectives of light conditions. This is reflected in the identified papers to a limited extent, with parents' views included in 25% of the studies. Nevertheless, children's views of their environments should not be ignored. The basis that IM is every child's right and that they have the right to be heard and to express their views on matters affecting them (Article 12) (4) should be a point of departure. Children's perspectives on light conditions captured through self-reports (e.g., through questionnaires or interviews), were included in 47 of the papers, indicating that their perspectives are considered to a great extent. Additionally, the included children's attributes varied between the papers, representing children of different age groups, genders and countries. However, a predominant part of the papers was based in European or North American contexts, and focused to a great extent on urban environments, indicating that the findings are mostly based on a Western and urban perspective. To better understand CIM during dark hours, research including children from different parts of the world and different kinds of settings is desired.

When studying the methods used for children's self-reports in the papers, it is also apparent that the opportunity for children to express their views on outdoor lighting are limited. Thirty-one of the 47 papers base their findings on questionnaires, but few of these include children's perceptions of lighting or the lit environment, and if they do, the children are asked to answer a statement about how they perceive the lighting (e.g., positive or negative) or if lighting is present or not [e.g., Evenson et al. (57), Kamargianni et al. (67), Onywera et al. (68)]. Instead, most of the papers that include aspects of outdoor lighting have assessed its presence in an area through GIS or so called “objective measures” on site [e.g., Sallis et al. (54), Goon et al. (58), Edwards et al. (76), Flowers et al. (77)]. There are fewer qualitative papers on children's perspectives on outdoor lighting, but these include more detailed descriptions on how children are impacted by light conditions in their outdoor lives [e.g., Mier et al. (65), Mecca (75), Thoma et al. (81)]. Here, however, technical-based environmental assessments are lacking, making it difficult to interpret the findings. As previously concluded, technical-based environmental assessments in combination with methods capturing children's and parents' perspectives [e.g., Derr et al. (133)] could provide better insights into the matter of the effect of outdoor lighting on CIM. Detailed knowledge of the lighting would aid in the understanding of how to promote exploration, play and socializing for children in their local environments in addition to supporting their essential commuting to and from school and other educational activities.

Another drawback in the papers is that the theoretical starting points are rarely explicitly stated, making it difficult to evaluate and interpret the findings. Recent systematic reviews on CIM have shown that it is associated with several attributes, such as socio-demographic, social and physical (134–136). It is therefore motivated to use theories that enable addressing the child-environment relationship. Of the 67 identified papers, only 12 refer to such theories. To better understand the complexity of CIM in relation to light conditions, using a framework considering ways in which parent and child attributes can intersect with neighborhood characteristics, is suggested in future research.

The limited findings about CIM during dark hours and the role of outdoor lighting might depend on shortcomings of the eligibility criteria and search strategy used. Additional themes might have emerged if papers written in other languages than English had been included. Also, the addition of non-peer-reviewed reports could have resulted in greater insights on how different light sources and lighting design might affect children's opportunities and needs. Nevertheless, the peer-review criterion was considered as the best way to identify and map the available scientific knowledge on CIM during dark hours and to identify potential research gaps for future studies. This is motivated by the fact that few studies focus on children's needs and experiences of light conditions in their everyday outdoor life, thereby implying that outdoor lighting recommendations and standards are informed by research based on adult's perceptions.

Due to limited possibilities in conducting full-text searches in all of the selected databases, papers that include aspects of light conditions may have been missed. There are existing papers that could have been relevant to include in the review that were not incorporated since they were not identified through the search process (e.g., found through colleagues, Google searches etc.). To ensure that the findings of this review can be updated and replicated, papers identified through methods other than citation searching was excluded. Despite these limitations, we believe that the most relevant papers have been found and included, when we consider the aim of this review.